Re: [SLUG] Wine & setup
On Tue, Apr 05, 2005 at 01:09:18PM +1000, Alan L Tyree wrote: > I have a windows based CD that I would like to install and use with > Wine. The setup.exe prog crashes. > > What's the best way to proceed? is there anyway to extract and install > without using the setup.exe program? If you have a spare Microsoft box which is relatively clean you can do the setup on that box, then move the files across yourself BUT don't just go right ahead and do the setup... do it this way: * Get a list of all the files on the Microsoft machine (before install) DIR /S C:\ > files1.txt * Dump out the entire registry as a text file (using regedit, before install) * Do the install * Get a new list of all the files DIR /S C:\ > files2.txt * Dump the entire registry as a different text file (again, using regedit) Now you need to get those dump files onto a linux box and use diff: $ diff files1.txt files2.txt | less $ diff dump1.reg dump2.reg | less Hopefully, that tells you what files you need, hopefully they aren't scattered about the place too much. Note that some registry settings may be keyed to a particular machine, some may be encrypted or encoded in some hex encoding. This isn't a guaranteed method but it gives you a fighting chance. Also note that if you install the cygwin environment on the Microsoft machine then you can use "find" to scan out the files which is better than "DIR /S" also then you get "diff" and "less" without needing to swap back and forth to Linux. - Tel ( http://bespoke.homelinux.net/ ) -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] running Konqueror as non-root + lan:/ ioslave + "illegal" port assignment for NFS
On Wed, Apr 06, 2005 at 12:22:26AM +1000, Jeff Waugh wrote: > You can't talk to an NFS server set up like that with any non-root program > in a safe and sensible manner. You can, however, set up the server to accept > connections from unsafe ports. This is pretty reasonable if it's a read-only > share, or you're in a trustable environment... but it's completely unsafe in > most circumstances. :-) Using a "trusted" port as a measure of security is unsafe under almost every circumstance. In effect you are trusting the physical security of the subnet plus every machine on that subnet. Suppose someone clips a laptop (or palm pilot, or wristwatch) to a spare ethernet outlet or jumps onto your wavelan... suddenly they have acess to a "trusted" port (and a "trusted" ip too). > Just add "insecure" to the nfs options list on the server. If only there was a "secure" option... *sigh* - Tel ( http://bespoke.homelinux.net ) -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] dealing with compromised machine ?
These suggestions are for next time, I suspect it is too late to take any of these on board in this particular situation. On Wed, Apr 06, 2005 at 08:16:14AM +1000, Voytek wrote: > I have a compromised RH73 machine, until such time as I can pull it down, > what can I do to identify and shut down any rogue processes/backdoors ? First thing I would suggest is some network dumping. Consider either ethereal or tcpdump, get an old, dumb hub and drop it into the ethernet connection to that machine, be quick so it doesn't lose connectivity for more than a few seconds. Then put a temporary machine on the same hub, bring up an interface (but give that interface no IP number) and start dumping to hard drive. Don't use a switch! Old 10Meg ethernet hubs are best but you can still get 100Meg hubs if you search. Save those dumps for when you talk to the police (of course you are going to do the right thing and report this) they will find the dumps very useful. Hopefully whoever broke in will make some contact with the broken box and might reveal something about themselves. NB: at this stage you do NOT want to do anything abnormal that might make it clear that you are paying great attention to this machine. The sniffer machine can be completely self contained with no contact to the outside world other than silently sniffing. Don't even think about trying to sniff on the same machine that is broken. > I've removed all the baddies, but, I expect there will be some open ports ? > is there a way to shut them in the interim period till I can get to the > machine ? There is a big problem with leaving a compromised machine active and also removing stuff while it is live. It is a much more dangerous thing than just leaving the compromised machine alone. Whoever has broken your machine has (approximately) the following priorities: [1] Remain undetected [2] Keep the machine active and stable [3] Collect information [4] Use the machine to break other machines Once they know they have been detected the above priorities go out the window and they really only have one thing that matters anymore which is destroying evidence and cleaning up their tracks as much as possible. By poking around and removing this and that you are spelling it out to whoever is on the other end of the line that they should think about filling your partitions with random numbers. So you sort of have to operate in two distinct modes... BEFORE you let them know they have been detected you are trying to watch from the sidelines and make notes... when you decide that enough is enough, then you have to pull the network plug clean, type sync a few times and just turn it off. There's no half-way. After you do turn it off, boot off a CDROM and take a full hard drive image which the police will also find useful. The rootkits are quite often customised and may contain links to websites, other compromised machines and bits of forensic evidence that might make it to court. Some people leave bash history behind, others leave temporary files and all sorts of stuff. They spend all day filtering through this junk putting clues together, often from multiple sites. You are paying for this, might as well keep them busy and get something for your money. By the way, in NSW the investigation of computer related crimes is the job of the fraud squad, see http://www.police.nsw.gov.au/ - Tel ( http://bespoke.homelinux.net/ ) -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: FW: [SLUG] Possible hacker Attempt
On Thu, Apr 07, 2005 at 09:48:48AM +1000, Howard Lowndes wrote: > My bad. Of course Linux apps can be just as badly written as M$ apps > and can have buffer overflows. > > What I should have said is that this attempt at a buffer over flow does > not affect Apache. Whilst Linux apps can be badly written, there are a few reasons why Microsoft apps tend to be worse. Consider that most of the people who program for Linux have also tried working with Microsoft products and they chose Linux because they wanted to do things the right way. This has a filtering effect that pulls the strongest programmers away from MS and toward either Linux or one of the *BSD variations. People who care about money first and producing a good product second will tend to go to the Microsoft camp, people who think that a good product should be first priority and then look at money as a secondary consideration usually gravitate in a free-software direction. Its a difference of emphasis rather than a completely different approach but it does make a difference. Then there are the Microsoft customers who have not (until recently) been willing to get nasty about software quality. With market forces at work, if there is no demand for high security, high quality software there will also be no supply. When Microsoft had a total monopoly they just didn't have to care one way or the other... now that their monopoly is weakening and some customers are getting aggressive they are forced to care so they are making more of an effort. As a practical example to all those C programmers out there, this is the common idiom (since early K&R days): int myfunc( char *stuff ) { char buf[ 100 ]; sprintf( buf, "My stuff is %s", stuff ); /* do something with buf here */ } Of course this is exactly what will sting you if you can't be sure what "stuff" might contain (in technical terms we say that "stuff" is tainted so we can't trust it). The trick is that the C compiler puts the "buf" variable on the stack and also puts the function return address on the SAME stack AFTER buf when it calls sprintf(). Using the same stack for variables and code pointers is a good optimisation for speed but it is bad for security because when the "stuff" is too big it wipes over the return address. With some care, it can replace the return address with a pointer into itself which sets to program running onto a completely new chunk of code. Now redhat (and others) have a few tricks to make that more difficult, the first is to limit the executable sections of memory and make it illegal for executable code to exist inside stack buffers. It is interesting that Microsoft were talking about how important this feature is and how they would have it real-soon-now at the same time that Redhat were shipping with it enabled. The second is using a randomised offset for various memory chunks in the program to make it much more difficult to figure out what return address to load into "stuff" to make it do what you want -- then the hacked program merely crashes rather than opening a back door. Redhat also ships with that feature. The easiest source-code fix is to use the snprintf() function so that the buffer size is known to the formatter like so: int myfunc( char *stuff ) { char buf[ 100 ]; snprintf( buf, 100, "My stuff is %s", stuff ); /* do something with buf here */ } But the snprintf() call is not POSIX, it is from BSD (and now adopted by Linux) and last I looked Microsoft C did NOT provide that function. Of course there are lots of other ways to protect yourself but the Microsoft programmers are forced to do it the hard way. Worse... when porting code from Linux and BSD onto Microsoft, the people doing the port find there is no snprintf() so they knock up a quick compatibility library that just ignores the buffer length and calls sprintf() resulting in a nasty vulnerability but only in the Microsoft version. Note that early Linux implementations (libc 4) had the same hasty hack but it got fixed and I doubt that there are too many of those old Linux boxes running on the Internet these days. So yes, anyone can write bad code but in practical terms Microsoft still has a bit of catching up to do. - Tel ( http://bespoke.homlinux.net/ ) -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Re: run levels in debian/ubuntu
On Fri, Apr 08, 2005 at 02:24:03PM +1000, Matthew Palmer wrote: > On Wed, Apr 06, 2005 at 02:30:21PM +1000, James wrote: > > when I install something that I may not want to use all the time eg.. > > Don't do that. Leave the symlinks in place, but rename from Snn to Knn. > I've also heard that if you leave at least 1 symlink in place update-rc.d > won't recreate all of them, but I haven't tried that. But update-rc.d > *will* respect your changes to ordering or S/K. If it doesn't, that's a > bug. > > Apart from being almost impossible to work out whether the symlinks have > been removed or were never added, it is more correct to change Snn to Knn > because that says "I do not want that service to be running in this > runlevel". No symlink says "Buggered if I know what I want done in this > runlevel". For what it's worth... the program "ntsysv" provides an interactive text mode interface which might be a bit easier to work with than renaming links. It's standard on RedHat... I'd be stunned if it wasn't widely available on just about every other distribution too. The only hard bit is remembering the name "ntsysv" which certainly doesn't provide any mnemonic in my frame of reference for "that program that renames the symlinks for you". Maybe debian has the same feature under a different name (that wouldn't surprise me in the least). - Tel ( http://bespoke.homelinux.net/ ) -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Re: [chat] Wireless
On Sun, Apr 10, 2005 at 11:34:03AM +1000, Terry Collins wrote: > Phil Scarratt wrote: > >I had someone (off-list) suggest install a "matrix" of aerials (eg 4) > >and run them all back to a header-amp with one output to the AP. Not > >sure that would work, and may be overkill anyway. > Perhaps someone can correct me, but I understood from my physics that > four separate aerials all pumping out the same signal would cause > interference with each other. They would, but the resulting interference pattern would not necessarily be a bad thing. Figuring out what that pattern would be in a complex building with reflections, etc is just about impossible but remember that even one single aerial in a funny-shaped reflective box (i.e. a building) will also cause an interference pattern with its own reflection. One problem with the "matrix of aerials" idea is that the coaxial cable that runs between them has to be low loss at GHz frequencies which is NOT cheap cable so running cat-5 out to 4 individual base stations is going to save money if the distances are long. Also, if all 4 base stations are running different channels, the clients get multiple chances to get a connection (try each channel and see which is better). I think it's possible for smart wireless cards to check several channels at once and hop around (not that I know enough configuration details to be able to set them up to do that). - Tel ( http://bespoke.homelinux.net/ ) -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] libcrypto, libssl needed for ProFTPd
On Mon, Apr 11, 2005 at 09:23:48PM +1000, Voytek wrote: > I'm trying to install ProFTPd from an rpm, but getting failed dependencies: > what do I need to do ? > > # rpm -i proftpd-1.2.9-1.7x.i386.rpm > error: failed dependencies: > libcrypto.so.1 is needed by proftpd-1.2.9-1.7x > libssl.so.1 is needed by proftpd-1.2.9-1.7x > > # openssl version > OpenSSL 0.9.6b [engine] 9 Jul 2001 I notice there's openssl095a-0.9.5a-11.i386.rpm in the package list, you might have to install this one to get the older versions of the libraries. If this works then that's probably the best answer. Don't uninstall your current version, they should both be able to be installed at the same time. Sometimes with these library version problems, symlinks can work: ln -s /lib/libcrypto.so.0.9.6b /lib/libcrypto.so.1 ln -s /lib/libssl.so.0.9.6b /lib/libssl.so.1 If that does seem to work then I would suggest you do a bit of testing because there might be some trivial incompatibility that causes it to crash. Then again, you probably just want to do regular ftp which is never going to use the details of the ssl system anyhow. > # yum update openssl > Gathering package information from servers > Getting headers from: Red Hat Linux 7.3 base > Getting headers from: Fedora Legacy utilities for Red Hat Linux 7.3 > Getting headers from: Red Hat Linux 7.3 updates > Finding updated packages > Downloading needed headers > No Packages Available for Update If droids could think, none of us would be here. Probably misquoted... - Tel ( http://bespoke.homelinux.net/ ) -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] libcrypto, libssl needed for ProFTPd
On Tue, Apr 12, 2005 at 08:20:58AM +1000, O Plameras wrote: > > Ah hah. > > The procedure is: > > 1. rpm -ql --whatprovides | grep libcrypto.so.1 > etc. > 2. If #1 is TRUE, then fixed /etc/ld.so.conf But if your system is missing the critical package you won't get an answer. It only works if you have access to a system with all the right packages installed. For a guy running 7.3 where everyone else is 4 or 5 versions ahead that can be kind of difficult. RedHat dependencies are easy to use when you have a reference machine "with the lot" that has been installed with the "everything" option ticked. This machine will always help you figure out your dependencies then you can run your actual production machines with just what they need. Mind you, these days hard drives are so big, running machines ultra-lean is pretty much a waste of time. - Tel ( http://bespoke.homelinux.net/ ) -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Using software raid0 on 2 sata disks installing gentoo
On Mon, Apr 11, 2005 at 11:01:55PM +1000, Chris Portman wrote: > Hi everyone... im trying to install gentoo on a software raid0 array > across 2 sata disks. the boot partion is just a normal partition > though on /dev/sda1. > > Ive read all the howtos i can find... they all go through the same steps > and the install procedure runs through fine however when i get to the > end and reboot... it can mount the root partition as /dev/md0 doesnt > exist basically. I don't know what gentoo does but on RedHat there is a magic ramdisc file called /boot/initrd that loads with the kernel and then goes about constructing the /dev/md0 device. It uses some tricky scripting and goes and looks at the partition names and stuff like that. Presuming gentoo works the same way then checking the initrd files is probably a place to start. Could be that you have an initrd that is suitable for SCSI but not for SATA (the ramdisc actually contains kernel modules and installs just the ones that are needed for your system, evil huh). - Tel ( http://bespoke.homelinux.net/ ) -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Exhibitors at Education Expo
On Tue, Apr 12, 2005 at 09:53:56AM +1000, Terry Collins wrote: > 4) The MS Tax is charged on EVERY computer in the school, hence there is > no money saving incentive to replace MS with FOSS/Linux. I thought that was illegal... ...how do they attempt to enforce this? > As my wife was a teacher in the NSW Dept of Ed, I've looked at helping > introduce FOSS into schools a number of times. Each time I have decided > that I have far better things to do. To be honest I found the same thing with UTS Engineering, and you would think they are in a better position to understand these things but their interest in taking risks is about zero. > Someone who wants to dabble in this area with FOSS might consider making > a Live CD, especialy for simulations, demos, etc. Or possibly start an online service for teachers since one thing that schools are doing right is providing lots of Internet access. Then you can use Linux under the hood and maybe encourage them to download a bit of open-source on the side, gradually get them into open office and gimp. Maybe get them interested in buying a server with the same online services for their internal use which will run faster (less users and over ethernet LAN). I'd be game to get involved in that one if someone wants to collaborate. > If you want to offer "certificates", consider setting up a Registered > Training Organisation. Then you can offer Cert I, Cert II, etc course, > but you do need Cert IV qualified trainers. You got any links to explanation of how this process works? How does the first Cert IV get created in a new subject? - Tel ( http://bespoke.homelinux.net/ ) -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] PHP load error
On Tue, Apr 12, 2005 at 01:25:59PM +1000, Edwin Humphries wrote: > We were having some errors in execution of a cron-initiated php script. We ran > it from the console, and got some errors. If PHP works when accessed through your web server then you can initiate the script by putting a lynx command into a cron script: lynx http://127.0.0.1/blah/blah.php > /dev/null I've often noticed anomalies between command line php execution and running it through a web server (although in theory both should be equivalent). If your PHP does not run through the web server then, yeah, you really do have a problem, follow the troubleshooting suggestions elsewhere in this list. - Tel ( http://bespoke.homelinux.net/ ) -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] MS Tax on every computer
On Tue, Apr 12, 2005 at 07:50:54PM +1000, Simon wrote: > > > -Original Message- > > From: [EMAIL PROTECTED] > > [mailto:[EMAIL PROTECTED] On Behalf Of Terry Collins > > Sent: Tue, 12. April 2005 12:21 PM > > To: slug@slug.org.au > > Subject: Re: [SLUG] Exhibitors at Education Expo > > > > > > [EMAIL PROTECTED] wrote: > > > On Tue, Apr 12, 2005 at 09:53:56AM +1000, Terry Collins wrote: > > > > > > > > >>4) The MS Tax is charged on EVERY computer in the school, > > hence there > > >>is > > >>no money saving incentive to replace MS with FOSS/Linux. > > > > > > > > > I thought that was illegal... > > > ...how do they attempt to enforce this? > > > > It was either a flat or stepped fee based on the number of computers > > they knew that your school had. NSW Public schools basically get all > > their hardware through the department. > > > > So there was no incentive for say "by replacing MS with FOSS on X > > computers I can save $Y". > > I thought he was being cynical in the fact that it is difficult to buy a > PC without Windows - you need to specify that and negotiate the price > (usually cheaper). > However this may be a reference to the 'Schools Agreement' which I > belive DET have signed up to for all their schools - this gives schools > access to nearly all MS software at NO extra cost - some server based > stuff still has some cost. You pay a licence for every PC in the school > regardless of OS or whether it is capable of running the latest MS > software (which is all the licence covers - you can downgrade by only > one version) BTW you are helping to pay for this through State taxes. > While this is good for the DET schools it is a very significant cost in > the Independent Sector. Heres a link from Victoria: http://www.sofweb.vic.edu.au/ict/software/microsoft/index.htm They say "State Wide" licenses but they don't say whether you have to pay per PC, the implication seems to be that license fees are all paid. There are certainly some details glossed over. I found it hard to get details from NSW (brief search only) found this which says all new computers are to have Microsoft pre-installed but doesn't actually say it is impossible to get computers on the side: https://www.det.nsw.edu.au/media/downloads/reports_stats/annual_reports/year99/chapter3.pdf There was a law passed in 1998 resulting in Section 51AC of the Trade Practices Act making it unlawful to engage in unconscionable conduct. It goes some way towards protecting small business against larger, more powerful, companies treating them harshly or unfairly in a commercial relationship. In some cases the small business can simply declare a contract void. In this case: * The DET made the agreement on behalf of the independent schools (i.e. they were forced into agreeing) * The bargaining power was enequal (MS and DET combined vs independent school) * The schools were deprived of their ability to negotiate terms * The schools have been deprived of their normal legal right to purchase software from alternative suppliers * The schools are forced to comply with unrealistic conditions that are breached by any additional computer purchase (i.e. under normal trade conditions MS would get paid for what they deliver, not for what someone else delivers). There's a chance that an independent school could fight it under these grounds. It may have to go through the ACCC. I think there's also a few other grounds like monopolistic practices and Microsoft already got told not to do this sort of thing with PC retailers (from memory their PC retail agreements were declared void both here and in the USA). They are just repeating the entire exercise with schools. You would have to find a proper lawyer to give you real details but if the independent schools are getting stung by these costs, I suggest they get together and fight it. Maybe the schools don't see how much they are getting railroaded so they don't try to fight it, maybe they just don't care... At least it would be interesting for them to get the full documentation together and plonk it in front of a good Trade Practices lawyer and see how much wriggle room they have to work in. - Tel ( http://bespoke.homelinux.net/ ) -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] MS Tax on every computer
On Wed, Apr 13, 2005 at 11:19:38AM +1000, Simon wrote: > > None of this applied to Independent schools in NSW, they are, well, > independent. OK, so there is a reasonable avenue promoting the cost-saving benefits to independent schools. But government schools are under the thumb of the DET... somehow I got the impression that the DET had signed a deal for the entire state. As I said, it's a bit difficult getting published details of how these deals work. - Tel ( http://bespoke.homelinux.net/ ) -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
[SLUG] SugarCRM not really Open Source at all
This really annoys me, here is a press release: http://www.osbc.com/live/13/media//news/CC417286 So they are having a special Open Source conference and picking their favourite Open Source products, and there is SugarCRM listed as an Open Source application. So what license is SugarCRM really under? Well that depends on where you care to look. If you look on sourceforge: http://sourceforge.net/projects/sugarcrm/ You see it listed under the Mozilla Public License 1.1 which is indeed an OSI approved Open Source license. So now you go and download the code and you get this: /* * The contents of this file are subject to the SugarCRM Public License Version * 1.1.3 ("License"); You may not use this file except in compliance with the * License. You may obtain a copy of the License at http://www.sugarcrm.com/SPL * Software distributed under the License is distributed on an "AS IS" basis, * WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License * for the specific language governing rights and limitations under the * License. * * All copies of the Covered Code must include on each user interface screen: *(i) the "Powered by SugarCRM" logo and *(ii) the SugarCRM copyright notice * in the same form as they appear in the distribution. See full license for * requirements. * * The Original Code is: SugarCRM Open Source * The Initial Developer of the Original Code is SugarCRM, Inc. * Portions created by SugarCRM are Copyright (C) 2004 SugarCRM, Inc.; * All Rights Reserved. * Contributor(s): __. / Hmmm, seems to mention a "SugarCRM Public License Version 1.1.3" and also seems to mention an ADVERTISING CLAUSE requiring you to display their logo. That's not part of the OSI definition and it's not part of the MPL 1.1 either. Seems that their source code does not agree with their dishonest sourceforge listing. Then there is this license here: http://www.sugarcrm.com/home/Public_License/199/ Which demands not only that you show their logo but that every single user interface page has to link back to their website. Has this been OSI approved? Is it likely to get OSI approved? How is it they can go around call this Open Source? I've emailed the OSI about this and Sourceforge too and they have both been quite sluggish doing anything about it. I'm starting to think that neither of them really care. I wonder how many other people have been duped by these guys... It isn't healthy for the community and it's sad to see misinformation being propagated. I'm posting this as a public announcement about how difficult it is to find anyone you can trust. - Tel ( http://bespoke.homelinux.net/ ) PS: this is a personal gripe, nothing to do with me being secretary of Slug, but I'm guessing that other slug members might also be annoyed at a dishonest vendor. -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] GCC question
On Sun, Apr 18, 2004 at 08:58:56PM +1000, David Bowskill wrote: > HI SLUGgers, >I have this little problem with GCC which I wonder if anyone can > enlighten me on. The little program below has been run on three > different machines all with the same result. The program compiles OK > under GCC but fails with a segmentation error. Under Borland 'C for > DOS', it compiles and runs correctly. >From the "info gcc" pages: `-fwritable-strings' Store string constants in the writable data segment and don't uniquize them. This is for compatibility with old programs which assume they can write into string constants. Writing into string constants is a very bad idea; "constants" should be constant. This option is deprecated. Compiling with the above option will probably make your program work. - Tel ( http://bespoke.homelinux.net/ ) -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Failed Boot
On Mon, Apr 18, 2005 at 11:05:21PM +1000, Simon wrote: > Hi all, > I have a RH system V9 which is hanging during boot at 'Mounting SMB > Filesystems' we have left it for over 48 hours and it still does not > progress. However it still responds to CTRL_ALT_DEL to re-boot, so is > their another command I can hit on the keyboard to tell it to skip this > part of the boot and keep going? I don't seem to be able to get the "I" > for Interactive setup happening either - it just boots normally. Put the RH9 install boot CDROM into the drive, boot off that and at the boot prompt type "linux rescue" which should get you to a shell prompt with your hard drive mounted under /mnt/sysimage (I think that's where it ends up, find it with "cat /proc/mounts" just to be sure). Then you can edit the /etc/fstab and put a "#" at the front of the line containing the troublesome SMB filesystem (that comments out the line). When you are finished, type "sync" a few times and then type "exit" and it will reboot. - Tel ( http://bespoke.homelinux.net/ ) -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Buying a Printer
On Tue, Apr 19, 2005 at 09:23:32PM +1000, Peter Hardy wrote: > On Tue, 2005-04-19 at 18:53 +1000, Richard Neal wrote: > > Also while Im here whats a good three in one printer that people have > > had experience buying and using with Linux lately. > > HP anything. > > Not sure about the cost, but HP have open sourced their drivers. > Originally on the provision that the code only be used with HP printers, > but that's since been removed. I've had excellent performance from the laserjet 1200 with a simple parallel port and Postscript. It's a really nice printer, pitty they don't make them anymore. However, I've also had a terrible time trying to get a laserjet 1100 working on Linux which is almost the same printer... the print engine and physical shell are identical but the 1100 has a USB interface and no Postscript support and it uses some sucky bit image-language. Maybe the newly open sourced HP drivers are OK, I'd say that the standard parallel port and Postscript style interface are heaps easier. > I've used a couple of different models, although model numbers slip my > mind. They both printed and scanned flawlessly, though. But the numbers are important, HP make a range of totally different interfaces on otherwise identical printers. I think that the LJ-1300 series are still out there and reasonable value, might be worth looking at. - Tel ( http://bespoke.homelinux.net/ ) -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] GCC question
On Tue, Apr 19, 2005 at 11:36:48AM +0300, Amir Binyamini wrote: > telford noted that info gcc shows that this option is deprecated. > http://lists.slug.org.au/archives/slug/2005/04/msg00404.html > > It could be that in newer versions (gcc 3.4,gcc4) this flag is built-in, Quite the opposite, the option is marked as deprecated because the authors of gcc don't much like the way you write programs and want you to modify your code so that it will not need writable strings. Then again, they will never be able to remove the option because they want gcc to remain compatible with other compilers so what they are saying is, "sure we will support your program but you are a naughty boy to write such code." > I don't know; but in man gcc in gcc 3.3.3 they also say > that this is a bad idea ;I must find time to check what implications such > a flag have and if > it can cause any troubles and in which scenarios. I'm pretty sure it never actually causes trouble, it's more of a "good programming" argument about whether strings should be regarded as constant or not. You can spend a lot of time arguing about what is good programming style and people do. I've been thinking about writing a book on as much but I'm wondering if anyone would buy it :-) - Tel ( http://bespoke.homelinux.net/ ) -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Group Calendaring Application
On Thu, Apr 21, 2005 at 01:42:56PM +1000, Edwin Humphries wrote: > A small customer for whom we're installing a Linux server wants to use > Outlook for Group Calendaring (only) - for this some other person has > advised an Exchange server. One alternative we have looked at is Scalix, > but I think it's overkill for Group Calendaring only - and it's quite > expensive for 5 - 10 users. If they are happy to go over to webmail style interface then maybe TWIG could do the job. It's easy to set up and works with most web browsers. Have a look at http://www.informationgateway.org/ - Tel ( http://bespoke.homelinux.net/ ) -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Parking wiki
> http://wiki.slug.org.au/wiki/ParkingAtUTS OK, I've done a bit, I'll go for a walk tomorrow and check out the times on various parking signs around the place. - Tel -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
[SLUG] Funny IRC log
Probably everyone has seen this, but what the hell... http://www.jellyslab.com/~bteo/hacker.htm -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] [OT] How to write a good IT resume.
On Thu, Apr 28, 2005 at 04:27:57PM +1000, Vini wrote: > If you know about IT resumes I would appreciate if you could share the > information. If only there was a group dedicated to helping their members manage their resumes and you could look at other people's resumes to get ideas for your own style. They would have to be secure, and organised, and not cost too much, and be an online community but with regular real-world social get-togethers as well... Could there be such a group? Would they give a talk at Slug? - Tel -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
[SLUG] Parking at UTS for Slug Meetings
Well I've done a walk around the block a few times and updated the Wiki. Haven't been hanging round the brewery much so there are a few more details to be added round that area. http://wiki.slug.org.au/wiki/ParkingAtUTS Remember that anyone who registers can edit the Wiki. It started out active but now seems to be just sitting around. At very least, when you find a good News link, add it to here: http://wiki.slug.org.au/wiki/LinuxNews - Tel ( http://bespoke.homelinux.net/ ) -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Old 486 DX 66 to give away.
This is probably another good use for a wiki page... each person with spare junk lists it on the same wiki page along with where to pick it up from and then everyone knows where to grab spares from. On Fri, Apr 29, 2005 at 11:40:21AM +1000, Luke Yelavich wrote: > Hi all > I have been given an old 486 DX 66, 516 MB hard drive, sound card, 3.5 > and 5.25 floppy drives, CD-ROM. Seemed to boot fine, appart from a dead > CMOS battery, but wouldn't boot for a second time. Haven't opened it, > got about 4MB RAM. I could use it for spare parts, but I have no great > need for most of the components in this system. You may be able to make > it boot and use it for a firewall. > > if you can pick it up from Croydon, you are welcome to it. > > Feel free to contact me with questions/pickup arrangement. > -- > Luke > > Get my public GPG key here: http://www.themuso.com/themuso-gpg-key.txt > -- > SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ > Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] wiki
On Fri, Apr 29, 2005 at 08:52:04PM +0930, [EMAIL PROTECTED] wrote: > Hi, > > 29Apr2005 @ 17:55 [EMAIL PROTECTED] thusly spake > > This is probably another good use for a wiki page... > > each person with spare junk lists it on the same wiki page > > along with where to pick it up from and then everyone knows > > where to grab spares from. > > > If only there was a group dedicated to helping their members > manage their resumes and you could look at other people's > resumes to get ideas for your own style. They would have to > be secure, and organised, and not cost too much, and be an > online community but with regular real-world social get-togethers > as well... > Could there be such a group? Would they give a talk at Slug? > Yes both of the above were said by me but other than that, they aren't really related. The latter comment was a hint about Open Skills and about Bruce giving a speech at the meeting. I thought it was a good speech by the way. > you could of course have a wiki page to announce new wikis. The wiki concept automatically handles the integration of a new wiki without needing an explicit announce page. You merely insert a hyperlink to bridge related concept pages. > seriously though, if I had to traverse 13 wiki pages to get > all the info I got by just being subscribed to 1 list, I wouldn't bother. But being subscribed to 1 list isn't really 1 thing, it is many messages to read every day. I've noticed a few people at Slug meetings saying, "sure I'm subscribed to the list, but I don't actually read it". > I like text based mailing list discussion better than web pages, Each has advantages and disadvantages. If you want to ask a question and get some answers or follow an actual discussion then the mailing list is good for that. On the other hand, the mailing list doesn't collect knowledge in an organised and accessible form. Yes you can search the list archives but there is no logical structure to those archives, the subjects often don't match the content and single messages may contain multiple unrelated issues that just happen to have all been interesting to those people at that time. The wiki takes longer to build and you probably have to search a bit for what you want but at least the wiki can be structured enough to make it easy to search. Once the structure is established it is good for all time. > I don't mind the occasional surplus gear posting, or side remark, but I can't > stand politics, flaming, or persistent off-topic threads. That is the other advantage of the wiki, nothing is off-topic provided it goes onto a page with a name that matches that issue. You don't like politics, I don't much like politics either but I've been dragged into it because people are trying to use software patents to take away my right to use free software. I'm not expecting everyone to be interested enough to join in with political discussion but at the same time, we can't start a new mailing list for every new topic that comes along. On the wiki, you just bookmark the key pages that relate to your personal interests (or make your own wiki page that links to stuff that matters to you) and ignore the stuff that you don't want. > anyway the real reason I replied to this post, > http://www.provenresumes.com/ > they really helped me get a good resume together, one that has landed me > several > interviews, I bought the $11 pdf book (this was several years ago, it's > probably > gone up somewhat) on how to put together a resume. This > woman has it down pat, I recommend it. I had a look at the website, seems to have a bunch of interesting ideas, kind of a course in "2-unit Marketing Hype". Probably a bit of hype is necessary if you want to get noticed. It does make me a bit nervous when the last chapter of the specialist "IT / Computer Resume Guide" covers the topic: "If You Are Not A Computer Pro" but then again, I'm an expert nitpicker. - Tel -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] running X11 app through Apache
On Mon, May 02, 2005 at 03:57:46PM +1000, Julio Cesar Ody wrote: > I tried to run a few GUI (X11) applications to run through my > webserver, but no success so far. They simply don't run, and no error > appear in any of my logs. Does anybody know how to do it? X11 apps need a DISPLAY variable set in the environment and it has to be a valid X11 display and the program (running as apache) needs suitable permissions. You can probably get something working by getting the web server to build a DISPLAY environment variable from the client IP address and then using the "xhost" at the other end to open up permissions. The result would be kind of ugly, well very ugly actually, but it could work. - Tel ( http://bespoke.homelinux.net/ ) -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] running X11 app through Apache
On Mon, May 02, 2005 at 10:05:12PM +1000, Angus Lees wrote: > At Mon, 2 May 2005 15:57:46 +1000, Julio Cesar Ody wrote: > > I tried to run a few GUI (X11) applications to run through my > > webserver, but no success so far. They simply don't run, and no error > > appear in any of my logs. Does anybody know how to do it? > The x.org guys seem to have even updated the Xrx browser plugin to > work with newer mozillas.. This reminds me, the vncserver program includes a small web server that sends you a java client and puts an X11 desktop into your browser. Vncserver exists in various incarnations but RedHat gives you the http://www.realvnc.com/ version. Please note the lack of security involved with this method, it does have password protection but no encryption. I've used it to run stuff over an ADSL line and it is livable, not fantastic. If you actually decide you like vnc then there are native clients for just about every platform that give better performance than the java client, you might also think about an ssh tunnel. - Tel ( http://bespoke.homelinux.net/ ) -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] ntp client test tool
On Tue, May 03, 2005 at 12:26:29PM +1000, Peter Rundle wrote: > Sluggers, > > Is there a tool on Linux that allows you to test if ntp services are > available at a particular server? You can do a fair bit with the "ntpq" program. Unfortunately, the usage of this program can be a bit complex, you will have to spend some time studying the man page. Some servers filter out which ntpq actions they let you do so sometimes you think you are getting it wrong but in fact you just aren't allowed to do whatever you are trying. - Tel ( http://bespoke.homelinux.net/ ) -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] gunzip error: invalid compressed data--format violated
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Thu, May 05, 2005 at 12:05:56PM +1000, Michael Lake wrote: > Hi all > > I'm getting errors trying to gunzip some files of xray diffration data. > I'm unzipping on a Linux box. The data was gziped on an SGI IRIX, stored on > a Novel server, > moved over to a Windows box and eventually to my Linux box. The most common destruction of gzip files is ftp in ASCII mode onto the MS-Windows box. If this is the problem then there is no way that I know of to fix it. Check that the size of the file is the same under Linux as it was on the original SGI box. - Tel ( http://bespoke.homelinux.net/ ) -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQnmILcfOVl0KFTApAQK3lw//fnNh0KCw86kTC0skSL4QFwjA6uZIdHCd ds6Pme5kDco6cTyWI5SqZl6xVERLGXHF4TsisvNjXvmdwWRn0B8y80o9wkSDzTSh tvnhZytq4nZj+nYTXO4ERVrpwNEC3nCdnMsVkUBVkbr8+PLg5KAIdUa1kD5MQ0gF nNLiDMUI8ZwDdrUdsr7jLwJqArZusagmsIsfOqBuV0KeE+C9WuLYcbV2ptBnPu4g NRouuKLyq7anbvMluLprjrL1LynsSHdwBi26Jhnz2wTsssgpRv0UJY3UiBQ6xkmY i6lpg1YEmGGrNzGZSWPAucbAlzS28WMFbEWLzROzz19Ksj96SHHqp5MOLvaVWo69 Uba9kgg/fTS2xqsKhhg0HQbeoHESJMf+3IzSPXPM6nmfyx8iQM3fflUGGI1yl8kA xpjhl5Oyuri4MdD9CMQMst0nywdgx/QWUVja3yCvQ6JsugsvxRFhdhYT9ZMneqvb k9ZE7m07FeLcgp3vCP7AErIk6F8xtgasggjh6qbSIEPFNaOvu3Jz6Ls0B7Kta62C I882WtuTqCuut0TIx36+CWJtELBFuqUV+kF/2VKvmjITFBM/AF6tfVtXdRmFYkB2 0xfpDnypfoD5y5w31XZ/NTS8Y3XgwQaZyoOiS1c7pqN+oqnA+hGoo/JV/IQth4df EMllPvzsFO4= =mInW -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] non-interactive password script
On Tue, May 10, 2005 at 12:30:00PM +0930, [EMAIL PROTECTED] wrote: > This works fine the first time, then times out when running consecutive times. > Does anyone know what is happening and why it times out if I run it twice or 3 > times in a row? It may not be cleaning up properly the first time through (i.e. it does not work fine the first time, it just looks like it does). I notice that in the smb.conf there is a chat script for samba to change passwords and they check for: *passwd:*all*authentication*tokens*updated*successfully* right at the end. Maybe you have to check that too. Also, passwd has some locks in it which cause some files to be locked during password changing operations, I think that lslk can detect the locks but I can't remember where they are. Another place to look is by doing a pstree before the first run, then another pstree after the first run and then diff the two, you might see a passwd process hanging around. Then you have to do lsof on that process and see what files it might be hanging onto. - Tel ( http://bespoke.homelinux.net/ ) -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
[SLUG] Sebastian Rahtz - UK Open Source policy maker - in Australia!
** Event at UTS on Wednesday 25th May 2005 ** - Forwarded message from Pia Waugh <[EMAIL PROTECTED]> - Date: Wed, 11 May 2005 09:29:32 +1000 From: Pia Waugh <[EMAIL PROTECTED]> Subject: [CTTE] Sebastian Rahtz - UK Open Source policy maker - in Australia! X-BeenThere: [EMAIL PROTECTED] Precedence: list Hi all, I've the pleasure of announcing that Sebastian Rahtz will be in Australia and available for a public talk on the 25th May (he is only in town for two days). Sebastian has a lot of interesting foo to talk about, and was involved in UK Open Source policy making for Government. - End forwarded message - For more details see http://www.slug.org.au/events/detail.html?id=240 -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Laptop recommendations?
I'm just comparing the spec against my Toshiba A10 which is probably obsolete but no doubt similar models are available. My main criteria were price and Linux compatibility. On Fri, May 13, 2005 at 09:46:13PM +1000, Michael (Micksa) Slade wrote: > - good linux compatibility. I want to be able to suspend-to-ram/resume. Overall the Toshiba A10 is pretty good. I was wondering why the wireless LAN was incompatible until someone from Toshiba who knew these things pointed out that the really cheap model (that I bought) comes with the case switch and lamp for wireless LAN but does not come with any actual wireless LAN (costs more to make two different cases than it does to just leave chips off the board). > - 1600x1200 display. I've been spoiled. I'd prefer something with ATI, > for decent 3D graphics, but it doesn't have to be the very latest radeon > or whatever. Bit of a downside here, the A10 is only 1024x768, very clear and crisp, good wide angle, not so good outside on a sunny day. Better models are available than mine. I've found that the intel graphics chips are pretty good, very compatible, play tuxracer and similar things very nicely. > - at least 512M memory comes with 256M but all laptops are expandable these days > - at least 5 hours of battery life (normal use, ie 0 CPU, display on, > HDD spinning). 6 would be cool. I don't mind if I have to have 2 > batteries at once in the thing, but would prefer not to have to remove > the CD/DVD drive to get the 2nd battery in. That will be very difficult on a modern laptop. My old gateway could do it no problems, the A10 gets 2 hours. All the CPU clock tweaking that you hear so much about doesn't actually get you much (maybe another half hour). From asking around, a lot of other people find battery life pretty poor on modern laptops. I hate to say it but Apple is ahead in this area (provided you don't buy the fastest models). Note that the A10 is a Celleron, not a Pentium-M (which is better), but the Celleron does have clock adjustment and those advanced features so I'm not sure what makes it bad. I think that part of the problem is that the graphic chipset consumes a lot of power even when it isn't doing anything more than displaying a static picture. > - at least a DVD-ROM That's standard on the A10 and just about everything else. > - decent keyboard and touch pad. The keyboard and touch pad on the > inspiron 8000 were fine. fastest keyboard I've ever used. On the 8200 > they sucked. The keyboard was too stiff, and the touch pad had this > neet smoothing feature which delayed its response by about 1/2 a second > and made it feel awful. I like the Toshiba keyboard feel and so far I have been unable to destroy it (which is impressive). Some of the keys have been moved around to make it all fit, so the "Windows" keys that no one uses are tucked away in the corner (and one is the ratpoison command key on my box), the tilde, insert and delete are all down next to the space bar (you get used to it). > I'd prefer a keyboard that had the cursor keys and the 6 navigation keys > in proper groups, much like some IBMs and Dells do. tilde and backslash > MUST > be in sane locations. That's narrowing the field a bit... the A10 has backslash above the ENTER which I personally find sane but preferences vary. Putting the tilde next to the spacebar is obviously a trade-off just to square up the keyboard module and make packaging easier. The cursor keys are good. > Yes, I really do want a touch pad. I don't mind them either, I just hate the "touch to click" feature and haven't figured out how to switch it off :-( > - built-in wireless OR 2 PCMCIA slots The higher models of Toshiba have built in wireless. Can't say how well it works with Linux. - Tel -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] EMF IPAQ
On Fri, May 13, 2005 at 11:34:06PM +1000, Visser, Martin wrote: > Basically a highly-conductive (read metal) box with no holes (or holes > smaller than the wavelength of the EMF you want to shield from). For what it's worth, that's a quarter of the wavelength (for non-round holes, you measure the longest way). For some equipment, they spray black, conductive carbon inside the box, it absorbs more than it reflects because it is NOT highly conductive. I presume you can buy the paint somewhere. With a shielded box, any unfiltered wire that goes through from the outside to the inside will puncture the shield and carry the EMF through. - Tel ( http://bespoke.homelinux.net/ ) -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Clipboards under X
On Wed, May 18, 2005 at 02:21:38PM +1000, David Gillies wrote: > I'm currently using Ubuntu Hoary (top distro btw), but I have one minor > annoyance. Up until I moved to Ubuntu, I'd been using Redhat/Fedora. > > The clipboard, regardless of whether I selected the text or ctrl+c the > text, it ended up in the same clipboard. I have exactly the same problem in Fedora Core 3 and I do seem to remember that it was not like this a few versions back. I'm pretty sure that fc1 was better. It isn't specifically an Ubuntu problem. It bites me when I run java in a web browser to access St George internet banking (which will work from Firefox with the Sun jre provided you also install the "User Agent Switcher" and set yourself to Opera -- weird huh). The bank statement won't allow you to "select" it at all but it does have a button marked "copy" which will copy it to the clipboard. But you can't paste that into emacs because emacs only accepts traditional X11 (i.e. the text MUST be selected). On the other hand, pasting the bank statement into gedit works fine (and that's about all I ever use gedit for). How anyone could have ever decided that traditional X11 cut and paste was difficult to use and needed to be augmented with something copied from Microsoft is beyond the furthest horizon of my wildest nightmare. That's the fun of free software, people really are free to do wild, woolly and sometimes really stupid things with it. They will have all on single button mice soon. Believing in freedom requires that you also believe in tolerance. - Tel -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] VoIP for home use
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Sat, May 28, 2005 at 12:49:51AM +1000, Simon Males wrote: > > I need some advise regarding Australian VoIP providers. I've spent a > good couple hours researching. Eventually I will go to a hardware > solution but I wish to try a software solution first (softphone). So > initially I wish only to be able to make outgoing calls, thus incoming > calls do no interest me. I've tried "skype" on my laptop (running Fedora Linux) and it is easy to install and gives a good clear sound. The lag seems to vary depending on who you call (average approx 1 second), if you follow their instructions and open up some UDP ports on your firewall (forward them to whatever internal machine is running skype) then you get better performance. The software costs nothing to download and calls to other skype users cost nothing. Calls to normal phones require you to register with their "skype out" service but you will save a lot of money if you call long distance. Skype is badly behaved with respect to network traffic, it actually runs a peer-to-peer network in the background and constantly connects to other skype users. It also probes firewalls (both your own from the inside and other people's from the outside), presumably all the network probe results get sent back to "skype central" where they track network topology and optimise their peer-to-peer routes. This is simultaneously a good thing (it gives good performance for the user) and a bad thing (heaps of information about your network is getting sent to a potentially untrusted third party). Needless to say, skype is closed source. It certainly would be good to see the open source world come up with an equivalent, I can suggest that if you just want to play around with VOIP then download skype and have a bash about with it... you get an idea of what is possible to achieve and it is about 100x easier to install and configure than asterisk (having done both). - Tel ( http://bespoke.homelinux.net/ ) -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQpffBsfOVl0KFTApAQJTIA//f2xVi5LwKH3dHsoqI2Q8NElMg1fBC/Ci mcKb168ESeL60oLi5dk/QBNv5Xk1X64Oqvnkx3LxL2IIuNCVEEhpLFhoYY/IL/8h kWA3OvW5Jf5qr0pm13Tw9CfR7IXpcIulxrx+/F0rGA2awxUjmT101DMBCDp1vyy6 rMA84guKDOHIYANkpaWjGOaUcvqjY7a6V/UfBBR/5CSPrarwmfsAnUjyLC9AKpaH tezQd1TcjZpJX7o2x8iUF7OQXUBh0Gp9xuJ/fCIiV7CpNcRrjvvvneMVkPETcA0w LWPfACqozxLbmNUaslMOOJfmQYccecBPdXVUmH5lf2D32GqEhCJmG47dCf3tPEsh i7sWuUnmT+QI8eJfHu7z+kzyMdUYbVeCITW5stpHZLyeQ2aiyv1WjU4sH1mAWRxQ bvsJyHEFQgAQYKjnoFy8jbXEKPkhuMqkM93T8ZVtvqdwrduM8qA5M+wQdXzv0FQs GiB+hpyU7cPVPSxxiPXz1/sHH7mPZKWYZUXWxrWS3KyUmf/nFbr6Z6L2dmTeav9p PN3v92b5bjXJpDOb7gi0e+p+BOZbgolxxjCgtGjtZ5DUCeaoeoXIy6BGjiYGD+PC lC4nGu9BH33OsNyYKCyuIAdffqzntIAvRqFpn5HJNRkpBZZcAc6+/3+HtIh8NguV YEmUXzOlkUI= =ewkB -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
[SLUG] Geographic Information Systems and PostgreSQL
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 There was a question last meeting with regards to PostgreSQL and the consensus seemed to be that it wasn't up to the job of handling GIS. I did a few performance tests on my laptop and didn't find it all that bad, I guess it would be interesting to race it off against other indexing systems on the same hardware but my results are really just scratching the surface of all the tests that could usefully be done. http://bespoke.homelinux.net/geo/ I did notice that the R-Tree performance is somewhat data dependent although I'd like the chance to go into that in a bit more detail. I'll add to the page when I've run a few more tests... I remember someone saying that they had the loan of a kick-arse server as a test bench (was it Matt? can't remember) if anyone wants to lend me an account on a test machine I can chug my test scripts across and give it a run to produce some comparison graphs (current test is only on my laptop which obviously has its limitations). - Tel ( http://bespoke.homelinux.net/ ) -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQp7QXMfOVl0KFTApAQKUuA//YFAgwEQuiX5VGgdtQ+BXc/G2AlO04o7c E38HwWGg0WRz/4V7GMHv+Fh2YoOQ9di6CaZR8NI7lEHHfbsWDOUTwG8FStQBKlRm aEPCBomykf4Qke1b/DB7VH0FZJwPmKjOTujwwFBDirTY1gmfGm9RvJ0IP4FMMdDO IHYNdZKEKv0WN7df8t9j1xC+bGEB08/zU//60aYauGOtbRbQDEUnChdMFJ5WqyXM Lxnf6oOB35hE14w8LZyc2cYvlwrgsKOtDkVKCQAIFr4Si9Gh5oQLaxIq6Ni+PN6t tzTj22sVhLaCg1VimSnLSLXRWyk3Q7cK2O/vozYkNBCY1kZ/QqpDzLbhiQiSEUY0 aAeJ0oocA0EbnMN6pP+wyXg5A9LLqdXJyhr2PPKAfZhhRCfcMW0ST7Vcn4uZ2uLt YvITQXsxKlf0z2ZdikMifZ1kVXhD7rP2UKWHyyRazmzaSgpeHsUgT/jNySrwIAz7 o1UWB/UPcQ94CdxAzu7LeXg/RzED9flllA+HxWCwYlQ5riMeQ2FTNa8qAK0PoflV dC+cC2m0eVFGDTS8ianSGF8NZowj9mMVHu5cpBnobY2aUtyV5/2G975iGx8BK7Wa jArk4RQvhZILfT1uIPnR9nYyGDJPBD4YVkH6NUVtiahbCPduTFaTcnoSf4u6pOe5 L5UvGVyLxio= =P96N -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Linux System Administrator Wanted
Not hassling out the job or anything, it seems like a good offer. Just one thing I can't help commenting on... On Thu, Jun 02, 2005 at 11:48:09AM +1000, Steve Waddington wrote: > Ideal Candidate > > - Is fanatical about system administration > - Has worked for an ISP in System administration for the last 18 months, > taking on all work and responsibilities > - Can demonstrate competence in all 'skills needed' areas > - Is now looking to take a lead role > - Is a reasonable person ~~ Why is it that every job description has to have the word "fanatical" in it somewhere, but at the same time anyone who is a genuine fanatic is treated like an outcast for being unreasonable and hard to work with? Why not just make it clear, "We want exceptional results but are unwilling to suffer any inconvenience in achieving those results." - Tel -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Linux System Administrator Wanted
On Fri, Jun 03, 2005 at 02:31:59PM +1000, James Gray wrote: > I've been called many things where I work: > Security Nazi > RFC Zealot > Unix Pig > Linux Whore > Rack Monkey > Packet Sniffer (I jest you not!) > Sneaky Little Bastard > Hey you! > ... > > Never been called a "fanatic" though...maybe I'm not qualified? Don't forget you have to also be a "reasonable person" at the same time. Surely a reasonable person is by definition not a fanatic. - Tel -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Telephone recording?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Tue, May 31, 2005 at 09:13:20AM +1000, Angus Lees wrote: > At Fri, 27 May 2005 13:04:21 +1000, Jeff Waugh wrote: > > What's the most sensible and reliable way to record a phone conversation, > > assuming a standard phone, using Free Software (and probably a bit of > > hardware)? > > You can get phone audio into a computer by using one of those old > voice-modems with the right AT commands (often not full-duplex, but > that isn't a problem here), or a telephony card with an FXO port. > Whether something like asterisk will make the software side easier or > not, I'll leave as an exercise for the reader -- but even using > asterisk for this step wouldn't require anyone to use VoIP, assuming > you had a PSTN double-adaptor thingy at your end. I have tested this with the Netcomm "Webmaster" CD1800 56k V90 and yes it works and the AT commands are moderately well documented (not by Netcomm, you need to search around). Roughly it is something like: AT+FCLASS=8 AT+VCID=1 AT+VSM=128,8000 AT+VLS=1 AT+VRX Then modem sends "CONNECT" followed by a stream of 8 bit unsigned samples. There are some other bits of stuff encoded in the samples as a DLE (0x10) followed by a single byte so you should strip those out afterwards. To hang up, you send: AT ATH The exact numbers after the "=" depends on the modem and what sort of codecs it supports. The CD1800 is about $70 retail and supports only one codec which is 8 bit unsigned linear PCM at 8kHz. Voice quality is reasonable, not fantastic but better than most mobile phones. You need to split your phone line and feed it parallel into the modem and the handset which seems to make the handset a little bit quieter (but maybe I'm going deaf). This is about the easiest way I can think of for recording a phone conversation to computer and sox will convert the resulting file into something more presentable: /usr/bin/sox -v2.0 -r8000 -db -fu $infile $outfile lowp 2000 or something similar... - Tel ( http://bespoke.homelinux.net/ ) -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQp/lVcfOVl0KFTApAQKZ0Q//dbL9eNDz3whX4Do6q1L6mjn3pO8suZss Wuf3F4VMdS3NGkTYwwBaRhksq6D67knrtyDFFKnCjwmCIpRgI1qW1tMQ6hVSQYX/ eUzveFgQtzoAP2TjtpNXOMvcUb9ChZnjW63aWkLIgXw9Cv4s884TMt6XJYaJoLN7 XkEu73kdGmkxBDVyWtI0I7Xd3fKzyKOnorbhE/3ttVJfomDblnOdf7CpGlVRF0DV yMTzwcgWNfX+pWl2DdPeLVO8ujrmWpaZ02n2gnYigezvYLV2G5KzITU7byc6Ah0j 7BQC3QkBMUXELewHJOoLaoKAh85eeiBIuf06SUZXdmEJHe3lkxDcutukUsLfbJEI Bfy3kc6pGnwuLJbR/r53eivZiFtgGPxMUiIQwUmBVKV63s6LJaKMlh7yjsULQl8j zJVfvbcqDetHsBIcqgkWaT2uVMjWfxjDnFu5u+oV9lj+Eqp8hAaiG7a8BVt/odf4 UnOn/KkGrE8k3qdTLPJjvVgV2lOnyfqRPPi+QO4N3WskMVBRm4Zt0k1bnDmq8qsf YhVf9zRY1nWTP8l6SQd0ah8E+6NCFK9Td3t+4V93ZwboMP6T3UD+dOdMqdIWW0Dq zI9zmZnWjQOJP9uYyx9cKgBTI/PiRzSJmP0929MiZMrc05dACU0WRYqMKVy4cc6V EAcTjVQt6aM= =Mpy8 -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] SQL Brain teaser...
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Wed, Jun 08, 2005 at 10:02:34AM +1000, Grant Parnell - slug wrote: > Essentially we agree on only looking at times of the day where packing > activity is occurring. The jury is out on whether you can assume the > orders were processed sequentially based on docketdate. Due to the nature of SQL not being a turing-complete language (i.e. you can be SURE that a pure-SQL query will always complete, possibly after a long time has elapsed) it turns out that it is much easier to detect the gaps between activity that to detect the overlap region during activity. For example: junk=> select * from orders; ordno | pickslip_printed | docketdate - ---+-+- 1 | 2005-06-01 14:32:16 | 2005-06-01 14:34:47 2 | 2005-06-01 15:12:27 | 2005-06-01 15:27:26 3 | 2005-06-01 15:12:28 | 2005-06-01 15:30:25 4 | 2005-06-01 15:12:29 | 2005-06-01 15:21:53 5 | 2005-06-01 15:41:29 | 2005-06-01 16:12:07 6 | 2005-06-01 15:41:32 | 2005-06-01 16:11:45 7 | 2005-06-01 15:41:33 | 2005-06-01 15:52:17 8 | 2005-06-01 15:41:33 | 2005-06-01 15:49:15 9 | 2005-06-01 15:41:34 | 2005-06-01 15:48:30 10 | 2005-06-01 15:41:34 | 2005-06-01 15:45:56 11 | 2005-06-01 15:53:00 | 2005-06-01 15:57:57 12 | 2005-06-01 15:53:00 | 2005-06-01 16:01:33 13 | 2005-06-01 15:53:00 | 2005-06-01 16:00:24 14 | 2005-06-01 16:02:25 | 2005-06-01 16:04:00 15 | 2005-06-01 16:02:26 | 2005-06-01 16:08:09 16 | 2005-06-01 16:22:40 | 2005-06-01 16:34:49 17 | 2005-06-01 16:22:41 | 2005-06-01 16:36:26 18 | 2005-06-01 16:22:42 | 2005-06-01 16:37:52 19 | 2005-06-01 16:22:42 | 2005-06-01 16:25:59 (19 rows) junk=> select o1.ordno, o2.ordno, o2.pickslip_printed - o1.docketdate as gap from orders as o1, orders as o2 where o1.docketdate < o2.pickslip_printed and not exists ( select * from orders as o3 where ( o3.docketdate > o1.docketdate and o3.docketdate < o2.pickslip_printed ) or ( o3.pickslip_printed > o1.docketdate and o3.pickslip_printed < o2.pickslip_printed ) or ( o3.pickslip_printed = o2.pickslip_printed and o3.ordno < o2.ordno ) ); ordno | ordno | gap - ---+---+-- 1 | 2 | 00:37:40 3 | 5 | 00:11:04 5 |16 | 00:10:33 7 |11 | 00:00:43 12 |14 | 00:00:52 (5 rows) junk=> select max(docketdate) - min(pickslip_printed) as whole_day, count(*) as work_done from orders; whole_day | work_done - ---+--- 02:05:36 |19 (1 row) junk=> select '02:05:36'::interval - '01:00:52'::interval as neat_time; neat_time - --- 01:04:44 (1 row) On the basis that the total time less the idle time must be the time spent working you can say that over the day you got a usable output of 19 orders processed at an input cost of 1 hour and 5 minutes spent working. That's an average of 3 minutes and 24 seconds per order. It's a backwards way of doing things but at least it can be all done in one single query (give or take a translation from my PostgreSQL code into your MySQL database). Note that the query that finds the gaps still crosses the table onto itself and then onto itself again so in the simple case it processing time will be of cubic order. However, you can safely work with one day at a time so if you build a view (or temporary table in MySQL) that has just one days orders then the cubic time probably won't hurt you too bad because the entire temp table will fit into memory cache. I'm sure that some clever index structures will speed it up. Of course, if SQL supported a "running total" aggregate function then it could be processed as a sequential event list but then the "order by" clause would have an effect on the data in the output lines and I think there is a rule that makes such effects illegal. It just goes to show that while SQL is a pretty handy database language, it does have its limitations. Which is another way of saying that procedural languages will never die because ultimately some problems only solve easily by using a procedural language. > We'll never know, based on this data, how long each specific order took > but then I don't care - baring exceptional cases (eg 2 hours). Looking at the average over a day is probably the most meaningful figure you can get. You can average by number of orders processed or by number of products picked or by some fugde-factor algorithm that combines both. - Tel ( http://bespoke.homelinux.net/ ) -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQqe+V8fOVl0KFTApAQIhcRAAmnqVT/LkQt9qEknCCG20i4U/+oz5+SJ8 0pTWaOo+308k8eezoF30N4tbJpkFWWKU1HQ/fg4lqcPlZnQbHml+DwFkOzrDQja+ Y+gRQyztzJUmoKEsojHoWidWka12kqrVqaJjMpxgVI9UMfMGqkgWKTJ8aay0S3RK 79p1YCJiBdjrsUT9Wj3BSXqWmViZfX6A8OXfWaet5uZCEMZjJT5TUUcQSvX5Lm5Y KvjlxJFFZyO2GaWdmJtH7zS1ONRSZj31WUzSkn
[SLUG] Why XML bites and why it is NOT a markup language
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 So JJJ is offering an RSS newsfeed for the "Hack" program and the URL is here: http://www.abc.net.au/triplej/hack/podcast/podcast.xml So here we have data made available to the whole world, put the data into XML format, perfect for compatibility, interoperability and etc. etc. So let's look at the practical example of this stuff really working shall we? If you look at the header of the file you can see: So no problem, I should be able to use all the existing technology to work with this stuff and gain the benefit of code reuse and all that good stuff. So I use perl and libxml -- proven technology. I can fetch the file easy enough using wget, no problem so far. OK, here's my microscopic perl program that reads the file, I'm including the entire listing just to prove I'm not doing anything other than the obvious: - -8<- #!/usr/bin/perl -w use XML::LibXML; my $parser = XML::LibXML->new(); $parser->recover(1); $parser->pedantic_parser(0); $parser->validation(0); my $doc = $parser->parse_file( "podcast.xml" ); my $root = $doc->documentElement(); - --->8--- You would think it would work real smooth but what result do you get? podcast.xml:39: parser error : Input is not proper UTF-8, indicate encoding ! Bytes: 0x92 0x72 0x65 0x20 All this week we re looking at the unofficial mental health system; ^ Sure enough there is a high-ascii item in there and some Mac user has no doubt used a proprietary bingle-bongle encoding for a single quote even though there is a perfectly good ASCII encoding for the same. Never mind blaming the Mac user... they do that sort of thing, it's a fact of the universe, nothing will ever change a Mac user. However, what really shits me is that the XML parser dies totally and completely when it hits a single high-ascii character. This is with the "recover" flag set, and both "pedantic" and "validation" switched off. Basically it is running in the most lenient possible mode that it can possibly operate in and a single bad character still nails it. Yes I realise that in an ideal world the tag would contain encoding information and yes I realise that in order to be correct UTF-8 it must encode characters above 127 in a special way and this encoding does not conform. OK, we don't live in an ideal world, the document has a problem. Having established that, how am I supposed to read it? man XML::LibXML::Parser search for "encod" -- nothing. Looks like the error message wants you to "indicate encoding" but the man page does not tell you how to achieve that. It explains that you can catch the errors using an eval block, but having caught the error there is no way to follow through and finish parsing the file. So one single byte has rendered an entire file unreadable... what a fantastic protocol, so good for inter-operability, so widely compatible. And here comes the gist of this rant... A markup language MUST be robust. Anything that claims to be portable and all-purpose and the document processing format of the future simply cannot be destroyed by a single bit-flip on a single character. The fundamental difference between a programming language and a markup language is that a programming language can have parser errors and syntax errors whereas a markup language cannot (by definition) have any errors at all under any circumstances. The parser for a markup language must be fully robust to all possible inputs and although it certainly can result in various severity of WARNINGs but nothing must stop the parser. Fundamentally, XML is crap as a markup language because it simply isn't possible to build a fully robust parser. Worse than that, you can't recover state (even approximately) in the presence of a damaged document, XML is brittle, as brittle as any programming language. Let's make a simple comparison... suppose I do all my data transfer by simple tab delimited ASCII files with one record per line. If a line gets damaged, I might lose that line, I might even lose the line after the damaged line but at least I have the rest of the document. If I jump into a plain ASCII file at a random location then I can scan around the local area until I find the end of a line and I can resynchronise to the local records. This technique can be used to perform a fast binary search directly into an ASCII file that is sorted by line -- can you do this with XML? Of course not... your basic parse-state is broken the moment you seek to anywhere at all, and that state is perpetually unrecoverable because something that looks like a tag can exist within a string or you can have a CDATA or some other stupid thing. There's a philosophical argument about whether brittle languages or robust languages are more useful for data transfer. Some would argue that if you are sending commands to a mars
Re: [SLUG] Why XML bites and why it is NOT a markup language
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Thu, Jun 09, 2005 at 09:13:49PM +1000, Jamie Honan wrote: > I think the key is 'validating'... If you checked my perl example, I specifically turn validation off in an attempt to get the data to load. It didn't help. > > I've got three answers to the above. Most importantly, you don't > > use markup language for talking to a mars rover... you use a > > programming language and we all agree that programming languages > > are brittle and always will be. Another (still significant) > > point is that you can always take a robust language (e.g. simple > > TAB delimited text file) and make it robust by adding a CRC or ~~ > > some sort of signature system... you cannot take a brittle > > language and make it robust. > > I'm having trouble parsing this :) robust and robust? Sorry about that, should have said that you can take a robust language and deliberately make it brittle by adding a CRC. There's nothing you can add to a brittle language in order to make it robust. Thus a "standard" data transfer language should be desgined to be robust and then have the option for users to add a CRC or md5sum if required. > That's got to be a different issue. You parse after you get > known correct data ... How is that possible? You have to parse what you get and then figure out what it means. Having a program that believes its inputs must be known correct before it will do anything seems to contradict the most basic rules of data handling. Thanks for the links... - Tel ( http://bespoke.homelinux.net/ ) -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQqjF28fOVl0KFTApAQL6UxAAj5J+Ef8u+ksKBtzBPY/7EfqZx/U1AgSv D+PcFC/cRCrTV4FdJZ7qCsWF9eoteffYPOfGOYz4Juz5k0mGjwJo+bipRy4WAjii Wgzv6Mz8CdXveV//VCdqZZsoboGQc7nr4aWMBltHKKkf3fZsqoikVtwNxrIDiGAe eCRFR8v6NtzcxI0Pk5Jrya9V2i91wpCzycyIynFJTd3jDguy3tWtULQ9u1KWVrJ1 E81c5qBsLfp0E6Wmj7861HNhGpIj8moi6lnMlPABhM01M+9jDQwoH0MrsIk9Y3x3 juibav2OEQb83aoHWrqZQKIftUDbNpje+ajDL1r/F75xl9xU9pA7fVsfJhQfNS+y 85SMbMxDBlFcIUA94y3Eqhtn8Be+/y2TO7+0URLZetKf34jjfXqiLLQoJgK4WddK y3EsvYYAZGe8f8fX2Rm2bkmJ7h2v8s60fcp30OyOC7IoF0mV3k39ACsZnkoIHWfj D4ipYVw+AC7zUGZ+mvEiqzzHxpsu4DgF9aO9b/HZhJaG+zjJu9d1ZYjOaXFEW32K h0flILtzx/brAIn0UrwXPh+mklPm12uL9ULOnmUbCBcQ9jGtswlpBMV2fvewqUVJ ElKLv09ICnVWPNWVsSXB3i8wPgUJmUk81jmJAzNP2FLXSuR1WwETCE0shqN0Wrfi DF0+uY0Lurw= =EVH7 -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Why XML bites and why it is NOT a markup language
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Fri, Jun 10, 2005 at 02:14:05AM +1000, Jeff Waugh wrote: > > > > > > Keep in mind that all forms of RSS are absolute abominations, most feeds are > completely broken, and it has not encouraged anyone to use XML properly. RSS is an example of people trying to make XML work for a real-world application. No doubt someone read the XML hype and figured that it would be good for the purpose of data transfer. > XML is > quite good in general. RSS and all its related muck (as well as HTML if > we're being honest) gives it a bad name, given that most people experience > XML through one of these formats first. But HTML has been amazingly successful in the "real world" where imperfection is a way of life. Generally speaking, if you send a browser bad HTML you still get some sort of rendered output... usually not too far from what you expected. It seems to me that XML is only useful when the same programmer has control over the sender and the receiver. For example when a program saves to XML then loads from XML again (such as gnumeric). This doesn't constitute data transfer, it is really just a library that makes it easier to save and load structured data. > Look at Mark Pilgrim's feedparser (used in Planet) to see just how stupid it > all is. But what XML people call "broken" is more accurately described as "normal", the whole idea of a markup language is to be flexible and to have ways to cope with minor incompatibility between sender and receiver. - Tel ( http://bespoke.homelinux.net/ ) -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQqjJ3MfOVl0KFTApAQLZUw//asb1PW7H4yOVL09cNuaekT6Kf7GJVJX1 5lvhG4aI2HizYFqwPv22BcsTu2H74xMDfQVLQ1qWTdIIccLItWih53fh5weWmT6B FtI9BHaoLDlzGXY8ipggQvTpx8hDowDduRkTFaH+cfuF1obgoeFgkKZGDWgd48+x ZyzjrD6WVsYpKqucqoNaqLnptGNxFFWu4vpfXCwbd/EgRQm3XsR+Ix7Q5Nfr5h1X dRYhbewMSb360rQbT+jgSCltu0omJfPzNb2yLfvpWXk0Yi+bZlo6AslgtB4+At5a Xij5mwbZorP2HTX+GBuYYgza1u22wrcf2SzBL8J/iKnmhIfbTc6aT0h5n9xbzIWB jWYDB+yFk7lRpXT4LSqulTSLYihbYTXq2NMLn+hOBP9nzSRd428EsAGEIa2DZvsh YlK5f8QqeBjKT0ylx5aOi3X60sn5GIwwrH6G/bTdr1ybRjI3GxWS1I3Iu/L5wPjt msxAEeTij63G66hjlbGI+opyCqzAem1gkcwy5iOBz0PtU1ERerbdBCNnlyIwWPHW rWW6b3ZzSAPMh5azqHQmQheBLqMx0GG1ONTMviz/G7PY02EusPms+FkZWNN/sfkg WUyG3V+YWWL7RpvB5DLMuxR3YgcRjD/Rh385ALLyW/AXius7EwVVe8FymE+BWEzx 4Tu1EdyBuis= =1jLn -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Re: Why XML bites and why it is NOT a markup language
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Thu, Jun 09, 2005 at 08:34:53PM +1000, Jan Newmarch wrote: > Don't confuse the language with the parser. By this criterion even TeX > doesn't make it as a markup language - I have had many, many files fail to > parse over a single error. TeX is a programming language, written by an expert programmer and almost exclusively used by programmers and tech-savvy people. Don't get me wrong here, I like TeX, it is very powerful, gives top quality results and handles equations better than anything else but at the end of the day, it is very brittle, and it is a programming language pretending to be a markup language. > One TeX parser I used even had an "extra help" > key: when you pressed it, the response was usually "I've given you all the > help I can, if you can't understand the error then see a human." The onus > is on you to fix the error, not on the parser - automatic error recovery > isn't as easy as it seems. As a particular case in question, TeX is a macro-based programming language which are the hardest things to debug and as far as I know there is no macro language that gives even moderately accurate error messages. In TeX you get errors like, "this token was not the sort of token that I thought it was going to be" and "something seems to have gone wrong back there". Imagine writing C by using #define headers that go 4 or 5 levels deep and never using any functions. I promise that you won't be able to debug the result. That's why TeX, powerful though it is, will never move into general usage other than (maybe) as a back-end processor for some other typesetting package. Can you imagine trying to explain to some boofhead warehouse manager how they should handle an error message from TeX? > So as a human, I suggest you change the > offending char and reparse it - and send a flame email to the originator. I can do that but I don't want to flame the originator too hard or they might stop sending me any more data. They aren't under contractual obligation to send me anything so although what they are sending me is wrong (in the strict sense), I'm glad to be getting it. Sure I pay for it out of my tax but since when has that given me any rights? If I change the character this once then that's fixes the problem once, since it is a news feed I'll just have to face the problem again which means what I really have to do is write a program to fix the data. In effect, I'm writing a pre-processor for the XML parser which makes me think, "how good is XML when I have to write a pre-processor for it". - Tel ( http://bespoke.homelinux.net/ ) -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQqjRJsfOVl0KFTApAQKFYw/+OSE4++5wZ7Oc2pLkOnSUpemBdXWB2Q15 jMXjzOEhm6D38NLUHSM+bnNyrwbZcrcPjZhY+5gwKsKrzADX0tgRIXjhsWxrhu1n 6WxHJlp0q+xdVL2tmJrmfGt2ccrDHrEONlF4KdOavYq7dLBGRrrrPsrkoCjc6wHu qu8BE74lpjIBEJErAau3vdvBoDhkbSnGqS6BzNCfdw9eTcjEuC7O84RxVvjMC1v5 EkFIdY/lXowPzmHXw+/3qPDLK7v68Eb2b0rASdo3JxCfEqs1rVe2VvOYtEzlpkgB E4YQpZCfE6couToRrLMHUL5bpmpuALn27hYNQtaeDeCPA/1yrIYCephlIec1N2fJ jELP0J2Mlnr9gacNLuF75SK0LMBsImNSD2s1GqTMyTvbYNVIv95bAJ1e/7U03Pr0 erx5vLX1qwgs0bcSu3cgzC9TeRWw8kogJ5kOrx2017Y7Rier13RR3+Njxz6lQKVZ rSRG+zf/JxBnEB/y8X0fWwxc7zMpyr4VqImnLfWE0dZXC5Tb4UN6A7n8195dPi+N 5mImzC1whh6vFfrPYqkQ6rcGZ5jlSPFp4alSH9CYeYP75mZz8a7AKv+FOCCVe6an bSnV3MgSzWc59PQ8oVAZyh8WyVsAICXNuP6VwoUfMqwljkVu4wTDx2iQNqAzQ/0l Ag6aXGOIKdU= =J13b -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Webcam driver questions
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Thu, Jun 09, 2005 at 07:55:55PM +1000, Leslie Katz wrote: > The documentation that came with the driver did refer to a program, > spcagui, that can be used to test the driver, but I'm told that to install > it, I need libjpeg, libsdl and SDL_image and I don't have the confidence to > know that I'd be getting the correct required files. Certainly, they didn't > seem to be available from the Fedora download site. Under Fedora-3 you want these packages: SDL-1.2.7-8 SDL-devel-1.2.7-8 SDL_image-1.2.3-6 SDL_image-devel-1.2.3-6 libjpeg-6b-33 libjpeg-devel-6b-33 They should be similar for fedora 4 (with different version numbers). All the SDL stuff is capital in fedora (e.g. libsdl* is libSDL*) and I don't know how normal that is or whether it is compatible with other distributions. - Tel ( http://bespoke.homelinux.net/ ) -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQqjUAcfOVl0KFTApAQI3GA//fZmId4GUej9ZLXiMS1FFL6R54wfQJDAr 9/n4vWbjPqIhaohK57PK6B92wWAVILBuHbAn3ecxugsZ/g8+St57ByWIZbF2y1+P +7dm4t1jtkw9Qap5XIhAO8uAAX/byZoHjCUAnd1gpQYvoeFBin61IupCAoD3+r1h CiRztMep6RPP3G6iXnsVpxOqK564mNeF6gSTXartnI78nRUeNLg9C4nIsFu6cC2x m0g0CMVkS/lxXum2zM2/hf3Y9uC9YlmYvJmyNzqAONTqsLXoMZiztHjPja/dod73 w0g3aJ58c4C7srMgMPNQACP1HE+tbeUM7R3bjLPXACgvdVngr32axfLYuoup+QS9 TUl/dwv79+AeH6zpyPz3VAe47mg8JTMWXCGh0uTred0NEWB2QmANwsnVXTpYk+WT gCN9J8lB9LuEd5jtPNoP2Wna4+UkbwWblMNeieTmfPUN4O6hjp2g+pzLvBBEE8aC woDCHHQ9/jqINoItdYBKp8x9GpQm2/kWQgbXtB1cvfQyQ0E1w30WdXhhm5cZkt82 +ecTUopAo3tlpXc/zoV8qTy8oHbC0lBVeMVJ2LFSyP+l+Z8FAcT0D9XPEJmtfdHr 2lTbPNfBDdC6OjMEqJ5xjw2IOT2nlWu3hCiOeDPJqqhs2h4vy78gj65aDQhP2IrY d/WxIJ5Z30c= =SH7V -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Why XML bites and why it is NOT a markup language
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Fri, Jun 10, 2005 at 09:48:12AM +1000, Robert Collins wrote: > Uhm. Sure. Heres a Gig of download, your 500K of usable detail can be > found spread throughout it. I think that even with perfectly well formed XML you will find that the ratio of usable data to cruft is worst than 50%. More importantly, if a crap data feed is the only data you can get then the choice is deal with it or barf and die. Which is preferable? > Seriously, XML itself is no more brittle than your ascii file, its what > you put in it that makes a specific xml environment brittle or not. Its > just SGML after all - which is precisely what HTML is. The parser you > are using sucks - sorry, but thats the root of your problem. I thought that libxml2 was widely accepted, used by gnome, etc. I checked the manpage and nowhere does it say "this parser sucks", maybe I should submit a documentation bug? Hoping for a quick fix, I tried the expat based parser instead (which also has perl bindings) with the following program: #!/usr/bin/perl -w use XML::Parser; my $parser = new XML::Parser(Style => 'Debug'); my $doc = $parser->parsefile( "podcast.xml" ); Which gives this result: Unrecognized character \xE2 at ./test2.pl line 4. Same problem, less detailed error message... at least the manpage for this parser does explain how to set an encoding so the following program does work: #!/usr/bin/perl -w use XML::Parser; my $parser = new XML::Parser(Style => 'Debug', ProtocolEncoding => 'ISO-8859-1'); my $doc = $parser->parsefile( "podcast.xml" ); Sadly, this parser gives output in a different format so changing parser has now broken the rest of my program *SOB*. On the wilde chance of an undocumented feature I went back to the original libxml2 based parser and tried inserting options from the expat bindings: #!/usr/bin/perl -w use XML::LibXML; my $parser = XML::LibXML->new(ProtocolEncoding => 'ISO-8859-1'); $parser->recover(1); $parser->pedantic_parser(0); $parser->validation(0); my $doc = $parser->parse_file( "podcast.xml" ); my $root = $doc->documentElement(); Frighteningly enough, this actually works... Woo hoo! I got XML to actually work! > convert the (probably cp-1252) text into utf-8, then parse it. or set a > encoding in the header, it looks like the perl bindings suck a certain > amount. By the looks of it, the bindings are better than the manpage is willing to admit. I still don't like XML because it is nutty that it should screw up so easily. My feeling is that if this sort of technology cannot make things EASIER to deal with then might as well go with something that does. My vote still goes to plain ASCII with single character delimiters (e.g. TAB or one of the DLE/DCn set) because of simplicity. - Tel ( http://bespoke.homelinux.net/ ) -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQqjdsMfOVl0KFTApAQIUOQ/8DUq90zzBlQCWhzEWJOxTfQPOwcSgDh/H QCwuKyvGBTnzI3UI6YbdDZ+gsezw/LcH8ltrsyq2QspyxwV0bkcweP9Tb0smRABp Ooivq9mBGE5dIe3WysdObkh9d0ZOiUFqiFFCYIcJYEJOxCVBDyeOvRwW+31TefPx GZmImrIktJmHxBBrpqbH8onRNK02cz/s6zXcKENYA/ixy7KzLwtVZiK5NX939E3F yVYP5+OIgvU6fgGK8xeQOqOQyTAVUAzYOKsjarE7tvHvlGg84MwGGaWkGIh7C5gz ZlvrE+dHKA8cgNfQl+goTLA214cKx3kAdADmEFVsz/509sacesKIlZAdOuuUDKxH rC+Cl0akqhCtr9L8pd57LGMHNDmZxFmMXYRXAuzS3/WrrqnFaOVpjJ2yTpPVmR+r 4z6VGjg2SZFYa5YeXlCNsjm22bHpszNgUpv0UyBaE4v9egCf+EX3P1GRvCLkdGmK mzjViA9qsO856uFCSsswWa/WmPrf9/Me3Lj3dxtp8uYKlqsaxGFlsoj7bGgBDMAU tRB0UBS2npOMjzF4dODXh1VesFVLzHC5mY7UFOo9Em4aaodID3cjId+6Um+TsCI6 miZFBUA7EswTWRxdMNeo5xPpl9xLNeJOlfuUyBqCptB0VqSWhRAj+k9uUyh3t9li StX7iRY3As8= =Yp9f -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Re: Why XML bites and why it is NOT a markup language
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Fri, Jun 10, 2005 at 11:04:07AM +1000, Matthew Palmer wrote: > The problem with XML isn't that it's a crap language, it's that people are > very poor at following instructions. When a spec says "thou MUST do it this > way", instead of doing it this way, people think "that's not important" and > don't do it. People are the constant part of the equation here. What you say is correct and also immutable. People might happen to mostly follow a spec if that spec is dead simple and following it is also an easy thing to do. The XML specification is overly complex, difficult to understand and there are lots of subtle ways to get things wrong... end result is that it is a reasonable expectation that most of the things you get with a "*.xml" filename will not exactly conform to specification. If they do conform to specification then most likely at some random future date someone will press the single tick key and what you thought was working will fall in a heap. > I'm not sure whether the problem is basic human nature, No, the problem is a refusal to work within the confines of basic human nature. > or because we've > been conditioned by so many really bong specs to ignore anything that > doesn't make immediate sense to us... And that too. > As for the comparison with HTML, web browsers have been written to accept > random garbage and try and make something useful out of it because that's > what the web consists of. Correct... and that's what makes HTML successful. The whole "world wide web" thing simply would not have happened if we started out with something as strict and breakable as XML. > While it would be theoretically possible to do a > similar thing with XML, it's a lot harder because you can "guess" what to do > with bad HTML because of the limited use-case of HTML -- describing a web > page. For XML it's a lot harder, because you can't make any assumptions > about what the meaning of the data is that you're parsing. Then we need to accept that XML is not particularly useful and we need to start looking for something better. I'd like to coin the name "RML" which stands for "Robust Markup Language" which should have the following desirable properties: * stream-oriented construction * byte-oriented construction (no 16 bit encodings at all) * supports arbitrary tags * supports parametric tags * never allow tags inside a tag definition * NO guarantee of tags making a perfect tree (but parser can provide information about tree or partial-tree structures if they exist) * when tags are all next to one another, ordering is NOT important (thus italic/bold is the same as bold/italic) * at most one parameter per tag and not named parameters (because named parameters bend your head and get very complex and require special syntax and further because it is always better to introduce a new tag than introduce a new named parameter) * supports guarantee of resynchronisation to tag boundary after an arbitrary seek into the file (scanning forwards or backwards) and something that "seems to be" a tag boundary always IS a tag boundary * case insensitive tag matching (for English at least plus any other language that sensibly defines mixed case) * damaged files can be recovered by an automatic process at least to the extent that lost data is proportional to the amount of damage * don't use closing tags at all, instead use the single parameter of the parametric tags to "update" that type of tag. e.g.: blah blurg is not good because knowing the font of "blurg" requires information going back an arbitrary number of tags earlier. blah blurg is much better because scanning backwards until you hit a tag will guarantee you have a full understanding of this parameter. In other words, you don't need to parse every document all the way from the beginning and thus large documents become managable * non-ascii encodings are passed cleanly up to the application level which can apply whatever translation it feels like doing (transation libraries might be an optional overlay after parsing is complete) * non-ascii encodings can never break the basic tag structure, so the parser can detect interesting encoding anomalies but can still continue to scan the file That's my wishlist... probably won't get done this afternoon but at least it is down on record so that when everyone is old and grey and some young guy says "I've invented this new tagging system that fixes the XML nightmare that has plagued the world for so long" I can give him a link to the Slug archives and say "told you so". - Tel ( http://bespoke.homelinux.net/ ) -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQqkYvcfOVl0KFTApAQLAQQ/9EyQ592wj1SpcJnaR9NG+6r5b3ZymJG7w lbo1wmMSQurCIoVqv2xsI8tQGhRAR4E2TFscPJq4Dm0oOq6BcDuRXD1/2CGvEX4G 4nOpTkxfTEGx
Re: [SLUG] the parable of the mudpile...
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Fri, Jun 10, 2005 at 01:56:54PM +1000, Rowling, Jill wrote: > Systems Engineering used to be a compulsory subject at both UNSW EE/CS and > UTS EE; clearly it isn't compulsory everywhere! > Unfortunately most small businesses (includes many telcos, computer game > developers and some dot-com survivors) are unaware that the cost of a > project is inversely proportional to the effort put into the initial > specification, and just cannot understand the leap from small projects to > large projects. It isn't necessarily anything to do with the INITIAL specification. Everything starts life as small (in the commercial world anyhow) and small projects don't need much initial specification and that is probably a GOOD thing because otherwise the small projects will not get off the ground at all. I've worked at the military side of things and they are the complete opposite, they believe in MASSIVE specifications. The problem is that because the massive specification is written up front (before anyone really knew what they were doing or why) it is guaranteed to be wrong. Everyone knows that the specification is guaranteed to be wrong so they try to build contingency into it for expansion and that makes it even bigger. Later on in the project, the pain of changing anything that might break the specification is so great that there is no room to get in there are actually make the stuff work. Getting back to the point at hand, the real issue is that as a project gets older, at some stage a specification must get written and the project architecture (whatever that might be) needs to be documented. It is probably a great idea NOT to try and do this up-front, or if you do it up-front then don't try too hard to make it correct and always be willing to go back and admit you were wrong (because you will be). On the other hand, it is a terrible idea to try and not do it at all, and that's where so many people make their mistake. The attitude is, "I got started without a design document, therefore I never will need one". When it gets to that stage there is a serious problem. A design document is part of the overall system documentation and it has to grow along with the rest of the project. The other problem (deeper than all the others) is that companies hire managers based on how they dress, who they drink with, who they sleep with, who they went to school with and a host of other parameters up to and totally excluding what they know about the tasks that they are supposedly managing. There is a belief in a "generic manager" who can manage any task simply by juggling numbers in a spreadsheet and does not require any task-specific knowledge whatsoever. Convenient though this belief is, it is demonstrably wrong. This belief has probably caused more pain and suffering than all the terrorists on earth. > The usual result is they cease to be in business after a while. Well you only have to look at Microsoft to see where selling high quality software can get you :-) - Tel ( http://bespoke.homelinux.net/ ) -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQqkg3MfOVl0KFTApAQIxXhAAnTsN4Icdquiq8++O57oMhMy7fMjQW3v+ YYgb3ZFGJ9RpoxC1Db7FBfOGpcBEmovUiMA9vasNGmxe2G6SMPyh87B4cbZvq9oa Vfj/m0C3qkqijqbMJHn4km8M2q0ZUu3xNAhQtuwbf4znDKEKcGfcfyoDfvy3NW2v t9fGP/O6xUMsE0QTuEuUvn0cfYOWzBHmt+a5SIQ/oiJMWVsEVjawj2Jwf14+gvSA Saq9muvHnTT2973h6S1sCpqutWKJJbKKJZQyq8yYblvNG1UV6ZJ9Z8FXsQNJjVeR rV2Jce0lUDupr/GQnEaq9RX5iPPBvpzBiNWNB4tNfc5uNlkbeH5eLb0PEG9xpfCO Wm0gXWpGtqFgHJZ3E0omk178UJaMZHyVSIbXFHzA7WrO5/hlGPl+ZYCdyHkc8szG QWMwyNyII76/DimfrrWWnB1vfoVz4f8GLAASVvn+BMGyPRPQyMhDLwIl604gIsJI /YScqarSXin+Me9QSP+Df7HEdRTS9EqQvrK0FjU71/auvIxGCtLuSKgXHU+tp2dh nCOoT4JkpfufL769eCwpy+bl6ci8RSeNPfZHBr/ZhxyMe/k3aBR0VeHSSAseK7DA /i3fN9e4IIqPduxocaVxh+4Vno7pi+c1tM5H44XlqPPYjjNEUfeLASuIWk9SUhRG +nfm8GEr0L4= =Kwsj -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Re: Why XML bites and why it is NOT a markup language
On Fri, Jun 10, 2005 at 08:58:38PM +1000, Ian Wienand wrote: > XML is so useful because it provides such good abstractions. You can > define it with a DTD, whack all your data in it, walk it with XPath > and display it with XSLT and some CSS. That's not really an intrinsic property of XML, that is a property of any nested and tagged data structure. The general idea of data abstraction goes right back to the first C compiler, probably back to the first algol compiler, maybe further back. Certainly having tagged data is a good thing, I just happen to believe that XML is a poor quality implementation of that objective. - Tel -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Re: Why XML bites and why it is NOT a markup language
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Sat, Jun 11, 2005 at 09:15:27AM +1000, Jamie Honan wrote: > > I can't believe I'm defending xml. A good mental exercise to remind yourself of why you don't want it and providing handy straw men for me to knock down. > > I'd like to coin the name "RML" which > > stands for "Robust Markup Language" which should have the following > > Don't be bashful here, Telford. I suggest "TOTRML": Telford's One True > Robust Markup Language. The acronym has to remain vaguely pronounceable, RML can be spoken as "rummel" without too much confusion. > > desirable properties: > > > > * stream-oriented construction > > Stream is good, yes. PNG, and JPEG are streamable. ASF is streamable > and AVI suffers cause it isn't. However, not all data is streamable. You can imprint a record-oriented structure onto a stream format by using tags in the stream but trying to support a stream by using a record format is really ugly (not impossible). It is desirable to have a format that makes it easy to build higher level formats on top of rather than a format which is already so high level that it becomes cumbersome for ordinary tasks. > Thus xml parsers present data as callbacks (kind of streaming), > or a walkable tree. Providing you don't want to seek an XML stream... > > * byte-oriented construction (no 16 bit encodings at all) > > You mean no unicode? As opposed to no binary mode? UTF-8 is byte-oriented as far as the parser is concerned, provided you use 7-bit markers for critical synchronisation (like start of tag, end of tag, etc). > > * supports arbitrary tags > > Ah. You've just lost validation. You now can't prove your data > conforms to an agreed DTD. Not at all. Validation is a higher level function and should be treated as such. Building a layered architecture is far more reliable, flexible and maintainable than building a monolithic architecture. Thus the job of the parser is to: read the raw data stream; identify the tags; identify the data blocks and provide an API that gives access to these entities (and nothing else). Any additional data analysis is another layer on top and there are all sorts of ways to apply data-specific templates over the top once you have the parser results (not limited by DTD technology either). This could include testing for particular sets of tags, particular sequence of tags, maximum data size, data formats, data checksums and any other sanity check that suits the needs of the day. For example, a DTD won't allow you to test a CRC against the content of a record so you need a higher level operation to do that sort of checking anyhow; XML does not remove the requirement to sanity-check the data you are given. I have nothing against higher-level libraries providing additional services providing they don't crash the basic parse layer. Think about a bison parser with no error-trapping rules, what do you get when something goes wrong? An error message that says "parse error" and that's all you get. How useful is this when you want to know what went wrong? Not very. Not at all really. I've seen people using Java XML libraries that allow extensive DTD and XSLT validation so you can define highly complex data structures very easily. When it tries to read something that might have a tag in the wrong place it just returns "false"... sorry I won't read that because it doesn't validate, and I won't tell you why so have fun figuring it out yourself. The yacc developers and users already went through this exercise about 15 years before XML was even invented and figured out that just a yes/no answer isn't good enough. > That's OK, because I suspect Telford is talking Telford protocol > to Telford at this point. Maybe yes, but from my point of view XML is a failure, I've tried it and seen that it is broken and I can explain why it is broken. I'm just identifying what needs fixing. > > * supports parametric tags > > Lot's of people don't like parameters. They think they should be > in the data part. I don't mind, we are changing the world here. The parameters are optional. Data should be in the data part, metadata should be in the tag. I agree that the boundary is a fuzzy one but in principle... "if the user of the program can ever see this text under any normal circumstances then it is data, otherwise it is metadata". You will note that HTML forms break this rule but I believe the rule is as solid as any you wil find and easier to express than most. > > * never allow tags inside a tag definition > > Hmm. You mean no heirarchical tags? Not sure here. Or you mean > tags are atomic... Fair enough. Yes tags are atomic,
Re: [SLUG] Seeking User & Group Management Advice
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Tue, Jun 14, 2005 at 09:39:16AM +1000, Peter Rundle wrote: > Perhaps something like login as peter then "su - matlab ; newgrp > peter"? When I do this it prompts for a "group" password, where is this > kept and how do I set it? (man newgrp tells me about the passwd but not > how to set it). You probably want "man gpasswd", the files are /etc/group and /etc/gshadow I remember that there is some reason why no one uses passwords on groups but since I never use them either, I can't remember the reason. > Ideally this would all be done in a "script" so the user > clicks on the "matlab" icon on their icewm desktop (served via ltsp or > vnc) and then they key in a passwd or two and volia matlab runs but they > retain the ability to open/save work in their home directory. Yes something like that should be possible. You could always teach them to use ftp. > Does anyone know of an open source matlab alternate? Depending on your needs, consider R... http://www.r-project.org/ It contains most of the BLASS and LINPACK algorithms and it is good at manipulating large data blocks in memory. Also R has some nice graphics and plotting capabilities. R also has rudimentary symbolic algebra capabilities, it supports expressions and derivatives and substitution into expressions. Since it is designed for statistics, it also has a large statistical analysis library and random number generators. That doesn't mean that it is only good for stats, any time you are manipulating data sets and matricies you need pretty much the same tools. On the other hand, R is not even close to matlab compatible. Personally I think that R has a much nicer high level language interface than what matlab does but if you have a lot of existing code that you need compatibility with then this won't help you. Octave is the best bet if you want close matlab compatibility (but the octave graphics were not much chop last I looked). Both octave and R will run on MS-Windows as well as Linux, I'm pretty sure they both run on Mac OsX too so it should be possible for everyone to work with their code at home before taking it to a big machine for running the more intensive jobs. Mind you, dual xeon isn't big by todays standards anymore, some people probably have bigger machines at home anyhow. This reminds me, the matrix alrogithms can be implemented in single threaded mode or multi-threaded (to take advantage of multiple CPUs). Some large matrix problems will bottleneck at the memory not the CPU so there is no advantage in multiple CPUs (other than a bit more cache), in other cases the multi-CPU is an advantage. Look for a thing called "atlas" which is here: http://math-atlas.sourceforge.net/ As far as I know neither R nor Octave use atlas natively but you can patch it in and other people have done it (in order to squeeze a bit more performance). For atlas to work properly you have to compile it (and run the self-tuning) on the same hardware as you run it. It tweaks internal buffer chunk sizes to match your system. > The Linux server will provide, *nix logins, Imap e-mail, possibly some > protected web pages with Apache, and samba shares of their home > directory to their desktop PC. I've been considering using LDAP but on > past experience found it to be quite a lot to chew, there will only be > about 40 users.Any opinions on whether LDAP is worth the effort. I > would need users to be able to set their passwords and have that ripple > through to each service. Currently PC's are just in a work group, trying > to decide whether to use Samba as a domain controller/active directory? > (I am not a windows domain/AD expert nor do I much care to become one). If you set the passwords through Samba then it can automatically update the unix login passwords. It is also possible to make both Samba and the unix login authenticate off an existing MS-Windows domain controller if you already have user accounts set up for MS-Windows users (the Samba box has to join the domain, their are instructions on how to do it). - Tel ( http://bespoke.homelinux.net/ ) -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQq5LccfOVl0KFTApAQLiqg//TVeek+TO3fxws8jylwF97r6QPz/MXrvN VV2fF7A+U6s9HQFaAs/TBMOnNKpOdE04i8qZUFhdpbfbIpQQVmLhd2L5Hh226O6V quhW5+bGrjdTFJNg8IlEXbJkv6VfnC7yf1R/CEhu+Gw04y5Q60YYT+7ngdxatlTm S+KFVY4Jerpmqsv4Er3ybX0fqbQlFboQHmISogFN+UnheLOxAgfgP99P8DOclUhp DDHq5LUkEXwj8sAJBwpPCSjqVT2Hb2+sVD2Dy/WwFnt3DrYyfet8zXUfvF0ZKuh8 jvUc1uMBdIqs+ywLGQoHQUsyGMWUsV5nM6LNHKL3+siWNPx8pFGNgzmeeEMZRRpw MiUuxfA9oWmdkBaz366KlOGYH0hM8p9liv0uhMiYQhwvQe+gjE6Sz2geza95AwkN 5x6o6ei6PKTaOw23iCtNFlSeBQgF3ZvQQqcXPzJqT/qW/MhPXlO2jik9CuHVYK7c PWLJnzzdSHlffywht9FIp4pwF1Jbd1HqDpUbBpKIgjd7x0YlOaWsghTTuztHR7N9 ntPaazTn+GGlZYO9HXXlHr+YZmuESiHUarsfmBocWDLjiakLP0i7YH/Ykg98ldHK ngNn9G9Pnvd4xrInozz1Pc9/uNn/CEI0Cpr4krZtDgekIewVNJq2uwWowq8iEFCg 8WjE6COqM1s= =xhtY -END PGP SIGNATURE- -- SLUG -
Re: [SLUG] Seeking User & Group Management Advice
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Tue, Jun 14, 2005 at 01:26:00PM +1000, Peter Rundle wrote: > Ok so far I've discovered the "gpasswd" command which allows me to set a > password for the group and apparently apoint an administrator of the > group though my idea is to have the user su to the matlab user then > newgrp back to their own group (each user being in a group of one). So > perhaps I can appoint the user as the administrator of their own group > of one at account creation time. > > So I login as peter, su matlab (without the '-' which means I stay in my > directory though my uid/gid is now matlab,matlab) I then $ newgrp peter, > which prompts me for the group password and I'm now still in my home > directory > with uid=matlab, gid=peter. I now run matlab able to read/write files in > my own account . When matlab finishes I need to undo the newgrp and su > commands I just need it all wrapped up in a script, hmmm. I just tested this on Fedora-3 and noticed that newgrp won't let you change group even when you do have the correct password. I found that RedHat have listed a bug that newgrp is broken here: https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=85280 The nasty thing is that newgrp was never updated to support PAM and worse than that, PAM doesn't even accept that group passwords exist. As a result, RedHat have classified this as "CLOSED WONTFIX" so newgrp is basically broken forever on RedHat. Someone made a comment in the bug report that it works on Suse Linux. It probably works in RedHat if you do NOT use shadow passwords. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQq5T4sfOVl0KFTApAQLWdRAAjU2o46ltm+9vGmKOOH9ByMFH63TxFlAt gNXMqYgCt78fP0kFmV5nN/cKWlk5ifWVrgMIDwgCa1o8BKA2OmIbrQ6oZNnEqg9+ DRjF5HrGbXr6OKJwQt4bTp2puxuvpACeD2PpJMObNqnlcaT4Ml71joAQxw4b811x DWXdQTHE+ghS1Y3RO1ELM+mfgJBlRHICXHEmKH40IF+bgyogeqrrA/FLt4G9KGev dH0hVhi0fUVQHQsPPQ5hqSyJA/YerSgdovqIlzEFpcboqPQRd7wW6j7HO7I+I4OY lUzcsJuJx1lTZoxgv7auEC7DBHWVJUXeZhPt/u79Xun5YVAhL0YTqwNvO4tcUgCP 4q62Q3dTZ7ODW/x7ZMzTN0VPV21IPSMRUk4AMs8tJhu5sp7i48ArHuZoznxNlqQC Dwhlby0EQisvEp/oGSLIoiP3BTeG8EME5chUviUBSKGy31tlvdJU+CpZQ1QGpBU8 CWtGUCHpmqRrsj1QZnFmS/ki6gcMxtOkYuz5Ud1dCH5f4fHP9MVXp/NWvABM2Wre uB1iTGH0sYSBIe+Zad+05dV9LWX1X9Od8o1tQOeSTr/Ggf1RkbOJk4ST9wMIQOYu vJIi/9dSQYF6LbJ+GXl0hYQr2/Lb/mC+by9m+agcvQb4ujuXUaTxuNYvVU+zH3dY 6rfpsT08Gt4= =nTx4 -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Call for volunteers for June Slug distribution roundtable
On Wed, Jun 15, 2005 at 08:36:43PM +1000, Lindsay Holmwood wrote: > At the June Slug meeting we're going to have a distribution roundtable! > > We need a representative for distributions to give a breif summary of > how their distribution handles package management. Once we're done with > the summaries the audience can field questions to the representatives. > > Right now we have a speakers for Gentoo, Arch Linux, and Rubyx. > > A distribution/package management system can't be represented twice, so > you'll need to get in early if you want to talk about yours! I'll talk about CentOS, it should be a short talk if it comes after the RedHat talk. But who is giving the RedHat talk? - Tel -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
[SLUG] Distributions and Package Managers
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Thu, Jun 16, 2005 at 09:37:58PM +1000, Ashley wrote: > > I'd offer to talk about YAST but it is so simple and user friendly that > it hardly needs more than "Here it is, the most user friendly system of > them all!" ;-p I think we want each of the talks to be shortish anyhow but there are a bunch of things about package managers that might be of interest: * How does it handle dependencies & conflicts? - --- file level conflicts - --- file level dependencies - --- package level dependencies - --- versioning for dependencies * Support for alternative packages that do the same job - --- e.g. postfix and sendmail - --- how to switch from one to the other without forcing dependencies * Support for back-compatible code - --- I have this program xxx that NEEDS some old version of a library but I want the rest of the system running on the latest libraries. - --- Can you install two versions of the one package? - --- Are there "compat" packages available to help you? - --- The various "libtool" and "automake" scripts are shockingly version specific, at any time you need at least 5 different versions of these in order to have the special one that each build needs. * Querying the system state? - --- checking where a file comes from - --- figuring out how to get a library file that you don't have - --- listing the files that a NOT owned by a package - --- listing the files that are owned by a package but have been modified * How does it handle configuration? - --- ask questions when installing/upgrading? - --- overwrite of configuration files on upgrade? - --- attempt to merge configuration files on upgrade? One excellent example is the X11 system when comparing debian to RedHat. On RedHat you install all the x11 packages but that won't give you a working X11 system... you can then either put in the config files by hand or run: system-config-display and then it goes through setting up the display. On debian, packages get installed and then configured by the package manager and it asks you questions right after you install. If you want to go through those questions again you need to use: dpkg-reconfigure xorg-xserver (Or something similar, there is also apt-reconfigure which I found didn't work on ubuntu, there's probably other stuff I don't know about). * Methods of upgrading the system? - --- Put CDROM in and choose "upgrade" (chunkular) - --- Track latest packages from website or ftp mirror (trickle) * How to roll your own packages? - --- source code archives - --- patch to source - --- versioning - --- distribution * How easy is it to get someone else's source code package and make some modifications of your own then use that package? - --- making a patch - --- testing the partly modified article - --- getting around an annoying dependency (e.g. I don't want PDF docs and I don't want to install all these crap tools just to get the thing to build) * How to include non-packaged code in your system but still have a working package manager at the end of the day? My experience with apt-get is that as soon as you break some dependency, it will never work properly for you again. It doesn't matter if you tell it, "don't worry, I have this under control", it always thinks it knows best. It doesn't even matter if you want to use apt-get to install something unrelated to the broken dependency... it still goes nuts. This might have been fixed in recent years, it was the reason I finally got the shits with debian (that and lack of CDROM distributions). Anyhow, the ability to include a hybrid of packaged and non-packaged code in your system it pretty much essential. Another issue to look at is auto-update facilities (e.g. RedHat network) and how much control the sysadmin gets over the update (e.g. up2date vs yum vs apt) * Can you easily check what is out of date? * Can you hold packages back? * Will the held-back packages also hold back other things based on dependency linkage? * Can you source from several mirrors and if you do, how does it know which version is the "latest"? To my mind, NONE of the systems have it really under control. They are all so "user friendly" until you want to actually do something with the thing and then they drive you nuts. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQrI288fOVl0KFTApAQJoCQ/+NxBTs0/OQ3aSq/XTJgfSStbwzIO3r4kG oTRa1H2TkE15r6U8ER5TAug8fNqCVtFMyvxaWpTR8FTo0svAmJhoaQ+5rDtuAC6F cf5mATRBPAZVn33CWB/M/tFQ4bfnVFh2rC9Cz0vOa2hRkb1XZ94/aNzclQFcZhiI PmLS8OHpnkFBjOXEf2qHl8ettWCB/cWZQZ1KPgVpdQeX1klq/fGrUHYM57E7Qqvb 6/NlwvAT+TJAmdDJ3CA+2wHlP0nI/rak7Sl8erv9ugQE4ixo9Gxiu4z7C4yfGfxE Gmsf8BZCHmNqBHCzWgx5gxHhPYvn6pUZAWx7OEBSmTUr6QOkVGUQjbg63kb7XXQu YOiCd2fuCFHCULjknLGnVnGNHeHKAVXUS22+mVYHRmZKT0ll0RX68TkG5Z97xZeQ 5OoM9Pm9TcYjbzPeMNDTPZORY9DlQKdEaiJgaB8xCp14PewXytBMgrq1/wAEWgKR 4rL/z5OnsCF4X7oZkZbZxEV+BUaYnGkcWlPJ1c/QIDGy2SZZtqLFuW7KDelDPPI/ BWFsrpKL3a
Re: [SLUG] stolen laptop
On Fri, Jun 17, 2005 at 12:23:51PM +1000, DaZZa wrote: > It also allows you to delete files remotely - not sure about anything > else. It's useful for tracking stolen laptops etc if whoever steals it is > stupid enough to connect to the internet. If he doesn't reinstall first... I remember talking to a guy who was selling this "phone home" software and he was saying that it will stay on the system (and stay active) even after a format and reinstall. He also said that it would keep working even if someone installed Linux over the top. Needless to say I didn't believe him but I still wonder if such a thing is possible? My approach is just to get a Dremmel and carve a phone number into the case. Makes it difficult to sell down at the pub and sooner or later someone will ring the number (maybe not the new owner, maybe one of their friends) then you have something to give the police to work with (since every phone call is logged). - Tel -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] stolen laptop
On Fri, Jun 17, 2005 at 01:09:04PM +1000, DaZZa wrote: > Apparently, this product does just that. Those are the claims, anyway. The > only way to stop it working is to physically replace the hard drive which > is in the machine. Pretty strange when you consider that they say they don't support Linux but then they also say that if someone installs Linux over the top of it then it will keep working. Somehow I have trouble reconciling those two chunks of information. I guess that bluff is half the ballgame in the security arena, certainly seems that the airports are working on that theory. - Tel -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] stolen laptop
On Fri, Jun 17, 2005 at 01:27:11PM +1000, Michael Lake wrote: > Yes. There are readable and writable areas on the physical platter that > are not normally accessable by the OS or BIOS. They are usually reserved > by the hard disk manufacturer for their use. By working with the HD > manufacturer and the laptop builder you could get a driver written which > will be able to access those regions. Supposing that you do get access to those regions, that gives some space for the program... there's still the boot process to consider and where is this code going to reside in memory? If I'm running some sort of NAT system (which everyone does), how is it going to know my local network settings? If I use a pppoe tunnel (typical ADSL user) then it is even worse. I guess it could try for DHCP but that requires the resident code to take over from the OS for long enough to get a reply on the DHCP (i.e. a long time, long enough for the OS to attempt a context switch or a device driver interrupt). What about network card buffer space and packet queues? Frankly, I don't see any way it could work independently of the operating system except for if it can grab the bootstrap process and get a quick packet out the door just as the sustem boots up, and then only if DHCP is working, and if there is no local proxy. Contrariwise, if it does work with the operating system then it must be hooked into the OS files somehow (and thus living in regular IDE disk space where the OS can find those files). - Tel -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Re: Distributions and Package Managers
On Fri, Jun 17, 2005 at 12:02:44PM +0800, [EMAIL PROTECTED] wrote: > > Another issue is how easy is it to set up multiple sources, so that if > > one is broken or incomplete another is used, and so on. Easy with > > apt, hard (or at least I couldn;t work out how to do it) with yast. > > So you are going to install some random binaries, packaged by some > random person/s onto your machine because it's often easy and there > are not lots of war stories good luck!! Where I choose to get code from is my decision, not the package manager's decision. I'm the guy who has hold of the on/off switch. I'm the guy who is going to reinstall if the distro doesn't do what I want (and easily). Lots of package managers seem to forget this. > I built a system, used apt-get to install mythtv and 50M later had it > working. > I built a system, downloaded lame-src mythtv-src and the total footprint > was 10M by the time it worked. Ummm 40M of 'something'! apt tends to drag in anything that even smells related ("tv" rhymes with "hippie" so you are going to need some hair-care programs with that). I've noticed that most debian package authors tend to set dependencies to be "everything that I have on my box right now" rather than thinking carefully about what they really need. However, we were commenting earlier about human nature and why it never really changes... if the system makes it easy to just drag in every possible extra package then thats what everyone is going to do because people always do what is easy. I think that the talk on Friday night should have a slot available for "war stories" because that's where some of the real grist starts to come out. - Tel -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Motherboard for software raid
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Mon, Jun 20, 2005 at 07:41:27PM +1000, Jamie Honan wrote: > My thinking is that I should use one IDE boot/root non raid > disk, and three SATA drives for the raid. Is there a good reason for that? I would be thinking to make a small boot partition (100M or less) and use the initrd to establish the drivers and get the RAID running. Then put a copy of the boot partition onto every drive so that the partition tables are the same and so you will always be able to boot (and put grub on every drive too). Then you can have your entire system running on RAID and use all 4 drives in a symmetric manner. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQriKM8fOVl0KFTApAQLhTA//cb8MV219OFK3j4AMI6MnYoelX2KKkmC5 YdMLfgriRRUrrOJmCaCtI+dcIDOCX2OsyD/oxX+w0UzvYUzKX4N1iRIEdslUZxsv NDI6ArDGwqdV/gtSPUHm1DN9HyfbaIG+9Cphj1TjnpTWWW4YKq8BkPEACRdAtCat hAuCRsm9svFS3tgrxbITncr4wmEnFsNufnh1+gd0CtrC8Wd/9wsd372tu3jl2c0w H0LAH/tYaDjkWi9BxYaSPmmB9+Q+wls2B+uzZxsxclyqasNKFj05SaqmRtew5Nwi PuDeWLYA8MQPeD7t9gcbWPF9lrS2UOvijJVh2tjdt1OjF4nH1Yp/QxxItHMLs/6u TAkB4ya24yWHz64x1YVmrvgJ7YWDf1GgGg8onlCoGas0GHab/ayxA+YbkWXtIcmn J00X2t5sFyp5JqYcvhDCiR+x06cICGPn7eGXXyulyI0I2PcX0RNvslnpDk+J/4Iq 3S7m87Ebf53hdoEaQVpHwQjlV2JY4MghxJj70mGq+Zr49KceyPPLVb8WZFG0YNeg yrMaY5YuP4+mjOOHSzIMJKGUfNPu9T4YS+vh6JVOrwqQuKWKlq5Kx15G4JNiq5g6 UoMYZDcT5Tb9NPDoUbumqX9H18DSrq3FrzBsdjIKsY61Fs0mMr6GPZZ/pJHC0KTg 9KoefOrohPc= =b2AP -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
[SLUG] Call for Speakers: "Email Server Roundup"
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 In the trend of our square/rectangular/round table discussions, I'm proposing something like this: We all know that "there is more than one way to get the job done" and when it comes to email that most certainly applies. Describe your favourate clan of email server apps and why you like them. In particular: The "transport" layer, SMTP, local delivery, etc. The "distribution" layer, POP3, IMAP, etc Webmail... is it worth having? Describe your best webmail server, what makes it good? does it play nice? Is it hard to install? User authentication and access control: is it really worth messing around with databases? Filtering: especially spam and virus removal... what works? Administration tools: how easy is it to create and remove users? can users admin their own mailboxes? Probably elements of this could become talks in themselves so to start with we will try to keep it quick and just get each speaker to cover one particular setup (e.g postfix/cyrus/twig with postgreSQL) and describe why they use it and what they like about it. Please send statements of interest to [EMAIL PROTECTED] and if you feel energetic then make notes on the Wiki page here: http://wiki.slug.org.au/wiki/SpeakerWannabe Don't know if this will be next month or some month after, depends on how much enthusiasm there is and whether the other talks (like the education stuff and the perl stuff) are more interesting. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQr5u3MfOVl0KFTApAQIr8Q/9Gwvb+hMUhTHOZOyuncLbJQlZ5osRzo1z 0HbKY3mRx3w5gRk0Mh6A2gcJpPM32nyh+Xk2f7BK7rBDngyQKXilAOoEJaDO8TFA rkv2uwD+WQphedO0BPbo23si7A+8CTQLPUwgOMqh9jQi+DbSszkuVc4gRxBXZxDO 1RxAgpgnauBW2v139SCcg2RAz0Qs01cjcq8h4LO+FW0K6/xnG2agy0lGOAjPM9SC fbl3TJWggNMqsAxenBn8RoeUiIOvKB7NBOuQcaEdcgtCbLmJKQQ2bwJ5tSVrGuCp zMV8FqrEVAUeJo659+I52GUCrAD9bEROkjjssMfvysYQwMC4JoWfmCr5PA3YXNqD 9xmx+7Cso3xYHRYW6OfQB5ya+QRLCAnsqIaoAfh7DTg5zIXwdJFGpEKKB/dTEMSr rtsrhchCMK0Gtiin0lDv1loWwA9t6fZBrkynjS5YP7eruYZmO+F/Ak6MzWVz+RBm /tYGQ4UWHnegmALxzfMoEw3zxf2byciFvjYLEykJvzg/pwOau2aUmEQw9uyE5fZR B5O1scjflWyio1OnMKAqs/4drCSOKIjQg2Yh9+MFWedDxHbxLIQ9ogu+i6mjN1Kb oe40Cx5wCEY8vwaRltQmuH5te/UfOQHSel/2gYiybF0KRKx6T4jbY3cnnPEnrw4B yjJh8GSaXnA= =oiUO -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
[SLUG] Linux storage systems
Couldn't help noticing the "JeffWaugh - Storage Dream Team" topic added to the wiki here: http://wiki.slug.org.au/wiki/SpeakerWannabe I added a list of everything I could think of just to make for a bit of a challenge, explaining that lot should be a fun exercise for any speaker :-) - Tel PS: Would be good to see more people whacking their ideas down on the wiki to build the list of cool talks. -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] htdig or other searchers
On Sun, Jul 03, 2005 at 04:58:10PM +1000, David wrote: > Does htdig go to the server (thereby parsing includes?). The site > navigates perfectly from a browser. Yes, in my experience it does, there are entries in the access_log. Beware if you are running virtual servers because usually it connects to 127.0.0.1 (but that's configurable). Also it has filters based on file type, usually it avoids all CGI references. - Tel -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] E Commerce
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Wed, Jul 13, 2005 at 12:12:02PM +1000, [EMAIL PROTECTED] wrote: > My wife has asked me set up an online shop for her small craft supplies > store. I'm a bit confused about why people are so hung up on internet payment options. We already have a vast number of payment options which are well understood and fully functional. Every shop must already be able to process a reasonable number of these in order for the shop to work at all so why should the internet be considered magical? * Come into the shop and pay cash * Write a cheque (either post or over the counter) * Postal money order * Come into the shop and pay by credit card * Give credit card details over the phone * Direct bank deposit at any branch of your bank * Direct bank transfer (at branch / phone bank / internet bank) * Run an account with the shop and make monthly payments as per above If you accept some or all of the above already then just make a list of the ways that people can pay and put the list on your website, also put contact details so people can discuss options with you. Thus solveth the internet payment gateway problem. What remains are three things: * Using a web page as advertising for your shop * Using a web page as a public catalog of your products with prices * Put an ORDER ENTRY system to onto a web page so the customers can type in their own orders (and check for correctness). None of these have anything to do with payments. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQtbhysfOVl0KFTApAQLqURAAk1um9wDtoMziOSroUhGcmwy2uRKBRU3Q JbzOtZCHTortGAFXR2OoJ3/WkyoL6gyLnpU9g/r0IVqmmYKxzh6He+D7T7Ipf5TY EDCRS+9Jd9Dg5LVVstkWcoo4yxnKmoUBZjgqHIao3VT6LCNv6BhX+yogZIMSfqAO +KIFmsuXq7dq6m2gI93eh74gd3n9OHpqH589crWZ+uAtI1CArMWTPnf6ND6UTBnD p6MdnF18duL41oHtKyE3M220c2X2zG4CABVcWRNk5qCfAdTU+p/ob9355FqoiKtU vscbtrGp3A+AMM1nKkYyhO9f4P6SXoa16tqJKPtAY7eTqI/cWgXPHoQf9BlmqQOf nVuOI7hx3ccgAev0NOrJ1CQDY2W0MMPt7vQuJerwO+M1TwfdDlzdou+MshjzTCDr lKoEadlGSCFtIwxGrEpCNhRXTXr9knWuAJn7d3H6jr88nUqKq6etwkw6PKQpgzod tsooJT1rcUN/0eiVEZpphhisZTOgiRkCjIhFg75/SFGIhN7A3LkNSVxwmB/4mqqt v/ScMD4RSLIWCWlhVKdB55jJgsjMxz97Wkd1WMBkcri1tzhoTitS1KKvV8XMWtEG fLxs6OfpBpK5Scl1to2mH46+9S0b+BBRozkcFoG1AyOpaEZCf1DB8NbmCFKRmDtw Hpb0m7sj2PM= =dCmU -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Pros & Cons of Unix databases
On Tue, Jul 19, 2005 at 12:19:09PM +1000, O Plameras wrote: > ldbm - uses neutral storage interface which could wrap Berkeley DB > (www.sleepycat.com) > or GNU DBM (www.gnu.org). Only sleepycat is considered > reliable, though. I've found gdbm very reliable but it has a few limitations. The most severe is that you can only have one process holding the datase open at a time, that really limits what you can do with it. Actually, I think you can have multiple processes reading the database but only one can write or something like that. Also, I don't think gdbm has any sorting facilities, if it has then I haven't used that part of it. There's also "tdb" which also does not support sorting but does at least allow multiple processes to read and write the database simultaneously. Samba uses it and sometimes pppd does too. - Tel -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
[SLUG] Security Video Cameras and Linux
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 I have yet to find a good website talking about security cameras and Linux (the hardware side that is). There are lots of software packages and various odds and ends that work with v4l such as frame grabbers, etc. I recently bought a Logitech QuickCam Pro 4000 (USB) which is a nice enough camera that is crippled under Linux because Phillips won't release their proprietary compression algorithm. It will still give 640x480 resolution but NOT the full 1280x960 that it says on the box. It seems hard to believe that they really have a cutting edge compression algorithm that is genuinely valuable, more likely it is just a matter of corporate disregard of Open Source and/or they have infringed someone's software patent somewhere and don't want the infringement to become public information. The other possibility is that the camera does NOT actually do any more than 640x480 but the additional resolution is done by software interpolation (like many scanners advertise a high "BS" resolution then a lower "optical" resolution, aka the real resolution). I do notice that the Logitech website says that the full resolution is only available for still pictures and on some other cameras it does mention software interpolation. Needless to say, it would be good to avoid buying the Logitech unless you want a crippled camera, but now the question comes up about what DO you buy? In the shops, there are no other USB cameras going anywhere near 1280x960 resolution in full colour. I can't understand why because USB2 has plenty of bandwidth and is hardly rocket science these days but yet most of the cameras are around the 320x240 mark. Another option is to go for a PAL capture card and a brooktree chip. That works OK with a regular video security camera but the resolution of regular video is still not much different from 640x480 so you really aren't much better off. Most of the PAL cameras are from 450 up to 500 lines resolution and I would presume that the horizontal sampling must be limited to something of the same order of magnitude (even though analog cameras don't really have a horizontal resolution, there is a limit to how much real detail is in the scan). I've also looked at a few IP cameras and they cost a heap extra for very little return (well, you can have longer cables than with USB so that is one useful thing). They also have low resolution. For example, a D-Link wireless camera selling for $500 is here: http://www.dlink.com.au/products/multimedia/dcs2100+/ Which has s many features -- except for a decent picture, when you finally scroll down to the bottom of the specifications it has only 320x240 resolution (which is squint quality). So is the Logitech 4000 running in crippled mode considered the best that you can get for a sane price? Is the crippled mode really the best it can do in honest optical resolution anyhow? I bought the logitech for $135 and probably would be willing to go to $200 if I was getting something noticably better. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQt4ca8fOVl0KFTApAQIC9g/8CrnacOq//xT07hriIkEBZwPJtBgVpgO6 hdZWMxmaa1myDFjDSuu5dB+j3PIPR9uXfJQjD+Ih7p4enQwvlhZQYR67waQrEWnc xPDgv3nAkrjp5JgnsYeXiM8E2IOuF02YDZn2L4EyKo4ItCtYBoFiY8eO2Y5HzNpZ F41CT/IFqGdHnXY8pnp02jcvsAWHxjgAW8+X8BzpZ+R2kqUHJPIU7zkh1/us4l1A m0pAM3Kx+a8TudkVHstI99tGYWMFxqscvYENtXyvKdQqhTbbKoc5iOyNtjwqN7Z+ BuynH8w3RCLxmsCmE7kAGFRWtBtJDyV76yf9chEYVMOOAXiIwFPG9dslU5nAqg3o GFhHj2qiRb83ir8l2jYYuPQCzdWprm7wEarJ1gc9Pn0d3wfZu9+i48B/CY9Co8Cm w7V74rCGPRqF3krOf4Nwi5+WB64qxq6ppJ5HweJ869x3IwbusdUyFaR/sV80uRJX xCXvyxem6/bqRxKTfJhD8Xy0sdiOURerrCz1zU9utHmyJY7ftpI/HJycW1CN5kGP WeFhTJ65Bgx/pTDY6q5Tk2VX/WPUO5uf8wOa2cT+uoka/rgLu3Soiaqnqf0dY0fp CYhhNaXXVYbDIxrd9mNsy6Xxe8gbxuQkmqYdMOPbNpwyGeMdmxEwLlGXsyxqLYPJ VHsvq2fWOFI= =pa7T -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] USB device mappings
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Fri, Jul 29, 2005 at 12:38:17PM +, Sam Couter wrote: > Ben Buxton <[EMAIL PROTECTED]> wrote: > > How can one go about ensuring that certain USB mass-storage devices > > always get linked to the right /dev node, irregardless of the order that > > they're plugged into? > > Irregardless? Yay for Bushisms! > > > I'd like to have things where by for example, my camera always appears > > ad /dev/sda, my USB key as /dev/sdb, etc. It seems that they are > > allocated in the order they're attached, but this makes things rather > > complicated as I have to keep hunting for which device something's > > attacehd to. > > Install udev. I thought the whole idea of the /media directory was to make it look at the name of the filesystem (as stored in the superblock, or in the header of the ISO image, or whatever you put in there) and automatically mount it under that name. Thus, the actual /dev/sda or /dev/sdb should be irrelevant because your data always turns up under a directory named by the media. Think of the situation if you have two flash readers on the one machine and you pull the flash card out of one and put it into the other, the result should be that the data is still available in the same place. The Amiga had this working with floppy discs around 1985. The Amiga also supported partial unmounts and other cool stuff (where you can actually unplug the media while it is still mounted and it will safely track the media state in memory, provided you don't go and modify that media in another machine). - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQuwnHcfOVl0KFTApAQKxkA//ZTQkZTBP55CrTwm9nW9+/J4IBjMVdUa7 ct07nh3R64lO81yh2kVxbn2Gq5LIkHCMeEQXo+w66R2duqHo4+BjKw8iUjRT5p5n ZtU++m7TaWph3vgfO0790AqK0mmXdn6qCebH96a8y4e+KTM6jbp6HAMoBiD/Oxsa IyVh58CrcwxgbfnfmSaXPG82iZWilcKDI5YjnFujnZ9GSRn6Ot5xiJn9DYzDdtxq qz4HhrAKp8PZf5aJZq7HXiSGmNXDo8Jp4KsTlIgmmnNn/0uIOr4/Xk++Wz4f9or/ Xmu21OIryPi3kiQR/KGIcuYwmTWU0Mny5+TNcuXrggiNDRNiv0akx+e/XAtsQE3n f1q18sx5OnmxdpBpDRwsQnYdrp9enS4z/d29FgbvBdrKCXTdMLw0sxtBSh5xbhJQ rP48NoJp7t7p2CqK1Tu4pwDZ2cIRsW5H8pNfjz5xQLvKLyOzhXIepYslF/e8hdgT RwlOlvD+Aewte2ed8aBewwxRuudPZ7/ZBItxoQ4ArRpUeVaJ0DkWNJgrO0CUk17C PFWi+YU18WHOuzbhjT1znALkD8QlmdhXpTBx7T8ov2iH07qoEUmSFgXekHPXCj4Y GsCFS7sCDVGUNqQf4Lwi1ASRfKkgy+A/vE0dPW+yg3gbvBt6BmZlZ1Lz3GOzgYXu Be3JtuQ8Veo= =8rZn -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Podcasting
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Wed, Aug 03, 2005 at 01:26:07PM +1000, purserj wrote: > >Seems pretty simple, which is good, but considering the size of the > >book, I thought there was more in it. > > Nope that's pretty much it in a nutshell. Podcasting is a simple idea whos > time has come really. I'd say that podcasting is a simple idea to do (in a slightly different way) what another simple idea was already doing. After all, it works just as easily on a normal web page with links to MP3 files, and hey there are heaps of those around already and have been for as long as there have been MP3 files. Right click on the MP3, select "download to ..." and you have your podcast. The main "special" thing is the RSS newsfeed which has some ability to merge and filter streams (which you can't do with normal web pages). This (by the way) is also an idea which duplicates an older idea in as much as usenet news feeds already supported everything that RSS supports but usenet got ground down by spam and cross-posting so people decided that what they needed was a different technology to do the same thing. Note that RSS is not actually any better protected from spam and other such things, except that (so far) no one has mered enough RSS feeds to actually cause a problem (I give it two years). - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQvE/LMfOVl0KFTApAQJdHw//fWvQZWWTtvf9yoHeLE4SkkV43O6fBwLL DO9NBfSq2zy6lbCHW/u3ZxtGQGrfFOb4G2zolydJcorxqt+9G3p9x3WynnOO10fx oYChfoOqyunf+BMGBIu2hkGbsmhvRPj9vlMH0eNWPYRQRjEfbNtS/F3DrFnhLvus hc1MdHR/SOFjn3f/7aPaHeu3f1RSZMVSQnOW42K+0ArddlQcPbbEQ0KEIy4Yu9G6 X7fU+PULjYQAJtJp2QI2b4K5ClYW2HYncNZVM+tAOpy9Uh/PhmzWIyj8+2uH9Ox8 kjlxOsdLeB5V4qMRxfv+/0poMNKi9DHmD05+oxwAZhGwhuFgQpq4pt90ZiFH2ix0 XBYfztWiqh2rtFXYdDCTswTedkLFaNg3yjzfChpXXmYlsrasnjuzvlysr2+xIAhC c+80FK1D7Hd5U5RtKw1rvqhlBxOjLJTkdBHl3HRj5l++NN5NfVu1v6EoXKKJA3rF 74WuVNv9NLP8JT/hkMWyy8LLqCXNRA1YgIpKCC+lgtVFLcWMV2ecdM3/wePN+I7i 6cwa51kqpw72J+2cQl5Qild9BUN9iaz+tgKrohdHQS3XgAfMGDfqL1JSa1mhg8xc mb3GZMJPkWngcA6NwoROIaO+MVfwdI04NdJao8WBANZ0ybU26bfw4TjBKIXi5dvg VuqDbPY5nic= =W40r -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] dual authentification
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Thu, Aug 04, 2005 at 09:22:24PM +1000, Ken Foskey wrote: > Got this one today, strange one: > > "Do you know if it is possible to setup a Linux redhat server to require > two passwords to gain root access? The responsibilities for the server > are going to be split over two different teams and we don't want either > to have root access without the other team knowing about it. Please let > me know if you can come up with something." If all that matters is the other team "knowing about it" then I would suggest that you send the syslog messages out to two different machines on the network and let each team monitor their copy of the logs. Then they know that if someone logged in as root and it wasn't one of their guys then they should be checking up with the other guys. On top of that, you add a polite procedure which says that each team promises to notify the other team by email before they do a root login so that if either team sees a login that they don't have notification for then they know the politeness has been broken. This forces both groups to prove that they can do simple organisational tasks like keeping track of notifications and checking them off against observed reality. Of course, if they ever get hacked, the attacker is unlikely to use a regular login thus won't leave a regular log entry but they might leave something in the logs somewhere... and with two teams who are actually forced (by their day to day procedure) to read the dang logs, there is just a chance it will get noticed. Also, sending the logs out across the network makes it harder for the intruder to edit history. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQvKc4MfOVl0KFTApAQJNwA//fbnKJ9yq6Ml3aS6mQ+zRDvtVkC/JYVD3 FM0kis0nhK3XQaEkfBqhoZaeQOKzwqOeFz2hVxDDUlLFsLqIwpUK5x6OVq1tqeta aNwoYsUYP2CTrS/gr95mSij+tXnGF78xI2Fu37w7bH40gora1n3/PQdNBycWhjWv iJ8NmKeu9oYYEVnydIvUjthms7TXocZPlZGgWMhQs7ZCGgTEgyAv5NpAl2thTFR/ hP7nZclMoVvuhWgeKV0FkVDEREQC9iVz/4Mhk2tKVE3/8s9TAjTKQOHk9xqxGaQ4 6HLQtvZEgLJgArAE+O875aEkoRinuE1yONrjfDUrC8rc9pfJK8WB1rP70x6emxVR hQjhICPe+DB1NeVChMmOA/bJVFzrnYLGLbKm31mD2au95auyt0+TKqW4LYz//e6t JiJ1K+ALdzTR5LChlp85FLasbrzhptTwq4t+7nyBBlECPGcsUD9tKMOHldXGVnc6 GuGhoe06I3Flo9qbSxwvzSeWV0ixTlUj3oS8l0FV9mN/2rONEBl5ZXmGw2N52ggP awsTH8bb6CNDwfsevMUHyIGfeozyUmBljm+dGlQX7GCx3WA2Zue6G/5Ffh+eRWoL P1pGQmbGgtuA9JnM9KE10GsmfiEW3wtRPZMEdYHFfSLstKfeJJUMtvK5RP5ZgRD8 8OoiRgQySxo= =mMc+ -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] [OT]: Database Design Question
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Tue, Aug 09, 2005 at 07:35:11PM +1000, Adam W wrote: > Hi, > > Can anyone put a name to the following type of design... > > I need to be able to modify the amount/names of data "fields" stored > for an entity so instead of representing the data stored for an entity > in fields associated with that entity you would relate it to an > another entity which would contain a datafield type field and its > associated value, so different rows in the parent table could possibly > have different datafields associated with it. Postgresql supports a sort of "Object-Oriented" table system where you can have a base table and then make an extended table that adds a few fields to the base table while "inheriting" all of the existing items in the base table. This sounds something similar to what you are looking for. You still have to spend time thinking out a suitable inheritance tree and all that design stuff plus this is a postgresql-specific feature so you sacrifice some portability. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQv28ucfOVl0KFTApAQIqZA/+IHiSGC2aeHeP7DRPkwvE7vIm6UIBmlde lrk0+yV8/SDUQu7g/ADskrjYHBYvJtnV24S8hoMsTQbGtnB9AjFjHmnXCEHHoFwi +xwHvTZ04vXiU8Rc+ozk/Pg0niYgeXIF0FBpWrjiNgjI2FX5iU7q+z5uLITolls6 QW4AP/a4kxRSNBvS/H4e3+1vR1Oe0UO97iyeWTttub/O1+l054QmDp08cnPS25hb NR9QC0QEi/s8b1SV9ePRgDvsiPZiGoi6DHKk7amcowkSK9EDs9XvB1AECnfWigy/ EEX+q6yu69AZoPsLT6ngo2BzyzDu+sM2aEbj7Ud/hnfd4mKpxhmVsQIgTWSQvpia HuU/5dr9NKs5Z1g0WGNkOgKyt/9u4BthNSDfXelYDVjKmHi6noEOucVLFfPGvRbl gSXBjLow0KyHws9pHVb+72mzMEbs19asHtptxUa431JaPiQpwmVp6haVizWYIKeS TpBSq4ApQiQTtkMyF0i2MwQi7w4erl4tO6Xe3aOTPVbHi8Lv//K4xe82DU42DGRy tK7o2H7SxmbhvGiy/NCwIO1jAF/Xj0mIGTrqRSDdCrcIt9EZ2TQUmBvVLoV0KgDo nplKpNDWhvMrNr20VsLRKNwAJD99yJzRDF+a3wc4+BRDevQNvbNu5ruN4Pq7OllE w0G9wgWb59A= =VSvi -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Lindows experience.
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Tue, Aug 16, 2005 at 08:21:27AM +1000, Paul Trevethan wrote: > > > While I believe that Lindow^H^H^Hspire is a wart on the face of free > > > software, I was shocked to see Ubuntu seemingly taking the same > > > path. Am I missing something? > > > > Yep - the difference between running every process as root and secure > > access to administrative functionality via sudo. :-) It's still kind of risky to have a normal user running with unrestricted sudo rights, not as risky as running everything as root. Malicious software that has taken over the user's account can usually find a way to trick them into entering their password, especially when they are in the habit of entering it at various times anyhow. It's nice to have root as a DIFFERENT password because it provides a warning flag to the user. > Also, is it not true that Ubuntu's action with regard super user rights > only applies to the first user created during install. All subsequent > users created do not display these "sudo" traits and behave as a > normally restricted user on any other Linux (apart from Lindows). > > So, on install create a user called "lord" or such. Then when > installed, create all the other "standard" users you require. Yes, this is a sensible idea, isolate the danger as much as possible. Probably most ubuntu users don't understand they should do this, then again, in a desktop-oriented operating system security is typically going to be a bit more lax than in a server-oriented system. > My view is that Lindows, in its attempt to be so much like Windows to > supposedly make it easier for 'crossover', has in fact become so much > like it to include its security vulnerability. Why not stay with > Windows? Price... freedom... attitude... I think it is an excellent thing to have a Linux distro that has the stated purpose of being as similar to Microsoft as possible. I wouldn't use it myself but I fully encourage anyone else to use it if (and only if) their main criteria for measuring technological progress is comparing things to Microsoft. For example, each and every time someone does a review of Debian or RedHat and comes to the conclusion "It's not like Microsoft", the reply should always be a resounding, "You should be using Linspire, go review that instead". This leaves the rest of the Linux community to go and do things that are not identical to Microsoft. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQwMfCsfOVl0KFTApAQIJiw/9FTfMh0bRImSBV80WinSEs1Lbf5p6/tPA Rts/lueO3towy49IxcABXbdOZzfFPwDadYRgr4sBwCob880G2wdNJau5hb27WJl0 z5P1dS3hPRUjkNPUPnN9l4Wc5JARhP8EpjW9qt5asdyRMK0xN+mGiVIu3I/cJbkm 1g1L3o+rvmQ95Ld9u63yeDJQyegGvB+GsMQdEIFcQEHdSFMOZXfclzGP7AIcl+Wl ViUjBkOj6q7Ga2qTVODnV78bvft0q8bSbpgGjksQ/25KVm7PfHQCiyHtGVfpzQBk +iaG1GsvgqQnaWPmuqY1LTvlXhdkUmr7tjEcGBYjDrL4uvDWYEZUNmKyv1wSiAqP XJ2BMSnXG2q3wFkdBXgWWOh2+Dk5boTddWKKli0O2IT3cumV+BxLjOzHaBrrLfcD HGd4uh9rq0GBIR2YHFKfIk0GlupU/usq2/PCHGPCvynhxfg40/6gE53b9d1/wp8k wSTH9ojWvwZR7vCuVeaYGQaJ0UvSHpob377oJRJWPiq/B1eYXRI6b2jYwRL+Lekq qtzB65Xk4HMB3lIZnd6XXDQeWquW0WaRrypSeptdd2/kdfVZWEozNR1AVqiMPH31 D6J6ZswYzc60l2f1a8M7047wa0VDsl2BMwkE3YkaSCJlqB2CrTqJDvibszy9VkPA CFJgogV0d3Y= =pcJN -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Font Server [Was: Looking for a Linux repair shop]
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Tue, Aug 16, 2005 at 10:50:17AM +1000, Jeff Waugh wrote: > Font servers are irrelevant these days, as modern tookits use client side > font selection and rendering (fontconfig and Xft). Once upon a time, it was > handy to have a font server running on your network so all your X servers > (hardware terminals) could have access to the same fonts. Back when discs were small and all fonts were bitmap and compression was black magic, it seemed a good idea to avoid having many copies of the fonts spread around the network. This was probably about the same time that /usr/share was actually shared. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQwMgycfOVl0KFTApAQLtTxAAnOpZorbzxiZN1Kc0+Y1/Hhg11RD1Tvi3 88cETIngNMeZB6nlYby+azSeLqkDxYwMTvaWaueaby/Spgjxox2eSRvaR5wV79DR yW1A9Q9rxkr6zgh1dHm/0EWBpswFNlKOUX3yoAZnLTsyhtm7Rr4DEYfJvFTpO+CT 3wq7y3eakn5MRgXaOqns75ukpF6+uJDSWIVatSZQvhQb85XDN5Tbb53uwjvqOY/O kxRNP33b35BU9MpMRtpJlIzJPP9BAmR3mZDkmb5MeXlvP9p4PYsim6+620xZeG7j c7I5Tnpi3E0QaczCvebYyQuFUPehRQKBRk5R1VZM0MKNm+HcPeRB+s7UPQifhsT0 RsC+bjvWjqN9F4RZyiGVEJrUF6zgf1DM6GjtnKgQ/eM5Dy6LDliTrCoX60vHYqxe Wf1xcyuMuqh853IckyxV3I2+Z6OZXWd/4ezb0XEtFzLJ/2U6V4v/vpHsEMd6n1d4 RsJp9AnW0sT4/XeG7L9acxNPqe/oiv7M/37tkV6IUL6pJDlUjMGpyq0o8u2GzvRO y4KtHgTCxE4AYcDvYr2+PZC4m7LzT6b81KO8Z3h93zkBGQqBp90pMKx7rYgwu/C7 0yLn62yYuBBGlF8IZdmoLIRHRgmCWaBTi0gai3MBau/M5qa07pVEWkqOAxhaqinn orIWhsc/F90= =n9Dz -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] turning html mail into text ?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Wed, Aug 17, 2005 at 10:04:07AM +1000, Voytek wrote: > what is the best way to turn html mail into text, preferebaly before it > ends in my inbox ? lynx -dump foo.html > output.text > if procmail, is there a ready made recipe for that ? You might need to use munpack followed by a bit of perl that looks at the unpacked files and makes decisions about what to keep, what to convert and how to pack it all up again. Someone else may have a suggestion for a good Open Source MIME translation engine, certainly procmail doesn't do it natively (yet). Remember that MIME email can contain a whole tree of mixed bits possibly including multiple copies of the same message in various formats. Also, spammers routinely send malformed MIME in an attempt to sneak past statistical filters (e.g. a large text block that looks "honest" to the filter, then an HTML block containing a completely different message that will be viewed in preference to the text). Might be a complex job to get it to handle all the variations correctly. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQwMlXsfOVl0KFTApAQIB9w/+LT1JAemw9VZLQ5MsGaYTO5go2/zgYLuA qZIWDYwAZ7fOgFyq9tJxloDGacNkRejfmY3pMiBcydS6DejLAErMbgT8r5QxiPyH pKgJ97zfdGKPV+Di2UMfROEH3NLsdUtAe4zSnruAlJhdPRpEeOpS4VpH2OjQMRkf HHymdheUjw7FqcRSXOaGC7rvPkrZ7xLLUYESj35ZIN7wk3KgG3aYe0VUaZDmv2ob 1qvSIb9Ct7///hG/eqwp9W6iDqDgL30yzmrj0zHtOQtJh+o5Y+90Odgd7FuN31U6 GNSUl3W5+Qa6qLCFodqCNW8BfYyAW9M3Q9444X5uUzgrhb3up3CjpB/J7p+G8VUK p3Uvat6KGz8IOgA2Od8rtwRN5Gm78UW1nCrlvBPS8Ecmos/S27z2KfBxDirWoa3e r84xy/+nNoL1Hyuln15abVjivMVbnVkaQZNAZvyoEOKsDpgoYoIwp4MHdDDGko89 W4gzpv7wnZxchFFavlk5/Ji7a17Ym1fZ6Db1/umkXB1WRfqLfZEXFfluIJSsOKrf cZH7umvjOuXhAGgKQZdTWQ2In+HFvU3Piwz9Rv8mxaw3nUrD2QFTaVqXUEQEgzLN Typ0FznZV/viEqXHglRKx9DkS+aSGSje3lpZIHLpF93VMMFdfxMLmctJ+Nx+diYn bs8QDlcPFuw= =LM8t -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Gimp v Photoshop
On Wed, Aug 31, 2005 at 01:26:06PM +1000, Jeff Waugh wrote: > > Are you doing 'save' or 'save as'? With 'save as', you will have more > control over how compressed you want the image to be, etc. One of the gimps hides some of the JPEG options and you have to click an "advanced option" button somewhere to find them. I've noticed this is version dependent. Coding coding coding Keep them buttons moving Rawhide! -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Aptitude command to TOTALLY remove KDE
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Thu, Sep 01, 2005 at 08:37:21PM +1000, QuantumG wrote: > Sam Couter wrote: > > >No, what I said is correct. The kernel is largely irrelevant to the > >end-user experience. > > > > > > Whether or not the kernel has real time scheduling or not makes one heck > of a difference to the end-user experience. And, (for I think, the > third time now?) although you can hack real time scheduling into a > monolithic kernel it's not a sane thing to do. MS-Windows (built on Non Trustworthy technology :-) has a level of priority called "real time" which happens to make no real time guarantees and for that matter is merely a "do your best" type of implementation. On the other hand, so many people rave about how great the MS-Windows user experience is... go figure. Does RT-Linux count as hacking real time scheduling into a monolithic kernel? Certainly the RT-Linux real time is as much a "hard real-time" implementation as you can achieve on i386 architectures without using dedicated I/O processors. I would say that the implementation is quite sane although not particularly convenient for coding to (but then again, 99% of code doesn't need it and the small amount that does need it also needs very careful design so the inconvenience of the API is probably the least of your problems). In other words, taking regular user-space bloatware applications and cranking up their priority to something that happens to be called "real time" is a meaningless gesture. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQxbnxcfOVl0KFTApAQJIhg//ZpWmdrfMKdxrwwr8HHxZGe4S5ZIEio10 O5XzGQO9ENu4D1Y4oNsnLLg/1uVzAIcYaLOu8IWYKIolZ2gtGD9LIO7TIaUHkzc+ MdEzFE4GY2He81RZLrIE9moS/RFJXjoAT/rdlbWMNMdyd9DeZLdu/kmGKT/RLohN J5cNVUALE9opvwOHwfZoAu2R9dmODgvypmc8vo1Yw6Ty9TUZ9nPuA3pmG0n6fqUe mjny3yRZsKFEItofxb6bSaEj3+DAZaPbBkWIhqN8eriM6Ehu6SpcudcaC/96qtDg NlguVcF7QHz/K0vil79NExNXj4mgfZSfIWRZjlogFiXeZ0AOFUYIwgOVYeze/kle zbgT3/wwhS4xGczuCUDj3WL63k2eAoe7mLzrsAROhNfl1RoQB3NkArfgYXfpbIeC gzUZ/3eS3H5p/lLoqeu+V2f7i7T9qW1o3wrnywUMzfOXK0VjA2wZsfxOa6JRTHoX WSm1P4u0524smNk9VezC1d9CiYSEcCPEMBIqzW90XDwfaN9kdY1VYdkiomytAEpr 5wepnEzH3ovg/2uUmVUHrtQzThcnHoYaNr8bGowhtob/GOYdCVVBrc4f0waild4R 4izbPnyR1U4bcX3GsxuOmQvYxzQ5DlLfIMBCD3Po7D3+omQ9Y/x//P0oUKiYijck TdnnVoj35VI= =Txjq -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
[SLUG] Re: Aptitude command to TOTALLY remove KDE
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Thu, Sep 01, 2005 at 08:48:41AM +1000, QuantumG wrote: > I didn't say it was impossible. I said it was insane. Yes, you can > hack real time scheduling into Linux. Yes, we do have a kernel threads > and kernel re-entry in Linux. Does Linux have the best architecture for > such features? Hell no. The result is a massive blob of complexity > running at the highest privilege level on the system. > > As for whether or not a microkernel really made a lot of difference to > BeOS.. it contributed a lot to the overall system architecture. It let > the developers compartmentalize concerns into communicating servers more > readily than a monolithic kernel does. Don't forget that a microkernel introduces communication overhead and usually some extra scheduling overhead which in turn eats into performance. I seem to remember there was a big squabble over who had the fastest webserver until Linux introduced a kernel level http accelerator which blew everyone else out of the water so badly that they first tried accusing the kernel developers of cheating and when that didn't work they just stopped playing that game, took the ball and went home. That just happens to be one example of a real time task that Linux turns out to be rather good at. However, other real time tasks (such as queuing live music streams) have proven a bit more of a challenge and to some extent the idea that "real time" means anything specific is in itself quite wrong because there are a great many different tasks that have special sorts of time constraints (e.g. low average latency, bounded worst case latency, consistent latency (low jitter), fast resume from a power-save state (power efficient), accurate and reliable internal timing sources, etc) and depending on the situation, different constraints will become critical. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQxbtMcfOVl0KFTApAQKoZhAAjecVbjt1TkSmg2caEmMCQNuShJxY8uPd /wC/ut9FYp3HqdZPJEFC15z27OkedkHV7zKorrCQs5vXzSgqckkXPkUeqK1yjUPd IROh7xtxkVr4TobCBR/sXI7ZinbcU66jPxvPAAibKAhYsbTk2/0AgptGNZK+Xylj G8o26E9228nTdnseUd8wqGWJxtCT1/1vcjHy9vdZuDexI/zQYC8HKiuShsf/92GX RvB+2pJnfgMK5VzmzKYuw77hxA62LHJamr5pt/9UwjsHeUZQ6Jklmw32RHxiCNDo +PC+7sqpSRiS1erSjsg32soS+NWVrpeHq3E0sLANDGzTPM+YRLGLep/XC30ay37H Fx/Q6SIKtJlxVguqA69BGWZSd7Oo5oitX5+P0EEBvAY+Thhr7Oxufmd1wvsVQrel 0lOXLus3zr1lkhZ5kUhuQBLNvvNCwuEVdn/kJ5D3V8fm90Ma2Yq76nxs/1Z5qee5 NiyLgHut8ODlpBa9mQJ/t1EnPhApTrX6jtbK6Bs4EbdZNdp0B2z+tJYrdO+qgpgH zeIvSL1sP7P4APhMUdzblzBeVzomJ2ughFoFjtEzXflkLrOsNm6k1zQtTX6TkQbc syKJySo9IrHmvUAqWsd+cmyy8OvTeTquNBR2xbHIqMrvR7Dbxz+Bmtoe3uQz3rcj wd3YlBB2hNw= =bZn5 -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Re: Aptitude command to TOTALLY remove KDE
On Thu, Sep 01, 2005 at 10:05:40PM +1000, QuantumG wrote: > [EMAIL PROTECTED] wrote: > > >Don't forget that a microkernel introduces communication overhead and > >usually some extra scheduling overhead which in turn eats into performance. > >I seem to remember there was a big squabble over who had the fastest > >webserver until Linux introduced a kernel level http accelerator which > >blew everyone else out of the water so badly that they first tried accusing > >the kernel developers of cheating and when that didn't work they just > >stopped playing that game, took the ball and went home. > > > > > > That's because it's a stupid idea. Really, it is. Really, Linux kicked everyone's backside. Really it did. Runs on the board. You may argue that the test was unrealistic in the first place, and many people would agree with you, but FOR BEATING THAT TEST it was a good strategy. > We can put all the > code of the entire system into one process and run it in ring-0 and > we'll have the fastest system in the world! Setting every priority to mostest is a common strategy in Australian management circles but it doesn't actually make things get done faster. Again, going back to the webserver test, I'm sure all the other participants were also trying their hardest to get impressive results and they failed. Getting high performance out of a system is a bit more complex than you suggest. > Until it crashes. Which is > what the argument really boils down to. When something which is part of > the "kernel" crashes, do you want it to take down the whole machine or > do you want it to be contained and replacable? A web server should run > in user space. End of story. Debatably a file system and a network > stack should run in user space.. if you can get sufficient performance. If you want to trade off performance for the ability to recover from bugs then you can run "user mode linux" and you have got the whole kernel running in user space. Mind you, if your UML kernel is crashing regularly then you are going to have to fix the bug sooner or later, just restarting the job whenever it crashes is a bit crappy. If you do not want to go the whole hog and run a fully isolated virtual machine then you MIGHT be able to recover SOME of the time but think about a situation where you have swap-to-disk implemented as a user space task and it crashes. Can you recover the situation? All memory on the entire system is now in an untrustworthy state so you must find a way of cleanly restarting every task -- most people call this a reboot. Same problem if you have a filesystem task in user space and it crashes -- what state did it leave the drive in? Restarting the task might quite likely create further corruption. Again, if you have bugs at the filesystem level then there really is no other answer than finding the bugs and fixing them. Restarting the task won't actually solve anything. Being able to make an attempt at recovery is a bogus sense of security. I might also point out that some hardware devices can have nasty effects on the system when used wrongly (e.g. graphics cards with DMA) so restarting the task isn't going to help you if your PCI bus is locked solid by a bad graphics operation. A microkernel, by its very nature, can't be smart enough to protect you from these things and as far as I know there is no architecture where hardware devices can be guaranteed free of side effects. > For a desktop operating system (which, remember, is what we were talking > about) today's hardware is so much overkill that you could run different > parts of the kernel on different parts of a local network and still get > adequate performance. In which case, why the earlier comment about the importance of real time scheduling? I completely agree that a typical 2GHz Celeron can execute any desktop task (except starting Open Office) in shorter time than I can react to the results. I don't understand why scheduling is such an important issue when the system is almost completely unloaded almost all the time. The only things that I have noticed killing destop performance are: * Swapping. If you are running big-memory jobs and you don't have enough memory then linux will swap out anything it likes including chunks of your desktop code. This is not a scheduling issue. * Choking the disk bandwidth with a large backup operation or file transfer and trying to open desktop tasks at the same time. Linux has no priority system for jobs waiting on I/O bandwidth (and such a system is hard to implement because of the many layers of abstraction) * Intensive graphics that are beyond what your system can actually handle (in which case you can't win, get a better system) I would be curious to know exactly what "real time" requirements there are on desktop systems in terms of something that the user can detect. For most operations, 1/4 of a second would be the smallest time interval that a human can recog
Re: [SLUG] Re: Aptitude command to TOTALLY remove KDE
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Sat, Sep 03, 2005 at 08:28:55PM +1000, Sam Couter wrote: > telford <[EMAIL PROTECTED]> wrote: > > If you do not want to go the whole hog and run a fully isolated virtual > > machine then you MIGHT be able to recover SOME of the time but think about > > a situation where you have swap-to-disk implemented as a user space task > > and it crashes. Can you recover the situation? All memory on the entire > > system is now in an untrustworthy state so you must find a way of cleanly > > restarting every task -- most people call this a reboot. > > The point of a microkernel is that for the swap to become corrupted, the > swap service must be buggy. It doesn't matter to the swap system how buggy > the graphics or network drivers are, for example. So if your graphics drivers crash out with a bad pointer was the error in the graphics driver or was the memory corrupted by a bug in the swap service? > Different tasks can even be using different virtual memory systems, so > if one crashes then processes using the others are completely unaffected. Except when those tasks interface with hardware and the hardware can do DMA which includes storage devices, network cards, sound and graphics drivers. Then there's anything that can access the i2c bus (and thus do things like reprogram your CPU clock frequency), anything that can reprogram interrupt delivery... Then there's all the various bridge machinations such as opening an AGPGART window. Yes, you do get memory protection and yes memory protection is a good start but it certainly is no guarantee of safety. What I'm saying here is that while a microkernel is a bit safer than a monolithic kernel... the microkernel promises a lot more than it can deliver. > > And because microkernel designs have been tried (e.g. MS-Windows NT) > > and so far they have proven both slower and less stable. > > NT was never released as a microkernel, although that was the original > plan. It has only ever pretended. NT's instability is mostly due to > shoddy graphics and other drivers being loaded in kernel memory for > performance reasons. In other words, the instability is due to *failing* > to really be a microkernel. > > QNX was the first microkernel to demonstrate decent performance, and > that came at the cost of flexibility and performance. More recent > research has lead to more portable and flexible microkernels with high > performace such as L4. I think that's a bit of an unfair comparison because QNX never had the same scope of applications that MS-Windows attempted to handle. QNX was written by a very small team without even attempting to cope with things like vendor-supplied Direct X drivers and the like. > When Linux first came into existence, > microkernels were still just a neat idea with a few big problems and > were a long way from being proven. And now microkernels are a little bit closer to being proven with a long way to go... and still only a small hint that they might be a little bit better than what we already have. > The high cost of context switches and switching supervisor mode (ring > zero) on and off on x86 makes things difficult for any microkernel. Most of those problems are solved with SPARC machines which are easy enough to get hold of. I notice that Sun never tried leaving their basic BSD kernel design even though they designed the SPARC chips with fast context switch in mind. Besides all this, the real problem with current Linux kernels is not that developers have problems tracking down bugs, or that they find the bugs are difficult to fix. The real problem is that cutting-edge kernels don't get tested enough to even detect the bugs that do exist. So many people are perfectly happy with older kernels that even though the number of Linux users is growing, the number of people testing new kernels is getting smaller. Changing to a microkernel isn't going to fix this either. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQxzDfcfOVl0KFTApAQJlkw//Vv70gdv7lD17cxQ88tEKftktxSyd/59o 6EAqs4FNytZH/NSGB+UkcRQT8n8A/G1FV2bWbC+hHsk5QAYWQoAgrLdWMgNVfMkd 6zI0kAjd0l/85V0YyoNg5fZ+2nYjba9dEoXHtAtFWxNuA9jBt9SqCzc1L2146z1t asiu5ZtsHsll4Bq5i9JCpcOpCxJ4yIMf1EsITlF0MBjeZCjsLu06Oql1FU31rRed NzXnLvTHf2ozBY2AI23QfzVKrDxx4CJiFBXXedJUv0DzAEH/B6/DZ/ZOQB70ba+w P6G2SSKDyIM+5esyZaXyPZrAv15OCitCkYFyKXxi8YUEgGvPuenDeI2e/6TJSbk8 RuUs2hXPTkcKaHbtNa4+jyrb0nrsMxS10hBgOXHHg/a08l9PEV359ZWuHELhOfIz 9dJutXryKS7etncBlRwKyRaBLq9e9t7rMd202G3fSRUJPYqs3pPqsRM/fttk1+5+ 6CHMGJkthgTT5Le2TLTGynSyqSHGz0PnRPeNG4mBphHEd0kd2reIPpxulsDLOBjE 87qpn7I9//PCkKPM25j70pa6qGEwtOXh0UyvQsCNvgCuVlPaV0PWBSBF6M069n4u HZDdNMaei2pyf9GN1MZaEFZ8SNaSfN60xJe+L8uO9AjKg0G4Em8vTLhmhVzowk0S jHtGkPmpIhw= =IISq -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] gzip from perl script
On Thu, Sep 22, 2005 at 02:50:13PM +1000, Erik de Castro Lopo wrote: > One thing thats been bugging me for a while about Perl is the > lack of a butfirst keyword. This would be *really* useful for > constructs like: > > { ># Huge chunk of code > } > butfirst > { ># Second huge chunk of code to be executed ># before the one above. > } ; Don't be silly, of course it does: -- #!/usr/bin/perl -w (eval{ print "foo\n"; print "baz\n"; }) if (eval{ print "bar\n"; print "bing\n"; print "bong\n"; } and 1 ); -- or just define your own syntax: -- #!/usr/bin/perl -w sub later (&@) { @_ = reverse @_; my $x; while( $x = shift ) { &$x; } } sub butfirst (&@) { @_ } later { print "foo\n"; print "baz\n"; } butfirst { print "bar\n"; print "bing\n"; print "bong\n"; } butfirst { print "zap\n"; print "atooey\n"; } -- All of which I'm sure has been demonstrated elsewhere. But then we all know that if a programming language can define what is *really* useful then it might as well just define the actual program (since at the end of the day, a running program that performs a task is the only thing that is *really* useful). Once we have such a programming language then we won't need programmers at all. In the meantime, I'll settle for perl and make my own calls about which features I find useful and which I don't. - Tel -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Your top-ten linux desktop apps
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Tue, Sep 27, 2005 at 10:23:16PM +1000, Erik de Castro Lopo wrote: > QuantumG wrote: > > > Erik de Castro Lopo wrote: > > > > >You will notice that something like the Array.mapi function is > > >much less likely to contain errors than the C for loop. > > > > What I noticed is that they invented syntax when they could have just as > > easily have used C syntax. Way to knife your language. > > Nice troll or was it? > > Actually, O'Caml is part of the ML family of languages Which is where > it gets its syntax. The ML languages date back to the 1970s at which > time the C language was not yet even a teenager. C itself started in 1972 which makes it quite comparable with ML, however since we are tracing ancestors here: FORTRAN(1954) Algol 58 (1958) Algol 60 (1960) CPL(1964 ??) BCPL (1967) B (1970) C (1972) So the roots of C go back almost as far as electronic computers and the language evolution was systematic and well considered at every stage. Well, everything EXCEPT for the operator precedence which remains to this very day, ALMOST annoying enough to fix. If the ML designers were going to borrow syntax they only had a few places to borrow from: FORTRAN, one of the Algol-like languages, LISP or maybe APL. Borrowing from CPL or BCPL wouldn't have been entirely silly even back then (although the smart money would have been on FORTRAN). The thing about syntax is that people just can't be bothered learning weird-syntax languages. Look at LISP for example, it's another really old language (in computer terms) and it keeps trying to catch on but never will because the syntax annoys people and all the really cool ideas of LISP have been absorbed into other languages by now. The thing that LISP got wrong was 5000 years of humans using infix notation. If you want another example, look at oaklisp (yes, go and search for it). When you read the design documents you can't help realising, "hey, this is actually Java," but oaklisp sat on the shelf being a great idea with no one using it for about 10 years until Sun came along and reworked exactly the same idea with a C-like syntax and suddenly it got popular. In many ways, syntax means nothing because it really has nothing to do with how a language works. On the other hand, syntax means everything because that's what people first see when they look at the language and that's what they have to stare at day in and day out. Think of a great program with a crap user interface, that's what a language is with bad syntax. Suppose I make a language where all numeric constants have to be entered in base 9. There's nothing wrong with base 9, it's a perfectly valid form of numeric expression... but no one uses it. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQzvcn8fOVl0KFTApAQL4yxAAgJ2W1W+Q6rL7hXmc+NpWbxIxEazK97cG bX67a03NjMxYmQxZIVq6IIl3g1XPmkCXeSAkueUM+2G8sylIqeLRyT/tfXw1vsst d4TTbPm52ZR4a31nJ4SITDcmuiTFN+wCMjmBHFMzWDp3drBq4EWn7hiltAn790oX VZpqphlqquoquMdcgCTFC0ffcRf5DdNSNPIXS4Xt5CZwUeLxi3e0fhw4OWKF6RaL D0npcq65hWVKaYvF/oqtXugOixIlmksRZa4kH3rl9TZbPhZgy385RjRG4SNb1Y5A sdSbZ4qpURNFdRzwRouvSeuWTiPDh84u2JWR1PWYbdZkzDL66VR81emWtcYWeY8p 6FaxTDBnUoVrrpjWX5ADwKOXXmfdOD241umJcVqeWi10oU7foENXfEoygLti53SC rQ0hy+ks1ao1udpa1bgWV+cZRU+CKhAFFTHpQ8yPiqj/e1xv+bROcVhBeLI6fass neShBVvA9hM671ZJxTwx0ImvwO/FoWWiQ11+8/SiZZcsvoD/lw3l77w04QGWE4aV 2qd4Gb9zJLFo1WWk/wDfZnZDpHfG8YPPgBiEWyCulTqM3bdwkcAdj4wYvGZfu01N RBsds1a8HDmd+c7O52V++Cs2qBKysyo3re9M7ArZdAqdZGOUuXghqHyQ5W+tT8dR pLCiBVTOJfk= =5SLL -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] C-Pointers and Perl ?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Thu, Sep 29, 2005 at 07:59:01PM +1000, Erik de Castro Lopo wrote: > For just about everything you can do with a pointer in C there is > a better, easier, less error prone way to do the same thing in > O'caml and write less lines of code to do it. How about writing a network protocol stack. You get a packet and all you know about it is that here is a block of memory. You then have to figure out what sort of packet it is, how long it is and what structure to give it. C handles this very nicely with pointers to structures that can be cast into whatever you need. > The one exception > I can thing of is writing devices drivers and operating systems > where you need to direct access to harware. If you are talking about I/O ports, you can always write a pair of functions to write and read from I/O and these can be added to any language (possibly even as inlines if the language supports that). There is no native I/O port access in C. Consider how useful functions like memcpy() and memcmp() are for writing an O/S. Consider also that an O/S is actually a form of self modifying code which strict type checking is designed to prevent. I isn't about direct hardware access, it is about having the flexibility to do what needs to be done. Try the simple exercise of writing an O'caml program that reads in a chunk of binary code from a file, and then executes that code to get a result. Less lines of code than C? - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQzxjfcfOVl0KFTApAQLhaw/+MudIVmEkOmHUGX61ILbt93q4v7uVUgpF WG9vxEReMxZMJNc/bMfAaTPI5M3p9msCgClgJnooGqGmLufToP2mmAeeGj2XXuvF abCyet6Gwr4xaFXbrboHfhWCUpDuIu7ItNQsSKaoNI/upjImA2lbV1K9sBixPfzS OTKkAWCQ+Av8ZGlYmvkiI01LRzFkIdkNbDZNDCOdJdtmsdJLzu2/dXuDscJGOQDi I48Q04zG5A4YADMCJMNQpN7GNOfPsfp6FtEKPTknJYKnOANrF2BkE9BOgXS4+kwN 7NugEOAskqmdBdfoisP/q4bdUArzSP4khdgRCtDGKSIx1hr57rUMfH+bgoswAbze fx7B8FAypyEbho4XazG+NIdQ6ez0ClnB/Dj/1fsDjxUNTPPTu9SuXIIz6Jk8ZVCP t6yyO+JgAzOZe3BgEoziSjvDpSodd3h6ORXQnmgx4FVM9oT6Qj0FXazCfN9BaFFz MBMq01mC72ISDbGSjDobfTt2XjgSXTjioQWw3hsWG8ao4OTfktyhF0Q/osd6ZAJm OyHulHuBC4e+pfycjt/U7E40iKcK2TaKm9rgQ105pmR/aLEOuD2MkKvqCVwJX2bp /vCjiNiVVzgeZvNc5bIAdtxMtr3/jvBIulsNrRJypt0QACFOfqHOQZLgmVQIHRWd Nbd/dzk/Hmc= =mpjJ -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Your top-ten linux desktop apps
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Thu, Sep 29, 2005 at 09:59:26PM +1000, O Plameras wrote: > I am now not sure because I don't have a 64-bit machine. > > It is easy to check if one has a 64-bit machine. I'm curious to know. Actually, just checking one 64 bit machine would not be enough. If you stick to Linux and gcc then you get fairly consistent results but C is bigger than gcc (only slightly). - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQzxkV8fOVl0KFTApAQL4zQ/8DpoD+Qhq4hLNJI3Fax1y1cTxYdrUygBT gXHCk0yLkXLjLMtzaYGNnRVV9Ob6JTCuGPP6llIBSSBhSWkusH+ruxKw4PtQgrOB bumYztj2KYU+/nSDWQTLSrbu365mukGO3G8nGJHEJYjYUqVx3d2AB6LZC6+qpe6y EsRNVOeOfz6kY674Ce9fFqIDoQ84Wgi4fhfZJM/WfPk951uWiGpFPK+dKJ5YPbaM 0J7IEPiCEg4PezVBPYsQpVWJd/eecZ/AP9Nnte6CkTxbDncUufBUV2uTpXj8Ivpa ZgHmPUUsX7+kTfe8dBgubm69wZt/MLjQM4p+uKURMCbWtvOhtoXWrDQz6O3tCTvq hvgM3Ox/z30AlXwMeU9YSat98StC7iaaCMJm9iBkmuVEBZ2LpO6SzHHHE88cLH7I g/Y6afBPQ0ogYjslkZHhhOT9KphKTmdSCogjDJKcfe+HMoMuJ0Y+YdnB6vn8jZ7V /ts1Efc3qbPUqKsxnR2WBRnrjKaEKab/TgPkRn3Z5Rb7/crkMLoqJ2WP8532GnTX e+IzMtW2zakJCBq0yi33qGp8jrhRhMj8fFxe98z0wYRTuLQ55LgO5D6+e6ijzkZn enDkxm7T+ajEOaDJWVVu9OjJhIYAXON4yPDoY/1w2WhsYm6T44DoKJWn22AqmBrH jmrI8lyEL0w= =4LgO -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Your top-ten linux desktop apps
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Thu, Sep 29, 2005 at 10:35:22PM +1000, Erik de Castro Lopo wrote: > The very existance and popularity of Python is a perfect counter > example. Python only got popular because Europe was so very desperate to write code in something that was not "made in the USA". Most of them don't realise that Python borrows heavily from "S-plus" which was made in the USA but better to let the sleeping dog stay sleeping. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQzxmMMfOVl0KFTApAQLUpA//dtLh4F8eK6h8RfFtcFfB7gVDcrPSvOvp 9Z7qLeFkfATYrwxtRN3GtuV8P9PtiaQQf4/+pgOdrkams12XwxeDX2EFsWh7R1zc UbObe07X9MO7H/TGRtvSKYva5hJUG4Sw4wVwPUHR+Ft/P3ec4Em2XeKW55ScgLSY 2zgA0bDt9pEzdgI5rMezlNvM3tRhNKw6YrBrEtRf5zuqQW8/n4e54eVT96FybsjQ 0B9n4w8AqiZGabS1ffQPtLKB8EiVRVYzCS/jRbcREh+OdOcUmtYJU11IoZPuMNSH yVS+Wjeu6gPjCpwudiuxS8q++GO7oYQOPCinOqR9wTqSK5t1qVTJ5YMltk8ibJE1 4iE9HCNQtDAsUGdst01moQjxlM7NFFYYc3GF0tNMdt2Q8ZBR9VaGvf+/nGB+Rpqk F3xAocirwEyqiJyc3Mk4rX0/xixvQJ4TNzEGQXkWB7hQXKIpkyuRztwIhlXbTgPZ N7l991ZC26fE4P8wNaKwM360glGy2hMVjDGddTg/swTYgoWlrLZvoGSJaPgQUvbg j2SuKcJFYMoRzUtH5N0UYfqFbC1c3zhHqzEqEN4cEGddfEL7RyYTRD6/EaBbPjiY YaCGdbEyilAwPZa3yNF0eKvxlK4OBOrxEVaRazE1dHOxGapQ7GjSDINVVC77h+xX 8PB8i/LgaB8= =V3/J -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Your top-ten linux desktop apps
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Thu, Sep 29, 2005 at 10:40:47PM +1000, O Plameras wrote: > This means that there is going to be minimal improvements from > a 32-bit to 64-bit PCs. Since all your pointers are now twice as large, any data structure that uses linked lists (or trees) is now also twice as large. Since memory chips are not any faster on 64 bit architectures, most of your programs will run slower. Thankfully, AMD supports 32 bit pointers by running in back-compatibility mode so you can still have fast programs with small data structures. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQzxnDMfOVl0KFTApAQLntBAAgc4iY4JCvWZUUarvZutU7Lx75wDzdXzD zV+kLgbK5E4jXbObMVDYmTv5S1cSgOleBl3X7ZxhvJc3M/h2N+9ufa/N5pFRqeWJ w+tfOuu2xIwSc9VEgDL4KoLm/ZADm/xrqU//yTz3rTZDF0vFXf0yzhxG711vPxJc zYQ3lws/7EKLICuPNHmbifZ0L+0rpPHW2yYETX8nPLN2ZmnBrJgvoK6y8CUR3A62 ovOiwsyS2+Y7vuxxx0Hax9YrdDgjOrNVLyqgyDZy5AtAXllX86LcjXcXIIy6fZ8b GEVEK9I8ZcGGFU6ygPuvRPvHi576Jl0Pi5qUAoccOgEW4ZWOuVrvUS/CJ2kxhjkX Rfy5spG7Ln6+1je0VqyzP8j1UgafrJqovOSbdGqtozvsUE//1cvQBofT1/ViyC2R sUcWR6u1XCBb11n9lMJBpYRK2FPXWSw5dWYdTfbQnO/jcoTASwZuDcDyqZcD7x6H 633CSBalfZ3D58p6n8hj5SpDBg254Ul1C6CtaE1fKdTGTt08+vdS0dlGw8TosfNS n5JAqt1MnDcyV0xo8W1YiRzf8050eIdGE+2y8zfuZme0arErjffoZaPktqb962cg qZ1jHnkco9tkHQ14gSWMWwN7p6NsH9uU3lghR4m3hpu8vYpZuGI2Er+Kwq7rlfdh e0AWt1/7M4g= =bwvj -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Your top-ten linux desktop apps
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Fri, Sep 30, 2005 at 07:44:01AM +1000, Erik de Castro Lopo wrote: > In addition, C is used for low level programming where the programmer > needs to be able to address 32 bit hardware registers. If int was > 64 bits, what would you use for accessing these registers. ^ (corrected) You would use int. You would also use a function to access the hardware register and the function would contain whatever CPU magic is required to access the register. The hard problem is when you have a 32 bit number inside a structure and the structure is packed so you have to access exactly that particular number with bothering anything on either side. Nothing to do with hardware registers. The structure could be out of a file, off the network, or read from any device. O'caml programmers obviously don't spend much time thinking about binary compatibility of data transfer to other machines. Oh yeah, I remember... file I/O is not a "functional" concept. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQzxpwMfOVl0KFTApAQI9BQ/+P05WcQJ3cTvSCkBGh93LAcI23DpIkgTc DZbCCs7h9Y0G2KChma49SsGz5a+AakXD324+i10u0MiSWaJC3sR3hpnZigNAMZqL UG85bgKXIc13DnWgsUaT9Ioaef7w1pAcqNTEU2RjoSw1TDSlE5KyxHP7Ko0RsKla 9A/VElDWt8uVo+eXP5rqSLNdaFi5QpkmgQRkWvsNlbBlghx8Rqz/DDMFt90IDg7W 5Y7ltyq8FEX+q9/TtJe9axl40fwofo+JykvEJhz4KGAW1/jNiGugwD/UBU92oOo6 CDbU4kXrfDTBqKR8ggUFbGH6C/dFuXSW3CW/UNzBxTmw8MY+ezrDDTsmB4XM+PY3 2bDdpDEWgihBPiwW/dMfCw1bZXjkkjAoPTAKYCwdS/uYg9TcPfF4tpylTisMLPnL 5YQw/87GPmO+5HyX0nlP8egVPqYoIFPfc/89NasqlHh8nIAashyOz1xUPWgHuleN Qg88nPwiY/u2u0FoUPoRQlAM5cQQ3w2GOXIqQKtw3MFQee5pEw3UvDpQwrdwYblO FAr9bM/Q76PAtykwJaMBUh7lnqWizrbJAVOtEdu7tr8NZbX9AytO36udL18M3DDa him+hOy0z/QOiz16UUXuS5pYUk+lX1gCj8grsqoiYyKwMM5RuW0SxDZOoEEn8kLM UOXOupu/zv4= =ZpjW -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Output from 64 bit machine
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Fri, Sep 30, 2005 at 08:12:53AM +1000, Peter Rundle wrote: > Compiled on a Dec Alpha running OpenVMS 7.2 > > VMS>run test.exe > size of a char is 1 > size of a short is 2 > size of a int is 4 > size of a long is 4 > size of a float is 4 > size of a double is 8 > > Bit surprised, I expected a long to be 8 but there you go. There may be compiler options involved, check the pointer size because if that is also 32 bit then your compiler is trying to be back-compatible. Not that I'm a VMS expert but I do remember a lot of tricky compiler compatibility thingies. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iQIVAwUBQzxqc8fOVl0KFTApAQIa5w/+OqK3bIF0ah93cuWKOc54IYFdMAxMyWAe 9mNFmHiy7yK8i9FXUCg09N4lmAa8rcuXJMB74370MEZ+j9+PbrGnNnt2dTF+BIPB VxlEQZKIuPsR2XVZvDCURzh5uesOtc3wG8ZleUZ0zPBUsHFi7OA0vkZtMNw8tFeX /TXUekchMyypvhNVH4nTN3Uuv3tfFZpcuPdCsRrVoSpbv8Dhlf25LplZjQVNKPs1 WC7o7mt6dd1X3emlvM5+buZssoxxvJMXu2fALcoDucPfM9njletK/53nizkzDAiU jCDhfCojBQBlr9TgCHq8zsazPSGXom02Au+yM86LBvuCUVBmGuyRRpzsIjCFscOU s8+s0G1a8v3wsxEh/udUD0FVovV7aNv/NleCk0QrIhSchjXMm/E9Crl9RmMsLpqJ SCM6zga+wJJsjauGvr0QByuKZXcwkqFcT5ht8o5Gh07bZKbtMSuTLMZ0JJXqDpAr uKJ/7agdtFjS1mQBo9ZhoyjKK2iq4ckWyh4CDIQ6e8nLp6G5+4SowiJ8ojc+S0v3 00M7PsrZ5H25iiNVPl/guj8BFCZh04H3/xoCiLrMeLwAebsTCER+AAISg9897oaT LjMr8CS4TMnxYvqphtJnBYzO3TUcTVtqPQOAQltkEtN4XxQCEOEkqyBJ/Y1CxrYd o9yt6DVb86w= =r+7k -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] C-Pointers and Perl ?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Fri, Sep 30, 2005 at 10:10:25AM +1000, Robert Collins wrote: > On Fri, 2005-09-30 at 07:58 +1000, [EMAIL PROTECTED] wrote: > > How about writing a network protocol stack. You get a packet and all > > you know about it is that here is a block of memory. You then have > > to figure out what sort of packet it is, how long it is and what > > structure to give it. C handles this very nicely with pointers to > > structures that can be cast into whatever you need. > > Its straight forward in python and smalltalk. you have an array of > octets, do what you need. What you don't have is something that can be > pointed at a random memory region and used to bypass the type system and > vm ;0 So with an array of octets you need an equation to reconstruct a 32 bit number from the array and another equation to reconstruct the octets from a number. Then you write functions something like: read_u32_from_octet_array( array, index ) --> returns u32 write_u32_into_octet_array( array, index, data ) --> updates the array ...and so on for all the different things that are going to be in the array. Excellent, you can now work on your network packets... but you have now reconstructed the entire concept of a "struct" and done it in explicit functions with explicit equations. Not only do you have the slowest network stack in the whole world but worse than that, you still have no type safety because if you write data from an index which is off by one and then read it back you get completely bogus data. Some of those bogus values could be packet length so when you get them wrong you read off the end of the array... Oh, goody, python and smalltalk are protected from reading off the end of the array but how is that going to help? You won't crash but will instead throw a runtime exception which is only mildly more useful than a crash. Sure, you can't have a bad pointer, and your program won't crash. On the other hand, you have to carefully juggle the array index values all the time and one mistake will still leave an unrecoverable situation. All your high level language with type protection hasn't done a thing towards improving the correctness of your final program nor has it made it any more difficult to make an error. A much better method is to do what perl does which is to treat the network packet as a binary "blob" and then provide special "pack" and "unpack" commands that turn a binary blob into local language constructs and then turn local language constructs back into a blob again. This is still slow, but each packet only needs to be packed and unpacked a few times. Then all the difficult magic is tied up in getting the right codes for the pack and unpack, once you get that right, the rest is reasonably stable. Mind you, if you want to handle a range of network protocols, you need a whole bunch of different pack/unpack combos that pop into action. At the end of the day though, C code to do the same thing using pointers will be smaller, faster and probably easier to read as well. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (GNU/Linux) iQIVAwUBQz9gAsfOVl0KFTApAQKvsRAAouogH1xkowFocR6MtUppt7DIn5aY2N2F 8pGUqMEnEVc61jDTzTfomU2OpGS948o97QtdSfg2owonOl5A4XjkPYESlmhmoren M9MKcHBEVz3Uo334SKjaih5rp17bpF0xFnGV7nWk5TAi1qznMlU57NE2wE34LzE5 FlbBXfceL8tctXewiRwQwFOsyb7MJp9NiFMM/l6qwE+amx+as23bUPyWYrmbjbvr fml5IQNlQbhMR2JxINHDtEkxGYUivkQQyNuBwGYfE+v8pezSKXEpWQtEP2f03D79 Wvn3aVR+JOFXwTF/XaoB7Rae/sTmY0eyBiVmlw8k8pH/jeg6Lcxl7M+atOD/jykj HRiXrIWyQ+Nj8mod6OvHvSMFNO4l93IYOXDOCNnshG3PaP9jeN/TGS8v8y2UtWzv Dm+A2iWEagwu+GAvOwMH2BzDzJKXVEdd9MtcMLXhdM9MQpGiiw2AUSqmubH32EuL KFDL9Kyzk0UYab92YgGSHiFiJe34i6Koa9uCvGhY1odPVo4X+3W+uINPUegB2Q/h viU9hLt849RMSjQjSXnAdxoCrmvcehE2oMMxGlI+d7f+KXnSkKqs1cEPr1zdeaWD /lVTaNMXukuIqkqBgxuYTU7pK/xyEXq4HUOH6TRCckvnL0oQkyAGN83ujVsKzyAh 4NHG1JJdTI4= =UQi3 -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] C-Pointers and Perl ?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Fri, Sep 30, 2005 at 11:32:36PM -0700, Angus Lees wrote: > > Since you dared me, here's some (untested) perl5 code that will parse > a TCP header, including network byte order conversions and bitfields: > > my ($source, $dest, $seq, $ack_seq, > $off, undef, undef, $urg, $ack, $psh, $rst, $syn, $fin, > $window, $check, $urg_ptr) = unpack 'n2 N2 C B8 n3', $tcpdata; > > $off &= 0x0F; Pack is fine for a fixed structure but TCP inside an IP packet does not have a fixed position (IP options) so you get: unpack( '.', substr( $packet, $offset )) Then you are tracking partial unpack() templates plus offsets and copying chunks of data back and forth. Consider IP fragments. Of course it can be done, but making good use of structs and pointers actually makes it easier (as well as faster). Keeping track of structured data is just something you have to do and high level languages only partially shield you from it. The same problem comes up in binary database records, graphics file formats and similar places. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (GNU/Linux) iQIVAwUBQz/HQsfOVl0KFTApAQIawQ/7Bv/3lB6etPaCi+W2HkJeMGCdGANVZhKy qQVLQoNdY9z5z2KbYCrZ5JzYHQok5fEFbV4u4goi/G+UpyVJSDFxP5KBZjBPmM1y ehu+u5aRW3K3eT4CBxMNw/9hnecgq9tO2XCjqdm0bBOiANLDSDwIo4oIMQnCztok wfOBMntU5CHAg5u0llMYS692lbTCmYRyvER1SeNZklv1181G7OG2EVNMftoihei9 NnmS4w5dvxaAbAeHGJLfVxnjWVSU1eOBGgyiBksGKSTobS46FCiM/+qOJVk/rQEw Ok/Pyp/nBdJM9vtMShB9Fb4ENbAVldaso+4QScjzJbSDS3cVbtfPk6VZM/VEFraL Gjtrc3b6+8nsqmKCGWR+wJUMGxWNwEkFa1GVaBctaLsuLGRiCMIpMNDUbhsRtG5O gTkz0taWBRfY1Bjqm5RxSSrCzLr+n4DYCiouT4m9YQDttORYmxpBW4AglvFDdnMy Bqb1K+ebUEmXX+MHbPcVugA2PfvGkf95G/CD0lRBWyi0V5OS7CFGgBpj5FxaQoJw BEd1ISN2s7IVzrxB+rjVrfDVZ00/dz7MzOo2gPUZs4rz8NBAZEuuJDBxcJh5Lsye bmimBNY3pv69LOSlxuJtvZXgZFh3kDvvYBEdOEraowIpSh/GSvsGwec+elpBlq3G WJKWTVxQeF4= =ovmz -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Keeping passwords safe
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Sat, Oct 01, 2005 at 11:29:56AM +1000, Erik de Castro Lopo wrote: > Hi all, > > Does anybody have any recomendations for a program that can be used > to store passwords, bank account details etc in an encrypted file? I've seen "bio-drives" for sale which are USB "thumb drives" that actually require a thumb print before you can access the data. Won't protect you against a key-logger though. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (GNU/Linux) iQIVAwUBQz/IPsfOVl0KFTApAQJ5lg/+NWAIk3/jEal6tBfyBLOJdB2fnxItuqJt EecIHNWyjQ8YndUd6P8Ofoxk4oqYOCf5YzGY0h4KrWA3GzvHQ9XT7cflqcmbwptw SB+UmYolpzSNHtjcAVu/0/+DNnePaeCCFQkuYgSNc418Npn/MQuR/BlmLDynZcgW KfvxFR8/6VuqBRed6wfSqReLUBEY9ar9dVAb+UwXdd25Lui+GPrl5v8PpRj7Yr58 prIMJb+KXmJVfacgpZrVBwBq36SlpdQst4zWZKDQzQ5lImb1aomcHJ4A6t++4RgO rH2OpkAVM5NlzYK7gc2BIRwZzZDtInWHHhZLrhEzPanf0CxW+ldectMJHtMCkOvK rePV1CmsiOb3Dr5urGGI+bZ1DigdadT0FjdiCu8Ow1DmxT01pgj2Zg8u5JftIVpI qQbcmGsH6Ml/ANBPNyuwBWXsaOPYqk6XfTGuCmSPnZH3sSTMGFxRlAnaYEgqsG9D lbAOlz0sAyu7fMD2V4NqsFn7/4TFTftkqDLdZ9AMymW9M2o4G2auHGjQ4rTF53V7 t/RC2brMADqoB3R5QuvcHPg8WkPaWeG9sHt+AQDwcwJExcdzlXz7uxj3fiJBOjIf zp0d0ERX7egM0ML6SIl67PwgkbP9FwGfk17/X9l3FHblvQ8uHwmhlvHd5ti7j+ze LUNAA5oQaf4= =iIDc -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] C-Pointers and Perl ?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Sat, Oct 01, 2005 at 01:50:20AM -0700, Angus Lees wrote: > At Thu, 29 Sep 2005 18:38:31 +1000, Oscar Plameras wrote: > > Is their equivalent codes for ff in perl 6 ? > > Sure, perl6 (just as in perl5) has coderefs. In fact, these can be > references to anonymous functions or dynamically created closures, > which certainly can't be done in C. Have you looked at how the mathmap plug-in for gimp works? You type multi-dimensional equations (handling colour vectors, space vectors, matricies, etc) then it writes out a C program to a temp directory, compiles the program, and runs your image through the program. Slightly scary but really no different than doing an eval() in perl and mathmap handles chunky image filtering with speed and ease (you don't even notice the compile time). Maybe that is a bit of an unfair example, a better question would be whether the ability to not give a name to a function is really so amazing or what makes a closure different to a function pointer plus a data pointer. Certainly a lot of C code is written as a function pointer plus a GLOBAL data block (and library routines such as qsort() and bsearch() encourage this by not providing an extra data pointer for the compar() function) but any language can be misused. Is global data really all that bad anyhow? At the end of the day, something has to be global. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (GNU/Linux) iQIVAwUBQz/OJ8fOVl0KFTApAQJpAQ/+L/IiUjDMOTtLV00nT8nc512GCLOKo9U0 OYq3TXabvgCa4o1zjT4OZH9LbwFCc7ogN1xDXzf2gnRcSIOSVw498tEXvFVNTmRh aEODvZd4w2r/LotbDm9q7a4CvW4bJhYV9sxtpzP9qe9lDZWtn/O+48rl3vlZspLX bXJSZ1dEQ0YEAcRdXj/twefVzLRDCTf/n75ycchcJOng62GOkxbFo0o/VOKP81tb vpsV0HzrYSQtqTN465HXmFCj0VgCeLzd4IZj9lJ04rdYHYV77qYOdkJx8TIDAt19 lmdyJgf5tnfpABaYiEIwQvWLIXu4t7uYWJqB5Ic/EwtRLv5aeInLtVUlfsQP5Ygr Sn2Fqg59TQR7ylFDDsA2k7Uj2NPW+peFnVTPKq/hrsac1i/AT6VnZVFGKN2L+Jh1 J4XN4lWvI0riM08Hk1qUJYr8OsG5uAeMlIcwhBf0dKELmNNDyvznkbyz/XMb+G58 P4g7Cda4ODqQldeHWpnsvlYkzjOw4ER97pmEXHQwn8d205yUTMvs0Ki3fdTfjPo4 PGutvU+qWB/kczWy14CbmIE6sgdtMWt5shonmLjJfRzRQDJodUJBzEcd9NwCGgpG tZl7Vo/ZazMZE7SEpDEECr3lp0CwJ50VLdq9KGhHJTyo6kE+6/VQgxWL/3dRwH3O brTLHjY0WwU= =tvUV -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Your top-ten linux desktop apps
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Fri, Sep 30, 2005 at 10:13:34AM +0800, James wrote: > tick-tock-tick-tock-bing! 64 bit ints are touted as being an easier fix than > re-org'ing the epoch, so 64bit ints WILL happen and 64bit machines are better > equipped to handle this > >Explains: in 2039 the clock (32 bit) will overflow. This is Y2K bug with a > vengance and a certainty. Not everyone codes so that time_t is equivalent to int so there's nothing wrong with time_t being a long (and fix the code that can't handle it). Furthermore, most people don't hard-code the epoch calculations anyhow; they just use a library routine. The epoch could move without much breaking. Finally, by far the most common usage of the unix clock is to subtract two time_t in order to get a time delta and this will still work correctly even when you overflow the register (the magic of 2's compliment). > Will 64 bit ints waste the cache. This is deep dark art, but I believe NOT: > cache is for speed. speed = 1 access to get the data so ... > c... a char in cache > ss.. a short in cache > a 32int in cache > a 64int in cache I believe that intel already uses larger chunks in cache than that. The Xeon uses 64 byte chunks for L1 cache and 128 byte chunks for L2 & L3 cache. I'm not an expert on caching algorithms but my understanding is that reading any part of the chunk will still read from the same cache entry. Any cache system which forces you to keep multiple copies of the same data on the same chip is going to be very inefficient. Cache memory costs a lot in terms of space on the silicon and power consumption, you really want to make that space count. Then again, not all programs will have their data conveniently small so that it fits in cache. Sometimes people want to work with decently large hash tables. And to cap off, there are cases where you are going to invalidate your cache and need a reload... it ain't pretty but it happens sometimes. Small data is fast data. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (GNU/Linux) iQIVAwUBQz/S68fOVl0KFTApAQIVvw/+NDemLGW3hfOMlsE8rjFa/aTzRJxpx7FH OJaqZAK3oAJwT3WIW9SJiEklLo/vmi0H8W5SSZPUxAJogVyxN8AaQ1n0r2M8vs5z A0ckg4mIOrOOUBJj0Jf8HQhWtrklIZWrMJ9GapnAniwoSYfLT6D2zuusutVCNWH5 TakIZiA6AcSx8pD758p4iSfZ5VTDVhXCwTomfzbF39oCaC++e8BLO5xyZRnqEYv4 oFm+xNmCssLM9p/Syhe7f08aYd0IVBsTwZBVTSd81F5UzQ5DOfDm+uDh62t3/Tm4 KIrz10+gyMTLzDxFZCVXtg366OYps20tBKxPGF8VKPNGuZHO7wlQUrjFjLoOdCXN d1nUqUFreHQsQ9ftsWHa78vOBYhoogJ8wo7k00+6X4IQPFVYYCclyMIk33y7yCsw ZdIidjEE8U5GYbGzp9oN9QUafkex+5HEB47NMWzexsUEZDvcSLDjJPiVhkug9HSz vltldirhoa0wnyGME+wSWr1tIj/a0+RT+0TW0lGVfJM4XeICv1adR6EQPhxEB1E+ x1Qp9z2AbYj8HUIqZF3/9pGcvfmiNeXr1JVqH5gf+WVqfuuxw0nwZlfgW14Q+7o1 SsnIlbluMoD8BS0Pw0xWjpUqx67WA2ReJ3dcst4A4rZ2zNJmR1bc3OvRHb8aj3u+ JO3WF0f8xZA= =Ri52 -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] C-Pointers and Perl ?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Fri, Sep 30, 2005 at 09:26:45AM +1000, Benno wrote: > I know, but I don't agree with him. If the language is good, there is no > reason why it *shouldn't* be used for device driver programming. > > Device drivers are, in general, buggy pieces of crap, so having a higher > level language to program them in would be a *really* good thing. Remember that when you are coding to hardware, half the "program" is invisible and out of your control (out of the compiler's control too). Half the program that you are building is the hardware itself, it is an active component, not a chunk of memory. The hardware does things and you have to also do things in the right sequence, at the right time. Often the device driver code is perfectly correct, it just doesn't perfectly match your particular bit of hardware. Either one manufacturer has copied another manufacturer and got it wrong or the driver was written for an older version of the card and the new cards changed behaviour (but the manufacturer never told anyone) or the driver was written to reverse engineered specifications and no one really knows what the hardware is doing anyhow. > I've written bindings to allow you to program drivers in python > before, unfortunately the result wastoo slow :(. Timing is actually part of your program when you write device drivers. > So a higher level > compiled language like O'Caml might be kind of cool. In theory, it sounds good. You hit a big snag when you look at what makes "functional" languages work, which is that each whole unit of code has well defined inputs and outputs and the behaviour of that code-unit does not depend on influences outside the scope of those well defined inputs and outputs. It's a nice mathematical approach, but once you put a piece of hardware into the mix, half the program becomes invisible and out of your control (i.e. the hardware half). The "functional" paradigm starts to break down because any bit of code that touches a hardware register is now capable of behaviour that the compiler doesn't understand. Functional languages are good for transforming a well defined input into a well defined output. They don't handle "real world" messiness very well. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (GNU/Linux) iQIVAwUBQz/XiMfOVl0KFTApAQLlcw//cpsVINdmQBZTrxXTcdKP9XVoA7JjOiqA EsPnY+yuzztIZzC2lzNJrNW4jiRaNQZDM8TwhaEwDnIZgGgqR2uTp94H4OVDEMbH bTG2FUhZ+m86ytcS7qTSeVhVVhNj+JKyBT9NaDtGGJf6aYz9BqSHOtdRrMBC0XZO rxdraJg7l/otyrArxWaLZeLW2k6Dchi5XfNE6RG/R3kesKIlupk7Zts8UsAygRrq P5e7ZPI7WuGQBghxBWGAU5kqRX/8FadtrMEfPb4b7Ait4W4bWYRlSVKdKKTuhY5j 42FaCmwWNj8gzCDW8e5kjBRClu7gGFUSSNHWqtXSSN9PsvCtmmoxsXmelwYc63og QVnPeFDP8iBUhzsCOr0k9pvLkD4Pf9uDxKEs8LLTciwy7PRmr8m6ilbWYPPk9H+R uSrYT9xo50IUnEoFCP7JgzbYoUYF8FkrcgsIdV8zPsyp5q1nevjK12CtsFu0RBu/ 2GooS8MRSR092YAqPRvP7dgvIzk6Jo9lCGP9zbtPyjEd5BMhN1ZnTFJsnprG2+9L qrSwrwfW7jc1IXjSq4BIDtiPRJDjMyDq3A6cbubG6X/88XnDxbyRo3x+klpx9Yf7 3wOj77i7OhyfhieAd9UAWsDRkBA8NaHGKsKg9pyFG4YGGfn+znf/YRHCFG2HjCy0 +UunZVJvKOM= =tphK -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Keeping passwords safe
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Mon, Oct 03, 2005 at 10:57:51PM +1000, Erik de Castro Lopo wrote: > Mike MacCana wrote: > > > openssl can do this easily. > > It can? How? Its certainly not obvious from the man page. Is anything obvious from the man page? - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (GNU/Linux) iQIVAwUBQ0L8P8fOVl0KFTApAQI9SQ/9GqoZcF461Mc9nDDUB2h36byeSwf7OHtb HDdUYBdswOU/yxiOlRMvVn4TMPWQaO9oqEtYHiNGPFvxDj30Q1++nTp4QQGv6r3E HKKQ/01mp+Vk9BJWYQKfy5hkeBxRUy2GwB6rMShVkGl+KGHNWx4losYEUqVQq71N K0JMZSojRWvoNN0NFGh3nkQk7V3vKnN0tIggzLlj+t1Z7scaLgG1yaSHKGQf39WJ YGofjL5ZWhO2PyEoGc2zO12aaQWWz5ZTf5t+QDrruwV328YKMamic3ogcXTmzl7m yDRnArrBobEIuUIRZSjEqYIZCpb9pJhp6GjgqhL9mjnA0pV1UPyCElYoq1XAW8+R k59qERqswAMVyLl+RgLwflXB2bCUpFb1myYofjWNVAqEd5hJcJC96jHhzS33jQNk stjIl68WkIZMb5hNkdzO/me5xtfLVX88lAog+gnYwaaCZbFmIY49J2PIQHPejvX1 SQab80gxDYFdavbxTkNTZ9lRlHwinLy5qvQSOGN68QvH0zrBumN0fqrXSzh3Mv06 y5hU9dd4tPFd7mksmL0fJLVeABDOi/q2TQeS9MXoq+3RqNN8THZ1GVKpd7rXMsMc k0+OZ5S8K3Om2D2Mlzli1Ms78e3U33ULgpZ0bR7aH5LpHW665MneIHFo8H3GYYok J3T4bxU1IHc= =TxBL -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] returning windows software
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Tue, Oct 25, 2005 at 06:42:07PM +1000, Russell Davie wrote: > On Tue, 25 Oct 2005 08:44:49 +1000 > Michael Lake <[EMAIL PROTECTED]> wrote: > > > James Purser wrote: > > > On Mon, 2005-10-24 at 13:28 +, l cheung wrote: > > > > > >>Get a life, get a power book. :) > > > > > > Yes but can you return OSX? > > > > It's a good idea to keep OSX on it as a small partition that can be booted > > by default for the first year. > > If you have to return the machine under warranty the techos at Broadway > > (thats where the > > Apple service centre is) will then be able to boot it and run their > > diagnostics. You'll get > > it back with your Linux untouched. Otherwise they will have to do a new > > install and wipe > > your Linux. > > o, never thought of that... > > good point I don't find it a very good point at all... Hard drives are completely interchangeable and there is absolutely zero excuse for the official hardware service centre not to have a special hard drive that they drop into the machine for doing diagnostics. Moreover, they can swap your hard drive into a different machine that does a sector read/write scan to prove the hard drive is working correctly and they can do that without damaging any of your data. They can even boot off a special disgnostic CD or off a USB device or off a network device. If Apple don't provide such boostrap options to their service centres then Apple clearly show no commitment to their customers. If Apple do provide the option and the service centre can't be bothered using it then time to look for a better service centre. I can remember boot floppies for 486 machines that provided extensive diagnostics in just a 1.4M image (including keyboard test). I first saw one about 10 years ago. The quality of computing services available to the consumer has gone downhill. -- I recently had to deal with the official Sony service centre in Lane Cove with regards to getting a warranty repair for a Sony Vaio doorstop, which had some sort of fault in the keyboard controller. Their first answer to pretty much anything that they can't obviously diagnose is "software error, probably a virus or something, warranty does not cover software". I even got the guy to shut it down, remove the hard drive and boot it with no hard drive and the keyboard problem was obviously still present (because the constantly repeating key was causing a steady beeping noise). Then he starts with "maybe you spilled something on this" and after that doesn't get a good response he goes back to the tried-and-true "must be a software problem, probably a virus". The only option he gave was to completely wipe the drive and reinstall from the original CDs before he would even consider taking the machine in for repair. I'll also point out that this laptop has only ever run MS-Win-XP and the install was already the standard default Sony install with a bunch of additional software added afterwards. I tried booting off a Linux CD and found that exactly the same keyboard fault manifested in Linux as well so I'm 100% sure it is NOT a software problem and sure enough after a reinstall from the original Win-XP CDs the fault is exactly the same. Basically, the reinstall is just their way of inconveniencing you and giving you a few hoops to jump through. It is utterly unnecessary and smacks of the unprofessional buffoonery that is typical in the computing industry and especially typical of anything sold with a Microsoft operating system. I'll also point out that the typical Apple user doesn't no squat about what is and is not possible so the probably get fed just the same line of junk because they don't know any better. Now I have to wait at least a week to get the machine back and then put in a good two days work reinstalling all the applications, reconfiguring the settings, etc, etc. just because Sony don't see a priority in providing diagnostic tools for their equipment. Sony do sell some decent equipment now and then but their attitude to the consumer is, "You pay money now and then #uc% off". -- Frankly, I would be very very happy to find a manufacturer selling laptops WITHOUT any hard drives at all who would take the laptop back for warranty repair WITHOUT any hard drive and actually do their job and repair the hardware or give you a replacement machine. I would know I'm not paying any software tax and I could get a decent machine that does what should do. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (GNU/Linux) iQIVAwUBQ2Qru8fOVl0KFTApAQK65g//XNQeoWwrr2ho5Ig76FY968wxJ7Ioinua +yg/lpgN7eB3Heo0CxZ9vHg+5qUBojBBIjdOMPUa818M7Lliug2IIEbbfnYpRuaY 7k3GsTD7cihj+yT0DoebE7FBTIdIKASLqVkVeORmCLRyuGOfne/aJKU/9QjoAlLO 0FFdtFZNukrizLCNui9PrT4x4f4soYyvA10MDa8m2NrgAHsgaJxGtZx9k8DWDTZx /6tftFcN3WXQ+40MbovLxcTiA+ZF65ZhFZsRp0N1MTYbKSWcpk9ntIbGrHiSwyqN d4oLKzyoIOo9utxyjF18fOE9dUYjoiKlf5VSgXEYsLjWa+YIG3OJ4Rx
Re: [SLUG] Linux friendly flash mp3 players
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Thu, Oct 20, 2005 at 11:23:06AM +1000, David Gillies wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > Does anyone out there have any recommendations for Linux friendly flash > mp3 players? I can recommend the omni music-stick which is basically just a USB pen drive stick with space for a single AAA cell and a headphone socket. It has a little screen where it tells you the name of the song playing and it will even scroll the lyrics if you have the right files. For linux is just looks like a MSDOS partition, you can use the mtools package to read and write, standard USB mass-storage. The one thing I will complain about is that my unit wasn't as robust as I had hoped, a shock or bump can reboot the unit and the headphone socket isn't a positive lock so you can get some scratch and crackle from a bad connection if your headphone cord pulls. Also, it does not have any removable media and has developed a few bad sectors. I'd be a lot happier with something that uses compact flash or some sort of removable media. Mind you there have been heaps of reports that Apple iPods tend to fall apart as soon as you look at them > And any that have .ogg support would be a major bonus. ogg support is pretty rare. I'm waiting for the Linux smart phones to start being available in Australia the I plan to use the phone as an MP3 player and upload ogg support to the phone if necessary. I'm carrying too many gadgets around as it is so with a Linux phone I can roll everything together into a single object -- data storage, phone, database, organiser, music, communications, etc. The Motorola A780 looks tempting but I would still expect an expensive item to have removable flash and as far as I can tell the A780 does not so backup becomes more tedious and if you get a bad sector the whole item becomes junk. How hard can it be to design decent gear? The ideal item for me would be a linux phone with a high quality stereo headset connection, a black-and-white non-backlit LCD and a compact flash socket. Using ordinary AAA rechargable cells would also be an excellent choice. If someone knows a factory in China then I'll design them the ultimate "volkswagen" phone. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (GNU/Linux) iQIVAwUBQ2Qxj8fOVl0KFTApAQJpkhAAj+y3p0zwBviEjun/hv2emOaNlxGY0emK QUT6A4OEtLyYFDLh8oIIK9+7RmRNLcrAERlLyvSJ1LuMVL4th0s+Z+RuVBAP3Kf0 4dJ1eECWotVUJELxyiPUfHT2PfWJ8MayaDuUC2MPNBZWkAhU139AOxWGBhvt5Pmp 2fpy5YQZAyMZ2SvfdirNvXBs2n/Im12jkGXEdG1VQqtluiVL722ilI9tmluLA9yg kvbJdT6CO9iSPVxFy6c2tqASP6JHIDdDqrlr2O93m0CpXLkJaqi5jXs7HanDL5vY +RuneHMOZ7LyEg6mdCK2nAJMwKN5eD5pLo7ZX1q9ww9ZGJEo8RSRT0iLaOy0Wzbe FCloRNxUKOD1BanqTJJaXW6J309j4gaghCz44khCehSYn+wFOHcv5gsrXg+b3ytK 7pyq8/4MTV6NSnx4TXoZY5bfD1qWJLYfxu38zS9hVzgUkyPeVklXhyKmjZ06u8aM Npd5SQFK9jXuQ66cdQzikXLqE/kXDGncUddr9Jfr8W2OJ+DUT0Cyyqfixo5YOEpq h2dRaiVDY+eDqNR8g0cjlUurnSf7ao/SAFpf887rjS+YzLeqer0Hmoc7WQnCFueY wTnLOiBxW1grO7b+xpvJZDHOTZ8D6tfeCI8xF5tmV9a/vEPogy2m/0fMx2Nnz2vi ay+dYCPekmI= =ohuM -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: ocaml vs python/ruby/perl etc. was [SLUG] Why not C
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Thu, Nov 24, 2005 at 11:11:42AM +1100, Robert Collins wrote: > Well, the only bugs I've found to date that cannot be effectively tested > for are concurrent operation bugs - threads and co-processes > specifically. None of which are detected by compilers, at least not the sort of compilers we have available. There are code analysis tools that can prove correct locking behaviour (e.g. prove your code is deadlock free, prove resources are always locked, etc). I haven't seen one available as open source. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (GNU/Linux) iQIVAwUBQ4WjYMfOVl0KFTApAQKPiw/+IgMDnXb0JZoHersTZQmC4fPX8rMiIkf9 rU6oqPK64K6pdlK/z5m1YQa0Fx7OFMXFJgeSSu2wiK/st3L89KDAFs5eVCWpO2jz NBW7nByd6Suh5+blrr9GZ8PKAvrYbWiKcVc8KzCrSXd7wyaUXT7spJeP244XCZ/9 eo6YRTF0LBrWm5SzSTTmfvHEHuSf3O7LcUjZKzLinBLpFSEa4qkttiU4Qyq5eTtj bdDG1/xbWMEBfD8sYUiH8b+GYk9fYtr4aRLLgusH4C6PCscvZy21xnHyit9XpXbb MpZfX4dQFgaZtjg/H2kXh9qEs2P4J8pamG6DQxHJNfthcXBniQ7cL8z8DOo8h/4T c6JfQRNQDbAOLtRlzofojhrJYE7FEyOWIwdK/r18Mb0GuuAu8pY1fsAF0kgfuMZN CAaOE0r51krFppPQBbcapM4k3WXfyjfghgdtos2OpX/3v0v83O59dF62+Czyybh5 X05CizYsSH3x0DfXbfVIK2Y7BFSgLTI6BQeXAlTnirGhmhL1+j2Yj6Xl6pwQQCIV 0ZaoPiy443R28KpzKUG9gn8qZCAmP8ZoueyHbCUjfq4tc8sPa5tWhDYXC4OEDukn s0TsYSznkHpL3WYr6hHFo0ewkk2kpVVM2+JecIDedY01XF+NBzCgwDPEbHyDh4Ji WNmKO8M9mcg= =r2JJ -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: ocaml vs python/ruby/perl etc. was [SLUG] Why not C
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Wed, Nov 23, 2005 at 10:42:54PM +1100, Del wrote: > > "Less" is a measure of quality not quantity. If I have "less" money than you do, the quality of my money is still exactly the same as yours. > So by saying there are > "less bugs" what you are implying is that there are the same number > of bugs, but the bugs are smaller. That would be "lesser bugs". > What you probably want to say is > "fewer bugs". This implies that the functional behaviour of a computer program can be measured in countable, integer units -- an implication that I for one would hesitate to make. There is a fundamental problem with the colloquial usage of the word "bugs" because insects are countable and and computer misbehavior is not. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (GNU/Linux) iQIVAwUBQ4Wp+cfOVl0KFTApAQJndhAAjpOoqxLFQyyKGSvmPaFuMLuLSd/da/Vh kMwojeryA+ZltaEmX3I4l4TeLZ4uZu3U/cbqkTpDL+/Jy3nkox/7pr5miSzhaVst T4Q52oyKl2b1C6cxpNPAEON6ejrOdA5UoHkVYlu4y4YOwR5M28H0Afq2HIXTUo7l uV4LwGlMJ6VM6VKfydypOxOd0yJX1ByFqYtE7bIREoFOAWVQRa4wpraWJQBPnITn 2lUUrmYAqEhGiEJ0j/kXMIJ4zJoyg5ZKMLUAjg/bihB5a74sSVMPCjs+7WV/luht FNAI/wAZdpVIlFw7jEzsYqeoItKejZedlgkAR6pUmWmYLn17zV7WCZgGeFHspWru TuHz4tBsCWSZwc6CNAXVnOehxF6mu+pyHtKvPsmlfAoInK2HhqJP7bMeZ8s2UCaE g6s/wnuFpQ5M4GDatvxiMNy85Xtp9ICI0Sg2lhG0xbP8dOuuVofONRz/f5Db6XSk EILD/dyi1SgKntRvl/Yx+CSKEHfeSgBCQtGws/AXbBh6oMy9KIJ2BdTckxcMeP8u InonJRGbOJJVaFZeoH3NBa/jHwOwbbAOtDewBv9pRa1KDcJDC4o+TiNQ/Kk9vLTd T/xxLiZU5QQWjLAj/0lxbCzu8s9VOdpH+dcTo0Y0HMAUpuPYaOVDZe79o1PcXjfc cuKI10pomXM= =zBiz -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: ocaml vs python/ruby/perl etc. was [SLUG] Why not C
On Fri, Nov 25, 2005 at 02:21:41PM +1000, QuantumG wrote: > > Dist thou not knoweth how thee shalt speak?! > > Prescriptive grammer is domain of the historically ignorant. > > Language changes, this is a good thing, deal with it. Spelling is one of those new fangled fads that will die out before much longer. - Tel -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Help Me - C codes
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Mon, Nov 28, 2005 at 05:53:24AM +1100, Tess Snider wrote: > What's totally crazy is that once you've been programming a while, and > really understand this recursion stuff well, you have to then learn to > stop using it. It's very sad, because recursion is good stuff, but > the trouble is that, in C, arbitrarily large recursions make your > stack the size of Godzilla. This is because every time you make a > function call, it has to temporarily tuck away all of the local > information in the last scope out of the way onto the stack, so it can > go back and retrieve it later, when the function is done. So, as you > dive down into your recursion, your stack gets bigger and bigger. You > can visualize this by putting a large number into the factorial > function, and then putting a breakpoint down at your boundary > condition, and looking at the stack trace in the debugger of your > choice. Holy cow! The stack will grow linearly with the number but the integer will overflow much sooner than your stack does. I've NEVER liked the use of factorial as an example of recursion in books on programming because it is such a bad example that can obviously be much better implemented iteratively and worse, (perhaps not obviously) multiplying up integers is a terrible way to calculate factorials (not much better than doing division by repeated subtraction). - -- #include extern double tgamma( double ); double fact( double x ) { return( tgamma( x + 1 )); } #include int main() { int i; for( i = 0; i < 50; i++ ) { printf( "fact( %d ) = %g\n", i, fact( i )); } } - -- Recursion only gets useful when dealing with tree structures and in that case ALL languages will grow their stack (because they need to store the intermediate values). In such a situation the recursive solution will be much easier to write and usually more efficient to execute than the non-recursive solution. Certainly it will be easier to debug because the stack contents are easy to inspect. Try writing a qsort() without recursion... - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (GNU/Linux) iQIVAwUBQ4rn+MfOVl0KFTApAQLT/w/+KxAxByWByNRrysFZG8o0FpLObCZp4B/g S3aQ/5Q/39jTK491NEjwisag4Bv2/M6VRGy5efYAc//j/9Tp6VXyQpQ+8MFQ5D+9 xk9R9njKp12h67zHIPva7jQV5A05Crt/i3c5FATHpPVVRTTEaABl+5r4VlINpKtE AmxKxg+DmjIOQMAgsymtDP0WwAam9fegYWPJds/2e2GLWO3outbvtA9qerXJZ2gJ BSTrzI2Y8tdsvLgfzQrl4T/2+uvKBD6E/ezNvMcF1TKeBtFE7nrpU70qIoTPIXEe mBMKuXqk8yi6A0IbqU+384wQNbeQ8znlWPxwFWy9AyvNWLiNul45ovvvnTD1iWYK u5xdpceVB+MgBr3qSEib1T3UP6ACtA+TsWWcMA52z5DzDGEx8dyRCVk3bxQQgb9P DQlu6jk3Hj0loAfmxR43M+fU5CIePfFK/iYLgNoCKY1+a3GD2nAu2s0DzS24D5Ad Rqfiyfr5+T/+y6btPauKIfNyt3mhKq1/HJDFwzdmiGnPzt2mvoccaXKhZ5Tc5q+S R0xtLohHJ8uOAEWgCUi8P+3YFgWTWX3/IIJRsMqV38ILzJKl629mhaQpEHL3YMcg 0vayNrwZcj6SqUki5h/GmzzV+mdKrQevyfg/wDj/qRqCULrAaRYmQkJ3/OEnKKS7 f5TxnDWV65Q= =NEdk -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
[SLUG] January Meeting Friday 27th
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Despite the exodus to New Zealand, there will still be a Slug meeting in Sydney for those who are interested. See the web page: http://www.slug.org.au/events/detail.html?id=257 We still have no idea about what the Spice Boys are doing, I'm looking at booking options and it will probably be at: House of Guangzhou Level 1st Floor, 76 Ultimo Rd Haymarket NSW 2000 Could people who are interested in going out for dinner after the meeting please email either [EMAIL PROTECTED] or [EMAIL PROTECTED] with a very short email so I can get a rough count of the numbers? I'm expecting it to be quite a bit down on normal nights, if it is under 10 people then it might not be worth even making a booking. If people have other suggestions for restaurants that open late in Sydney city on a Friday night and cost approx $20 to $25 per head then this might be a good time to suggest one. - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (GNU/Linux) iQIVAwUBQ9AnY8fOVl0KFTApAQJPKQ//W/xj/QxmsQmZVyPVAupL7Qpz6/SXkkhj Ur5tyCoMfRaswfq8LdevPhFvcj8lqJzlFstwIcfKblKBHyvaBQ0emrxQsmBVKo4I 9puRQ1B1Xd5Kj7WbddJtD3KlGS7imGvFT0EEUhpDZ9OK3WvVhMhY4FYsKqx0bGRs Y8Ok7bZw7XS1bq50VEAts2eIMxQk6nt2PF/s9QtInzdtvLXBwnU4PtlwrVjPslm9 Y3hsZN7N2mUUds5gndwsI6ZJe1QEVspGefFThxOLY0rCafPjcV6uoEdkLlTl7zJm Wj/p12qHueb+t4qJs8V7YmEFeFWv8nYK2TJuClVsooM+cmzMFw8wDp9PgzwIe2/T RUMBSOs3XjMeBKRJjwc4Yekk40hw7EoHMZl7p5NL0+0QtjtURyJIo2X0VB5zYA6g s2WBxpA2ETOR6dW7aIcY6ANBPsWAjhDG1ep3zMRFwCyvDtBnlg836U4XQ9P21z9v RtK68vxqOdnmQhqq0dD/0206dWycZRt4J04w6spwzURzqHF4KztLdkweQTelwdMK HqYOvNH+uI2S5bza6/eBoceCIfG7C4H16K7KJCHFUUvRdGMPjy6U38J17CSdfK04 uHBV/5QPPoG+eW7mOWoshVh8KyubCRz7xo9AfuNwj9lpzlwm7VPl4UZ+AQpdmckk QHMhgIM3yc4= =cEOK -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Fedora vs RH Enterprise - consultants advising to change
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Wed, Jan 25, 2006 at 11:53:01AM +1100, Simon wrote: > Hi all, > AT the risk of starting a flamewar.I am being advised by consultants > that I need to 'upgrade' my Fedora Core servers to RH Enterprise as it > is 'more robust', 'better supported', 'easier to upgrade' etc etc. It is `more robust' because the new features go into Fedora first where they are thrashed around for a while then finally trickle down to RHEL which is always running 6 to 12 months behind the leading edge. As an approximate rule of thumb, RHEL is old but stable and predictable whereas Fedora is new, fast moving and sometimes unpredictable. At the risk of starting a flamewar Fedora is filling the space that Mandrake Linux used to occupy. You should also note that although RedHat back-ports bug fixes and security updates into RHEL, there can be times where Fedora has bug fixes sooner simply because it is running newer upstream versions. > We > are currently running them as our webserver (informational only - no > transactions), mailserver and intranet webserver (this one is a bit > slow, but just needs more RAM). RHEL won't make your system run any faster than Fedora. It you find it stable then there is no reason why you can't stick with it. Basic services like email and web have already been heavily tested in any Linux variant and your chance of running into an issue with these is relatively low. > I am unaware of any major differences in the products that would require > us to change over and start paying for what we now do for free - > maintenance has been trivial, yum runs regularly via cron, downtime has > been non-existent. There are two main issues to think about... [1] security: you don't know what's wrong till it goes wrong and RHEL is better tested. On the other hand, most often configuration errors are to blame for bad security and no operating system will protect you from yourself (although some will pretend to be able to). Taking regular backups is an excellent security measure. Regularly checking logs, processes, network traffic and the root mail account are also excellent security measures. You can take these security measures with any operating system. [2] bug fixes: RedHat seem to take bug reports seriously when they come from RHEL users and you get a response (and usually a fix) in reasonable time (a few days). If you don't mind fixing bugs yourself, working around them or you just never have encountered any then probably no need to worry. RHEL is good for the "I demand that someone fix my problems" type mentality. Don't forget that RHEL is a support service not a software license, so you have to pay for the service every year (else your machine becomes unsupported). There is also the minor issue that most RedHat consultants train for RHEL and that's the thing they are most comfortable configuring. You can always choose different consultants, I'm building a list of Linux related organisations in Sydney (probably still incomplete): http://bespoke.homelinux.net/buslinks.html Feel free to ask around further... - Tel -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (GNU/Linux) iQIVAwUBQ9dissfOVl0KFTApAQID4w//fdx3LcHkobRFDHKv0/sz/l0yCPQN3C/j 8xrt319T1mtZcDbImYYpyWwAOve8XNpEzkB92CNjGPn/uQUvboYxdXgMszPHnKE3 4J1i5+ljsNvqKrXQvu6dcl9diqFZ9swGx0Ua5ijgyH+fl6qVu0QKRPh306Hc4yC3 bwMudOXk29SME6nn69WceZYmVE6HaPsF+d3RdsZ5Q8tZ7sEGVMf03fCMuI/5hEbI DQoceB9YoQbdY+b4QWG0EPGeVyb1n7gZAedGYnNH7MhTr9lxiDrERupE/5qHpBTE Ps/Q2FGUzGfmOEN4D4uKc6fcIoF4uUGpjaEir2aiyMGD/feNcDUYZxj8DfN8fefa E4pfa1SlwsqSmFFa58uDX1gyNiLzTUgEYLGTfVueQV9psP4GyVcwMy609ssrLHIc 6dDbeNgpPXQpwmc+RtygP0LHJXvT/0PPjEuCmaSGKbMpz//kmWclhIFg/ydKwFYP +y8cLrW5sG6tUKq0mbgm8homWJt92FdHN3ofh6GqKI9uKSWVDYavcHutb6V4r8t7 w0AMsT2e8qDRd4vbZE+0/p/Coaubu4rhVDuv041Ld6FVLvBuLBnnorWMjyczr+iT 4jHb2+VrGB9i/CrRsvxddz58j3Tsog+GH0PhGJThXnymqf7wwo2mEfajqL00/bAh t5H5Oz4L4SQ= =LYla -END PGP SIGNATURE- -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html