Re: [eug-lug]Looking for a PCMCIA Ethernet card
On Wed, Mar 24, 2004 at 10:20:11AM -0800, Bob Miller wrote: > Anyone have a Linux compatible PCMCIA Ethernet card they'd like to > unload? 10 Mbit/sec is fine. > > Alternately, who in town sells them? Staples does, though probably only Linksys and Netgear. I don't even think they sell them at the UO bookstore these days. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]HELLO!
On Mon, Mar 22, 2004 at 01:24:08PM -0800, happy life skills foundation wrote: > hi again! i tried to look in my setups for the chipset. the whole thing is > a CMOS setup, by Award software. i looked under "chipset features" but > didnt' see a brand name or anything. > > if i need to get a modem what should i get? If you need another modem, it should be the external serial variety. I had one of these, but I think last fall I gave it to someone or other. The external serial port modems are the ones that are essentially guaranteed to work. Non-winmodem ISA modems work, and with the right drivers about half of the PCI modems out there work. USB is hit and miss for modems. PCMCIA usually works, but not always. External serial is the safe bet. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug] Unix/Linux Optimization- Programs&Need for?
On Sat, Mar 20, 2004 at 11:53:47AM -0800, Harald Sundt wrote: > In Unix/Linux is there a need for Optimization and which are the > best programs for this? > > I come from a traditional Mac Background. I run a Mac Household. > There optimizing really helped. The work I've done for clients (I am > a Home Health Aide), using Windoze has shown me that they can do > their optimizing without going to a Boot-Up Disk, unlike a Mac that > can't do optimizing of a drive you boot from. I VAGUELY recall a > Linux maven saying "Unix drive doesn't need optimizing they're swap > disks are so efficient." Windows' "optimizing" involves defragmenting the drive. MacOS X optimizing is simply updating the prebinding so that programs load a bit faster. Prebinding is available for Linux and some flavors do it. Most didn't, although the advantages are big enough I'll expect that to change more and more over time. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Store and Forward
On Fri, Mar 19, 2004 at 05:49:02PM -0800, Cory Petkovsek wrote: > Also, with procmail, I used ! to forward mail: Cory's right and I'm apparently on drugs today. ;) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Store and Forward
On Fri, Mar 19, 2004 at 08:53:02PM +, Bob Crandell wrote: > > Dropping the X-Spam-Status header requires that you do a bunch of pipes. > > > > Just sending the message on would be > > > > :0<--- no need to lock > > * ^X-Spam-Status:.*Yes<--- technically you should have .* not " " > > [EMAIL PROTECTED] <--- isn't that easy? =) > > > > ___ > It looks easy. What's the catch? ;^> > > I'm dense. So > * ^X-Spam-Status: Yes > becomes > * not " " X-Spam-Status:Yes X-Spam-Status: Yes etc Since you know the only program to define something as spam in this manner is spamassasin, you can probably leave it. However, it will only match if that space is there and there's only one. Safer to look for the header and for the word "yes" in it. This is a regex after all, though it is by default a case-insensitive one. > There are no local users so 'other' needs to be a variable that I can attach > to the front of @there.com so it will be delivered to the proper mailbox on > the other server. Does Procmail have that stored? > > [EMAIL PROTECTED] > > BTW, the Milter is no longer incommunicado. It is happily fitering away. I > would prefer doing it with Procmail because it will almost work out of the > box. Why are you trying to forward all spam to users' accounts at another box? ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Internet Annoyances Needed for New Book
On Fri, Mar 19, 2004 at 10:30:33AM -0800, Marsee Heon wrote: > Dear User Group Leader: [.. noting that this is a UNIX group ..] > Pictureless Pages Predicament [.. pages without loaded images, here are "fixes" ..] > * Internet Explorer may be configured so that it doesn't show pictures, ** [..] > * An invalid value in the Windows Registry is preventing pictures from [..] Fleh. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Store and Forward
On Fri, Mar 19, 2004 at 04:42:26PM +, Bob Crandell wrote: > I'm having trouble getting Sendmail to understand I want it to talk to > SpamassMilter. In poking around looking for an answer, I got to thinking that > Procmail is already filtering the way I want. How do I have Procmail forward > this filtered email to another box in the same domain? > > This section sends mail to a folder: > :0: > * ^X-Spam-Status: Yes > $HOME/mail/caughtspam > > Can > $HOME/mail/caughtspam > be changed to: > forwardto:[EMAIL PROTECTED] > > Dropping > * ^X-Spam-Status: Yes > of course. > > What's the correct syntax for: > forwardto:[EMAIL PROTECTED] Dropping the X-Spam-Status header requires that you do a bunch of pipes. Just sending the message on would be :0<--- no need to lock * ^X-Spam-Status:.*Yes<--- technically you should have .* not " " [EMAIL PROTECTED] <--- isn't that easy? =) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [Eug-lug]Notebook LAN
On Thu, Mar 18, 2004 at 11:41:26AM -0800, Bob Miller wrote: > > Although it's not immediately clear to me if it can work w/o cable > > detection support, you might want to checkout ifplugd: > > > > http://www.stud.uni-hamburg.de/users/lennart/projects/ifplugd/ > > Thank you. I just installed ifplugd on my new laptop. It works very > well, so far. The problem with ifplugd was that it tended to keep my HD from spinning down on the gateway. I don't know if this is still a problem or not. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Semi-interesting SPAM articles
On Sat, Mar 13, 2004 at 08:44:47AM -0800, Ken Barber wrote: > > You realize that the reason I stopped signing mail was that NO > > VERSION of MS Outlook can handle any standard method of PGP > > signature correctly, right? > > No, I wasn't aware of that. I'm also a little puzzled by your > news, since it worked perfectly the last time I used Outlook. > > I also know that PGP is working quite well in communucations > between my elderly mother and me (we discuss many things that > John Ashcroft doesn't need to know about). She's using some > version of OE but I don't remember which one. If you don't have PGP installed, OE sees all messages as attachments which cannot be quoted for reply. The signature, being "application/something" is of the type that people fear may contain a virus. Most messages of mine were simply deleted. Those who did not simply delete always felt it necessary to explain to me why my email was broken. Not one of the idiots considered that it was their email client. > But I haven't been a 'doze user for a couple of years now, and > haven't kept up with what M$ is doing. Still, I'm a little > puzzled by your news because it used to work well. And MS hasn't kept up with what the rest of the world was doing or bothered to implement RFC standards... ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Semi-interesting SPAM articles
On Fri, Mar 12, 2004 at 04:23:23PM -0800, Ken Barber wrote: > I'd ask John Sechrest to set up the MX at PEAK to reject all > non-PGP mail coming to me tomorrow... > > ... IF I could only convince enough people out there to adopt this > solution that already has RFCs in place. You realize that the reason I stopped signing mail was that NO VERSION of MS Outlook can handle any standard method of PGP signature correctly, right? ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]SATA Blues
On Fri, Mar 05, 2004 at 06:47:04PM -0800, Ben Barrett wrote: > Bing bing, correct! Knoppix is known for good hardware support. "How did > they manage to do *that*??"... and for those of us who are colorblind: I > ahve no idea what you're talking about, Joseph ( = Knoppix's console can be described only as "angry fruit salad", in terms of its colorings. The CD image is worse than Gentoo even, if that's possible. It's better once you install it to HD. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]SATA Blues
On Fri, Mar 05, 2004 at 12:53:59PM -0800, toman wrote: > > I may have hosed myself. I bought an AOpen AK77-600 Max motherboard which > has a Promise PDC20378 SATA controller and a Via 8237 SATA controller, and > I bought a Western Digital SATA drive to go with it. The BIOS recognizes > the drive fine, but neither my Debian 3.0r1 or Mandrake 9.2 disks see it. > Ideas? I saw indications on the Linux-RAID list that there might be drivers > for the Promise controller, but I couldn't quit figure out which kernel > they were talking about. I'm sorry, but bhahahahahaha! You expect DEBIAN to work with hardware so new? Yeah, sure. Debian doesn't add new hardware support in security revisions, you'll have to wait for a new release version or somehow get a less stable Debian installed. Mandrake I can't abuse for this because I have no idea how they behave, but I will point out that Debian doesn't even support ATA133 devices in stable, even though I reported that they didn't work and offered a URL to a fix a full month before woody was released. That was some two years ago. If you want a Debian that works with current hardware, try Knoppix. The CD has some issues with the creators being obviously colorblind, but beyond that it's really quite nice. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]beagle.j procmail recipe
On Wed, Mar 03, 2004 at 05:31:22PM -0800, Larry Price wrote: > the legitimate one was > a. from mjackson(@)efn(doh-t)org > b. did not include a zip file > c. only sent to accounts that are (not currentlly_paying() && not > logged_in_2004()) Of course, I was being a wiseass. Because efn people still use my efn address when emailing me at times, I am still getting messages from the account. Of course, sometime last fall, I decided that forwarding my mail off efn servers via SMTP so that I could retrieve it without broadcasting one of my passwds all over the campus LAN was probably a wise idea, so I haven't logged in since sometime last year as a result. =) Still get @efn.org emails now and then. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]beagle.j procmail recipe
On Wed, Mar 03, 2004 at 02:24:48PM -0800, Larry Price wrote: > A mass mailing virus that mail's itself out as being from > administrative addresses > claiming to be an account termination notice Wow, that was a virus? I thought it was Mike looking for inactive efn accounts ... ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Greylisting
On Sat, Feb 28, 2004 at 12:45:12PM -0800, Tim Howe wrote: > I can't figure out if I like this idea or not... I think I would try > this for my own mail server, but not a server with thousands of > customers on it. I think it is probably best as a private solution. More and more, actaully, I think the things we consider private solutions will need public deployment. I mean, speakeasy for example refuses to accept any message from a host until it has verified that the host not an open relay, and then continues to check now and then. Sure wish they'd make daily snapshots of their relay list available. I was reminded that the more widespread thing comparable to SAUCE is TMDA. I don't mind an initial mail delay if I don't have to, eg, prove I'm not a spammer within 24 hours or other silliness. This more automated solution is only a few steps from my own suggested replacement for SMTP which works on the spam filtering principle of modern IM clients. That is, your message doesn't get through until I decide to allow it. Of course, my solution is impractical because it involves replacing SMTP which is quite entrenched. When I concluded more than a year ago that it was the only effective solution, I saw another flaw: it would almost require a seperate machine just to manage the database of who ma send mail. I believe Larry mentioned last week that efn now has four machines devoted to spamassasin. Suddenly a server to manage a database not dissimilar to the one used by this greylisting technique becomes a reasonable alternative. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]TV input cards
On Thu, Feb 26, 2004 at 09:48:42AM -0800, Bob Miller wrote: > A third is video capture. Capture the current TV show to a disk file. > In this scenario, audio is needed, A/V quality is important, and A/V > sync is important. A TiVo or another DIRECTV receiver is the source. MPEG decompression from the TiVo then recompression on PC is very costly in terms of quality. If you're not afraid of what you might learn, I can dig out the TiVo hacks book still in my posession and bring it with me next week (much homework tonight) so that you may investigate the notion of directly transferring captured video. I answered truthfully the TiVo survey about this that I haven't done that, but I didn't say that it was only because I've not had time to install the proper software. ;) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]TV input cards
On Thu, Feb 26, 2004 at 09:31:07AM -0800, Ben Barrett wrote: > For some reason, I thought that s-video carried audio as well. I've no > experience there, so "no blood no foul"... That wouldn't make much sense given the evolving standards for both. Aside from the PAL vs. NTSC vs. SECAM issue, video currently comes in these major flavors: - Composite (single RCA plug, usually yellow) - S-Video (mini-DIN connector, seperate chroma and luma) - Component video - YCrCb as 3 RCA plugs - RGB (sync on green) - 3 RCA plugs (some pro gear uses BNC) - VGA connector - RGBHV (seperate sync) - 5 RCA plugs (some pro gear uses BNC) - VGA connector These are from worst to best, acutally. The most flexible is the five signal RGB cable since it it gives equal colorspace to the three primary light colors and offers seperate sync for various resolutions and timings. This is basically VGA/SVGA. Older fixed frequency gear (and anything designed to work with today's televisions is fixed frequency) can get by just fine without any vsync signal and depending only on the hsync which is multiplexed into the green wire. Adapters to use old Sun monitors with normal video cards do this multiplexing, and adapters to use multisync monitors on old Suns extract the hsync and use some intelligence to create a vsync. I digress.. YCrCb is interesting because it applies the technique used by broadcast television. When color TVs came out, B&W had been established and they wanted compatibility. So, they left the B&W signal alone and added the colorburst signal B&W TVs didn't even see. Basically, color TVs treat the B&W signal as brightness and then use a signal for the difference between the brightness and the red, and another the brightness and the blue. What's left once you subtract these out is green. This leaves you with high green resolution but pretty low red/blue. The Y (Luma) signal contains the brightness and sync, the other two contain the color differences. S-Video is what happens when you combine the two color information lines together into one signal and put the result in a single connector. It's not quite as good as component video, but it's still pretty good. Most say it's good enough. Four pins are used--two signal and two ground. No idea why they adopted a single connector for both, but anyone who remembers the old Commadore 64 color monitor has seen S-Video done as two seperately shielded RCA plugs. Someone just decided to make one connector out of it along the way. Composite is the two S-Video signals mixed, giving you one signal. Short of RF modulation, it's the worst thing you can do to the signal. If you haven't guessed yet, Composite happened when someone decided to not bother to modulate and demodulate the TV signal. S-Video happened when someone decided not to mix Luma and Chroma. YCrCb composite happened when someone else decided to output the signal without mixing the Chroma channels together. RGB simply was what all of these things were before we tried to make them sane for both B&W and color. THANKFULLY, Audio is more sane. Sortof. - Analog - Signal level - Line level (what you're used to) - Mic level (weaker, you won't use it, usually needs preamp) - Connector madness - XLR (you won't use this) - RCA (1-6 depending on channels) - Headphone type - Size - 1/4 inch (dying except for mono for musical instruments) - 35 mm (probably only stereo anymore for computers/headphones) - Bizarre sizes and 3+ channels for camcorders and other weirdness - Channels - Mono - Stereo (seperate or combined connector) - Rat's nest of (almost always) RCA plugs for 3, 4, 4.1, 5.1, etc - Digital - Signal conduit - Coaxial copper (RCA plug) - plastic fibre (TOSlink) - Signal type - Stereo PCM - AC3 (basically multi-channel MPEG audio) Currently, non-embarassing AV gear accepts S-Video and stereo RCA. S-Video plus TOSlink patch cables are becoming more common, but I think those won't last because DVD players are making people start to demand component video over S-Video more and more often. TOSlink will win over coax because it's cooler (not because it matters since the signal is already digital) and YCrCb vs. RGB is still up in the air. Both have a strong push behind them. Although DVDs happen to be encoded in the YCrCb format, the playback devices quite often render them to an RGB framebuffer before re-encoding them for display. This is amusingly wasteful, but lossless. ;) > So, what are some reasonably good ways to determine how many milliseconds > out of sync one's audio and video are? When you figure this out, talk to the people making DVDs. Note that the beginning of Fellowship of the Ring was out of sync in the theatrical DVD. Most weren't terribly annoyed and those who don't work with digital AV didn't even notice. ;) ___ EuG-LUG mailing lis
Re: [eug-lug]TV input cards
On Wed, Feb 25, 2004 at 10:15:16PM -0800, Ben Barrett wrote: > For Hauppauge and other BT8x8 (BrookTree, which is now Conexant?) chipset > cards, which are well-supported, actual features vary. I think the 848 > chipsets are full video in and out with tv tuner; the 878 work with a > secondary sound chip, which often has an FM audio tuner. It works well, > although it can be a little noisy when switching channels... but that might > just be the software. > (I'm not sure about the 848/849 and 878/879 differences off-hand, either) Actually the tuner is a seperate component, but whatever. =) The 879s do have extra support for FM tuners if your tuner provides the feature. Basically the bttv chipsets handle sound and NTSC/PAL video signals. What is supported and how varies by implementation and the exact chipset used. Example, I've never seen an 848 card which supported seperate chroma/luma inputs (S-Video), so I am not sure that the bt848 does have seperate inputs. Of course, you could still composite the two signals with a few cents worth of isolating components, but it's not true S-Video in that case. Any card with an S-Video port has had a later chipset which I know has seperate chroma and luma pins. > I'd read that Hauppauge's PVR cards, which do mpeg-2 encoding in hardware, > had some limitation, possibly that they *only* output the mpeg2 stream, as > opposed to writing the video itself through the pci bus. I want to find out > more about this one; the PVR-250 cards are getting a good bit cheaper now. I think they can support framebuffer output, but the drivers for anything else are kinda weak. Don't trust me on that though. > Bob, I think your needs might have been well-covered already, but if I may > digress and over-extend the thread: An old school-buddy had told me how he > got satellite tv on linux going; it was well-enough developed set of tools, > using some older hardware. I'm not sure what cards handle that, but there's > even an HDTV tuner pci card out there, which is possibly the cheapest way to > get HDTV. If the HDTV card is a bttv-class chipset, color me interested. We even have MacOS X support for those things (though it's a bit beta-quality at the moment..) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Greylisting
On Thu, Feb 26, 2004 at 07:23:14AM -0800, Tim Howe wrote: > "The Next Step in the Spam Control War: Greylisting" > > http://projects.puremagic.com/greylisting/ > > OpenBSD's spamd supports this in -current. This is basically just an automated version of SAUCE, Ian Jackson's annoying spam handler. Fortunately, it's automated, which means it's almost sane at least. This is probably the future of spam blocking whether I like it or not. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]TV input cards
On Wed, Feb 25, 2004 at 04:32:33PM +, Bob Crandell wrote: > I'm have my satellite plugged into a Hauppauge WinTV-GO model 190 with nVidia > GeForce 2 MX 400 and Sound Blaster live Platinum and XawTV. I got them at > Circuit City. Sorry Mr. O. > > I don't watch much TV with it. I mostly listen to music. Mostly. The USB WinTV-Go thing isn't so good, but nearly all of the PCI cards work flawlessly in Linux. I think the PVR card might even work pretty well nowadays, but the difference there is that it has its own MPEG-2 encoder which needs extra drivers. I just can't imagine kbob trading his TiVos for a clunky Linux box with a crappy interface. ;) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Request to review Oreilly book
On Sat, Feb 21, 2004 at 11:39:38AM -0800, Jason wrote: > I'd like to request an O'reilly book to review > (apparently they allow LUGs a free copy if reviewed). > It says I should contact the group leader for more > info. > > Is this Bob, Jamie, someone else? The book is ethereal > packet sniffing. Bob does that sort of thing, but if you could save him the trouble of looking up the ISBN it might help. ;) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]procmail help
Bad Cooper.. On Sun, Feb 15, 2004 at 09:19:48AM -0800, Cooper Stevenson wrote: > # Mails with a score of 9 or higher are almost certainly spam (with > 0.05% > # false positives according to rules/STATISTICS.txt). Let's put them in > a > # different mbox. (This one is optional.) > :0 > * ^X-Spam-Level: \*\*\*\*\*\*\*\*\* > /var/spool/mail/almost-certainly-spam > --End Recipee- > > Hope this helps! You need :0: there, anything in /var/spool/mail is an mbox, and mbox requires dotlocking. The results of not doing so may not seem obvious now, but someday it'll blow up on you. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]autolearn=yes?
On Mon, Feb 16, 2004 at 05:39:33PM -0800, john fleming wrote: > Ham? Good point, it's not a very good name. Tofu would probably be more correct, given that tofu is 100% the opposite of spam. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]autolearn=yes?
On Tue, Feb 17, 2004 at 12:58:06AM +, Bob Crandell wrote: [spam headers] > There are 2 issues here I would like help with. > 1) The Subject line has "**SPAM*:". This is from a server that is upstream. > It tags messages before it passes them on. I have a procmail receipe on my > box that says: > :0: > * ^Subject:.**SPAM*: > /dev/null First, don't lock /dev/null. =) Second, try this: :0 * ^Subject:.*\*\*SPAM\*: /dev/null > This happens before my SpammAssassin gets ahold of it. As you can see, my > SpamAssassin is still getting these messages. What do I do to fix it? Yes, the upstream server is tagging rather than deleting. That's probably a good thing unless you're on a modem and paying by the minute. Delete it when you get it, before you run SpamAssasin on it if you like. The above procmail rule will do it--note that * is special, we're talking regex here. > 2) You will notice autolearn=no on the last line. How do I make it say > autolearn=yes ? BTW, the score doesn't make any difference here. That's a spamassasin config issue. As I don't use it directly, I can't help there. =) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]procmail help
On Sun, Feb 15, 2004 at 10:02:52AM -0800, Ben Barrett wrote: > | So I'd have to do something similar to a switch in procmail? > | X-Spam-Score: * > | X-Spam-Score: ** > | X-Spam-Score: *** > | .mail/maybespam > | > | Can procmail do something like this? > > Yes, and what will it do with *this* message, which clearly states > X-Spam-Score: * (ie, confusing but possibly through the roof) > > Sorry, just the Sunday morning heckles ;^> :0 * ^X-Spam-Level:.*\*\*\*\*\*\* /dev/null :0 * ^X-Spam-Level:.*\*\*\*\* Spam/ Note that I am not locking on the second rule (:0:) because I use maildir. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]procmail help
On Sun, Feb 15, 2004 at 09:00:00AM -0800, Rob Hudson wrote: > Hello, > > As I get inundated with spam, I'm changing my server settings a little > so I have less to deal with. I was getting a couple megabytes of spam > in my spam folder every 2-3 days. After reading up on Spamassassin, I'd > like to do something like the following in procmail: > > Spam score of 10 or more: send to /dev/null > Spam score of >= 4 and < 10: Drop in my spam folder for analysis > Spam score of >= 1 and < 4 : Drop in might be spam folder Throw away anything 7 or above. Just trust me. 5 would be a reasonably safe bet, but nothing that isn't blatant spam makes 7. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Not Linux Related, Assembly Related
On Sun, Feb 15, 2004 at 06:25:48AM -0800, nyal wrote: > Fits on a floppy, why not try it! You have a floppy drive? Five years ago, Apple said that the floppy was dead technology and that people would be burning CDs instead, even for small files, because the CDs were cheaper than the floppies already. People laughed at Apple. Most new PCs today no longer feature floppy drives unless you add it as an extra feature--a feature which is steadily becoming more expensive and harder to find quality media in quantity. Amusing. =) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Navigator install error
On Tue, Feb 10, 2004 at 11:46:22AM -0800, Darren Hayes wrote: > > Yech, why don't you just use firebird and tell it > > what to set the User-Agent string to. > > You mean Firefox ;-) > > http://news.com.com/2100-7344_3-5156101.html I can already hear the filk of a U2 song about browsers with no name.. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]near Eugene DSL questions
On Mon, Feb 09, 2004 at 02:24:55PM -0800, Ben Barrett wrote: > I thought qworst's line charge was ~$30/month, then you add on ISP costs. > > ...and being that the lowest DSL ISP monthlies I've seen were $20/month, I > had added up that cable internet is cheaper... but I'm guessing that you're > too far our, in pleasant hill, for that... Yes, it's cheaper. And the downlink is so much faster than DSL that it's barely in the same league. After a recent upgrade, Comcast users are reporting up to 3mbps downlinks for a price right between Qwest's 256/256 and 640/256 prices through efn. The uplink speed is now topping out around 225kbps for these upgraded areas, which is too bad. Somewhere the idea was floated not long ago of Comcast dropping to 128 uplinks to give the file sharers some pain, but that hasn't happened and was discussed long enough ago that I hope it may not. The situation is that you don't know for sure, and frankly I wouldn't dare do anything that someone might even think possibly bears any resemblence to what might be illegal or possibly just unfavorable to large media interests on Comcast. Their AUP is fscary--they can kill your account for any reason at just about any time. As far as I know, efn's AUP still does not contain anything of the sort, though former management did seek such a change. I don't watch things at efn closely enough anymore to know if the proposal left with the manager. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]near Eugene DSL questions
On Mon, Feb 09, 2004 at 02:42:39PM -0800, Patrick R. Wade wrote: > My fellow worker Jay recently got QWest DSL via EFN, and got the Actiontec > device; he's been pretty happy with it. It can do what you're describing, > simple firewalling and routing, and has slots to add PCCard wireless cards > as well. Too bad it only works with Actiontec 802.11b cards. They cost more than several 802.11g routers do at this point. The firewall feature is generally no crappier than I have come to expect from most network appliance router/firewalls, but the average free UNIX variant provides a lot more power and control. You'll find that anything which has a simple Windows-common workaround to firewall issues (regardless of how pathetic) is probably going to expect that you use them. I don't recall even that non-passive FTP was supported. I'm picking nits obviously, but it was once possible to turn off the Actiontec firewall and do the job yourself so it can be done right. The other concern is that Qwest waited 9 months to tell people about a security flaw with the Actiontec because of a bug in their software which made the new/fixed firmware incompatible. > With my own ReachDSL setup, i have the endpoint feeding a little NetGear > NATbox that also does wireless. This apparently isn't possible anymore with Qwest DSL. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]help w/ weird mouse configuration
On Mon, Feb 09, 2004 at 11:30:51AM -0800, Bob Miller wrote: > The right mouse for me, now that all our old SGI Indy mice have died, > is a Logitech 3-button. But I'm the only one who has to use it > (except for Anne, who insists on using my 2.4 GHz Gentoo box instead > of her own 450 MHz Windows 98 thing (-: (and yes, I've offered to > buy her a new PC)). Clearly Anne was married for her intelligence. =D ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]help w/ weird mouse configuration
On Mon, Feb 09, 2004 at 01:18:58PM -0600, Christopher Forsythe wrote: > >You mean with only one button? Are you serious? > >(Are you using this under linux, or OS X?) > > Using under os x, but it can be configured for linux. It makes sense, > because my hand is always on the keyboard, and if I want a contextual > menu, I can easily hit ctrl. It makes some ergonomic sense. Plus I do > not see a reason to buy a mouse when this one works perfectly fine. I do, but I like my higher DPI, my dual LED tracking, and my scrolley. AND my right button, dammit. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]help w/ weird mouse configuration
On Mon, Feb 09, 2004 at 12:48:31PM -0600, Christopher Forsythe wrote: > >Apple's acrylic > >disk mouse is the poster child... > > > HEY! I love my apple mouse... It is one of the more usable mice that I > have. I would rather use that than some other ps/2 mice out there. I think he's referring to the hockey puck. Nobody likes that. The buttonless one is a different story, though I still prefer the scrollywheel 3 button variant made by somebodyorother. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]help w/ weird mouse configuration
On Sun, Feb 08, 2004 at 10:51:11PM -0800, Brian Baker wrote: > One other note on this. If you are running Kudzu, it may detect your usb mouse > during boot up and want to configure it for you. > > Don't let it!! > > It will remove your ps2 mouse and you will be stuck with an X windows that > will only work if you have the usb mouse plugged in. Just tell it to ignore > the new hardware. > > I've lost count of how many time I've made that mistake. The joys of 2.6... That problem goes away. All mice are /dev/input/mice. And um, why would you ever NOT use a USB mouse? Laptop maybe? ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]euglug.org ranked # 40 in Google for linux training window
On Thu, Feb 05, 2004 at 09:38:16AM -0800, Brad Davidson wrote: > This message tripped my mental spam filter. Come on, she obviously > didn't find EUGLUG on google. The keywords make no sense, they look like > 3 random words. She also obviously doesn't know what EUGLUG does, how > would she 'compete' with a mailing list/user group? Ahh, but she did. The three random words do give euglug as a result. Indeed it is #40. This person or persons are involved with a google poisoning attempt. Look up anything to do with any of the top search terms and you'll be swamped with links to sites like freeringtones.com (which isn't free and has nothing to do with whatever you were probably looking for...) These people look for top n results for a given grouping of common search words (like "linux" and "training" for example) and try to get themselves linked. If they can fool ten sites out of the top 200, they have a good chance of showing up in the top results for a given search related to one of those words, and a higher chance of showing up in a search for both. That's how these slimeballs operate. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
[the_tick@brok3n.org: Re: [benb@nu-world.com: Re: [eug-lug]euglug.org ranked # 40 in Google for linux training window]]
: 1, ADDITIONAL: 0 ;; QUESTION SECTION: ;elizabethrichson.com. IN MX ;; AUTHORITY SECTION: elizabethrichson.com. 10800 IN SOA ns.elizabethrichson.com. root.ns.elizabethrichson.com. 2001100102 28800 7200 604800 86400 ;; Query time: 353 msec ;; SERVER: 211.152.14.68#53(elizabethrichson.com) ;; WHEN: Wed Feb 4 23:27:35 2004 ;; MSG SIZE rcvd: 82 kermit:~ chris$ end dig stuff/ /***Port scan info [EMAIL PROTECTED]:~$ nmap -P0 elizabethrichson.com Starting nmap 3.48 ( http://www.insecure.org/nmap/ ) at 2004-02-04 23:35 CST Interesting ports on echeckservice.com (211.152.14.68): (The 1646 ports scanned but not shown below are in state: closed) PORT STATE SERVICE 25/tcp open smtp 110/tcp open pop-3 135/tcp open msrpc 139/tcp open netbios-ssn 445/tcp open microsoft-ds 1025/tcp open NFS-or-IIS 1026/tcp open LSA-or-nterm 1027/tcp open IIS 3372/tcp open msdtc 3389/tcp open ms-term-serv 4899/tcp open radmin Nmap run completed -- 1 IP address (1 host up) scanned in 261.092 seconds [EMAIL PROTECTED]:~$ ***End port scan info***/ The only thing ev1servers has to do with this is that it is the registrar. Obviously the other contact info is not legit either. You can report domains on that, but I fail to remember where. And if it is legit, well.. you guys can figure out what to do to someone who spams a linux user groups email addy :D. Christopher Forsythe On Feb 4, 2004, at 10:11 AM, T. Joseph Carter wrote: Didn't you say your domain was hosted by ev1? From: Ben Barrett <[EMAIL PROTECTED]> Date: February 4, 2004 2:29:57 AM CST To: "The Eugene Unix and GNU/Linux User Group's mail list" <[EMAIL PROTECTED]> Subject: Re: [eug-lug]euglug.org ranked # 40 in Google for linux training window Reply-To: "The Eugene Unix and GNU/Linux User Group's mail list" <[EMAIL PROTECTED]> Ya! Egg on face. This has been going on for a while; google on her name gave these examples of similar situations: old: http://www.rcuniverse.com/forum/ Weird_spam_RCU_guys_with_website_should_look_out_for%25/m_1449911/ tm.htm (RC hobbyists? ~2 years ago!) just recent: http://www.rb21.com/news/index.php/t/31489/0/ (tiki dev) http://sourceforge.net/mailarchive/forum.php? thread_id=3768760&forum_id=1471 (sax users) The whois record shows this: Registrant: apple 2198 Apple dr. Columbus, OH 43212 US Domain name: ELIZABETHRICHSON.COM Administrative Contact: Johnson, Muliya [EMAIL PROTECTED] 2198 Apple dr. Columbus, OH 43212 US 410-678-8768 Technical Contact: Customer Service, EV1 Servers [EMAIL PROTECTED] 2600 SW Freeway Suite 500 Houston, Texas 77098 US +1.717873Fax: +1.7139429332 isn't there some law...? (d'oh) but HEY, we're still cool. Yeah, we rock! (d'oh) On Tue, 3 Feb 2004 18:45:28 -0800 Ben Barrett <[EMAIL PROTECTED]> wrote: | Cool news, but time for some sort of consensus here. I'd like to know if | she's running her site for-profit, or what ?? I don't want to see our | group become link-mavens for sites that get advert-based funding or | otherwise are commercial. Does she know about LUGE, or the other LUG | directory? Cool we R0X0R !!! #40 AO AO $) 40! | | congrats, people | | | On Tue, 3 Feb 2004 15:46:45 -0800 (PST) | Elizabeth Richson <[EMAIL PROTECTED]> wrote: | | | I'm a web master, and I was just searching Google for linux training | | window. I found your domain, euglug.org ranked 40, which is pretty cool. | | | | | | My site is all about Computers - Software, too . Maybe we should link | | up? I wouldn't be stealing any of your sales, because all I do is write | | informational articles...not selling anything on my site at all. And | | most of my visitors write back to say that they love the fact that I | | only write good, quality info. As a matter of fact, I've got a pretty | | loyal following of people that come back over and over again (they use | | the site as a reference), so if you link to me, you should get some | | pretty good traffic from it -- which is always nice. | | | | Anyway, let me know if you'd like to swap links. I've already linked to | | you, and will keep it up there for a few days until I hear back. Hope to | | hear from you soon! | | | | Elizabeth Richson | | RAC IM: 391574. | | | | ___ | | EuG-LUG mailing list | | [EMAIL PROTECTED] | | http://mailman.efn.org/cgi-bin/listinfo/eug-lug | | | -- | ___ | EuG-LUG mailing list | [EMAIL PROTECTED] | http://mailman.efn.org/cgi-bin/listinfo/eug-lug -- ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug --- End Message --- ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Naive BSD question
On Tue, Feb 03, 2004 at 07:43:58PM -0801, Jacob Meuser wrote: > > Yes, that's what he meant. > > > > Now what would be cool is a journaled file system for bsd. Any one out > > there? Perhaps when bsd supports xfs, reiser or ext3. > > Probably won't ever happen. UFS/FFS has been around for ages. Many > BSD'ers take comfort in that. Look at the FFS in FreeBSD 5.x for > state-of-the-art BSD filesystems. There's already a form of journaling present in BSD. It's not quite the same as the stuff used by the Linux offerings, but it does log the changes to assist when you need to fsck. I'm sketchy on details. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]euglug.org ranked # 40 in Google for linux training window
CLUE: This is spam. On Tue, Feb 03, 2004 at 03:46:45PM -0800, Elizabeth Richson wrote: > I'm a web master, and I was just searching Google for linux training > window. I found your domain, euglug.org ranked 40, which is pretty cool. > Hello, category match alert! vvv > My site is all about Computers - Software, too . Maybe we should link "My site is all about Computers - Hardware..." "My site is all about Beer - Imported..." "My site is all about Toys - Toddlers..." (Hey, a site about WinXP!) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Re: EUGLUG FreeBSD users?
On Sat, Jan 31, 2004 at 12:45:34PM -0800, Larry Price wrote: > (Metacity is billed as "the window manager for the adult in you") Translation: It has no features that Havoc doesn't think you need. If any current feature seems useful but not absolutely necessary, it will probably be taken out soon. ;) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]a few questions
On Thu, Jan 29, 2004 at 10:02:06PM -0800, Bob Miller wrote: > Video comes in many formats. I doubt a VIA CPU can handle realtime > decompression of most of those formats. And future codecs will be > more compute intensive. If you're building a single-purpose box that > only uses MPEG compression, and you can get the hardware MPEG decoder > to fly, the VIA boards might be suitable. I'd disagree to some extent. An intel chip of the 266 MHz starting point can usually handle MPEG-2 decompression in real-time without dropping frames if that's all it's doing. A 366 MHz processor in a laptop can do MPEG in Linux at least without hardware help, though in some other evil OS there is help from the hardware on this particular notebook. A VIA chip around 900MHz or so should be more than adequate for MPEG video. It should do DivX fine as well. More complex formats (which nobody is seriously using at this point in time) are out, but unless there is a major compression breakthrough, DivX video is currently about as good as it gets for quality and compression ratio. Of course, with hardware decoders, a dinky 50 MHz processor is almost fast enough to be useful for some practical purpose. Like presenting a simple GUI to the user, perhaps. But I wouldn't expect you'd know much about that, right? =) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Help the out-of-towner
On Thu, Jan 29, 2004 at 01:40:50PM -0800, Ben Barrett wrote: > Can also suggest Graybar for cables and networking tools (ends/crimpers, > etc), although they claim to be wholesalers, I think they'd happily take > your money. I've found them to be more helpful and also cheaper on a lot of > things, than Norvac. Graybar and Norvac are both out west 11th, Graybar is > past Fred Meyers. Also out that way is VOS, Virtual Office Systems, which > some here may poo-poo, but they've been providing many stable PCs to my crew > lately, and are pretty responsive to problems and returns and the like, when > they do arise. The only complaint *I* have with them is that they seem to > like to zip-tie the heck out of the cabling, which can pose problems for IDE > cables fortunately it hasn't yet! VOS sells the PC systems through the > UO bookstore, but is located out west 11th, between Graybar and Norvac IIRC, > in the same little shopping center as a Radio Shack, near the DMV. It's probably okay to zip-tie and IDE cable, really. I used to do an awful lot of that myself just for the sake of trying to get airflow around components. Not everyone can have the airflow design of a Mac. ;) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Microsoft and Fun With Trademarks
On Mon, Jan 19, 2004 at 08:58:24PM -0800, Patrick R. Wade wrote: > http://www.theregister.com/content/6/34955.html > > Now, one wonders, will they come after MikeOSoft.com? ICANN accused of secret dealings, double standards, and other misconduct. Film at 11. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Combo Message - Comcast DNS and Brewpub/EUGLUG Meeting
On Wed, Jan 14, 2004 at 04:31:43PM -0800, Hal Pomeranz wrote: > > Secondly - are normal meeting folks going to the SAO > > brewpub meeting instead this Thursday? It sounds > > interesting and I wouldn't mind checking it out. Do > > other EUGLUG'ers attend these events? > > I generally plan on going to the SAO events, and I ran into Ben at the > last one. So yeah, there's some of the EUGLUG Underground > infiltrating SAO... We infiltrate anything that involves the possibility of pizza, beer, or software. Interest rises if any of the afore-mentioned is likely to be free. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]open source pascal
On Sun, Jan 11, 2004 at 10:47:18PM -0800, Neil Parker wrote: > >Object Pascal is probably the target to aim for, and a compiler that > >supports Object Pascal is (or at least was) necessarily a superset of > >Borland Pascal. Neil can comment about how much Delphi has evolved since > >the 16 and early 32 bit days I hope. > > Not as much as you might think. I've only been seriously exposed to three > points along the Borland Pascal evolutionary continuum: Turbo Pascal 3.0, > Delphi 4, and Delphi 6. Though Delphi 4 is clearly a descendent of Turbo > Pascal, so much has been added that it's practically a whole new language. > Lots of new classes were added between Delphi 4 and 6, but the language > itself doesn't seem to have changed in that interval. I generally meant between versions of Delphi. I know the evolution of Pascal through Delphi 2. > Delphi's support for classes feels like it's been ripped right out of > Java. I consider this a plus: Java's class model always struck me as > much cleaner than the complex and unwieldy C++ class model. The > predefined classes that ship with Delphi, however, don't bear much > resemblence to Java's class library...you get lots and lots of support for > Windows API objects and COM and Active X and web services and the like, > but not much in the way of abstract data type classes (the latter is > sometimes annoying--I have occasionally wished Delphi came with something > like a Perl %hash or a Python dictionary (well...TStringList can be > shoehorned into service, but it's just not the same)). TStringList is definitely not the same. =) It probably wouldn't be hard to write one, but like C++ you'd need to write objects for all of your primative data types which do not have one already conveniently usable. This is annoying about Java as well, but they at least thought about it in the design of the language. > Yes, Pascal supports pointers. But you don't use them as much as you do > in C--Pascal's ability to pass parameters by reference eliminates one of > the major uses of pointers in C. Delphi implements classes with pointers > just like Java does, but (also like Java) it works very hard to prevent > you from needing to know that. Pascal's pointer syntax is a little confusing for newbies. "Was it ^foo.bar, foo^.bar, or foo.^bar I wanted?" The usual advice I have seen newbies who have managed to get them to work (but still don't understand them) is "don't use any ^s at all, then try to compile it. Put one where it complains you're missing one." I went on a long explanation which involved borrowing a set of those brightly colored plastic things from a kid for a day. You know, the ones that have aa connector on one end and a hole on the other to form a chain. Less than half my audience got it. The other half followed that other piece of advice. *shudder* Pass by reference is one of the major saving graces of C++ if you ask me. Most programmers tend to use a lot of globals in C because it's easier than dealing with C's passing mechanism. Bleh. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]open source pascal
On Sun, Jan 11, 2004 at 04:36:03PM -0800, john fleming wrote: > Joseph I vaguely remember a conversation that you ,neil and I had about > problems with the classes in C++ . It had something to do with the > impossibility of new classes of certain types, like C++ has boxed itself > into a corner ? Do you remember what that was about? Yes. > John > PS. If so please elaborate a bit instead of answering yes to a long > question like neil likes to do, thnx Oh, damn. Basically, there hasn't really been any coherent agreed-upon standard for C++ ABIs until fairly recently. That's ABI, the Application Binary Interface, the thing which allows you to link to a library and read its function table. Short explanation.. In C, a function in an object file is referenced by its symbol name. If the function's name is do_something, its symbol name may be _do_something, for example. (In fact, that's what it should be for C.) C++ is a little more complex because a function named doSomething would not be named _doSomething in the object file because there can be multiple functions with the same name in C++. Some people decry this as one of the things about C++ that is evil, but I for one think it's actually a good idea. You can doSomething(i) or doSomething(str) as long as you have a version of doSomething that can take an int and another that can take a string. It's possible these functions could do different things, but that is just bad coding style. The thing is, if they have the same name, you can't have the same symbol name. The only way to tell them apart is by what parameters they take (and also what namespace they're in, but we're assuming the simple case that they're both global..) Let's say, since I don't know C++ name mangling, that they are stored as _doSomething_Int and _doSomething_Charptr. As long you're consistent with this pattern of mangling the symbol name, everything works fine. I just said that there was no widely accepted standard, didn't I? gcc 2.7 had one way. egcs had at least one other. gcc 2.8 used that, but I think gcc 2.9 may have changed that again (I'm not sure..) What I am sure of is that gcc 3.0, 3.1, and 3.2 have all used completely different and incompatible symbol name mangling for C++. Each time you change this, every single C++ thing on the system MUST be recompiled or the linker will break when it tries to look up functions according to the new mangling scheme. What gcc 3.2 used, if it was not somehow broken, is the now agreed upon standard. 3.3 should work with it, and now that gcc adheres to a standard, I expect they won't break the ABI again for no good reason. Now the language, on the other hand.. Let's put it this way, the C++ focus groups and committees are so anal that they feel it is their duty to engineer and re-engineer C++ so that it can "remain competitive" while correcting for "bad programming practices". ie, things that used to work tend to suddenly stop working. gcc does its best to remain current with the latast standard, and the gcc C++ people are about as anal as the committee. It is not uncommon to write some code, upgrade your compiler, and find that things which were not even warnings previously are now errors. Additionally, a whole bunch of silly things are now glaring multiline warnings telling you that the next gcc will disallow these too. So nowadays, the real problem with C++ is not in the ABI, it's in the strict adherence to the C++ standard (which is basically a moving target.) While it's fair to say that, for example, C99 is more recent than the latest C++ standard, C99's designers refused to consider any ammendment which made code which compiled under previous standards suddenly not. Even Java is such that code written for the original release still runs on the latest JVM, though Sun did do a bad when they made 1.4.1 refuse to bytecompile code that was technically an error, but silently ignored in 1.4.. 1.4.1 should have warned and 1.4.2 should have refused. That's just my opinion though--at least the stuff that broke was already an error, even if a silently ignored one. Ask a C++ old-timer about the introduction of namespaces and how little code was broken by that change. Better yet, make it a Linux C++ old-timer because it's much more likely that you'll ask someone using MSVC and they'll tell you that there was no problem: If the code doesn't use namespaces at all, it compiles. If it does, then the specified rules for namespaces are enforced. Old code doesn't break, but new code follows the new standard. With gcc, it was a forced march to the new way of doing things. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]open source pascal
On Sun, Jan 11, 2004 at 09:25:23AM -0800, john fleming wrote: > I read a n article about relative performance between languages, though > pascal was not one of the nine > the article said that it should be included in future tests. So I did a > little more research on pascal and found out it does pointers and is > faster to program with than C , It seemed that that might be a good > second languuge after python ,but it doesn't seem to be used that much. Pascal has an annoyance or two.. There are four basic variants: ISO Pascal (useless), Borland Pascal (which has been somewhat evolutionary over the years, but whose last version can be considered the authoritative implementation at this point), GNU Pascal (which has a ton of similar but not directly compatible extensions to ISO, and Borland Object Pascal. Object Pascal is probably the target to aim for, and a compiler that supports Object Pascal is (or at least was) necessarily a superset of Borland Pascal. Neil can comment about how much Delphi has evolved since the 16 and early 32 bit days I hope. FPC (I guess they've dropped the C now and it's just FreePascal?) was originally written in Borland Pascal 7, at least until it was complete enough to compile itself. The problem is that of bootstrapping. Anyway, when I last looked at FPC, it had a couple of minor extensions over Borland syntax, and I'm not sure this is a good idea. They have added support for Object Pascal and it is even possible to recompile Delphi components for it on Windows. My complaint with it does not produce standard .o files for the UNIX systems to which it is ported. It has maintained a Pascal-convention stack, and is basically interfacable to C code in a complex manner. No attempt has been made to make, eg, a clone of the Win32 Delphi controls which happens to be cross-platform by virtue of wx or gtk+ or whatever. Why this mattered in comparison to C is that because FPC did not use native object format and calling convention, there was more work to do in order to port something. Much may have changed since I last played with it, though. By the time I got it to work the way I wanted, I knew enough C to port all of my old code directly. I admit that I kinda miss the Pascal streams, but they were basically weird special-case stuff. Also, at the time there was only ia32 support. Let's face it, x86 is dead. The world is moving to 64 bit chips and if FPC has not been ported to at least 32 bit PPC and x86-64 by now, it isn't likely to be useful for very long. Again, some or all of these concerns may have been resolved. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]open source pascal
On Sat, Jan 10, 2004 at 08:59:07PM -0800, Bob Miller wrote: > > Is there such a thing as open source pascal with oop features? > > GNU Pascal Compiler, gpc. http://gnu-pascal.de/ > > Free Pascal, fp. http://www.freepascal.org/ fpc is a little annoying to set up, but it's much better than gpc last I checked. Also, it helps that fpc was written to be 100% Object Pascal (eg, Delphi, Kylix) compatible, language-wise. Different units though. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]unbelievable sound
On Sat, Jan 10, 2004 at 12:07:02PM -0800, Mr O wrote: > Where does ALSA mixer store it's settings? PCM likes to be muted > on startup which isn't helpful. If ALSA is configured properly, shutting it down and restarting it should save and reload your mixer settings. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
[eug-lug]Hmm...
0403 1207 1729 2701 7663 As I have a boatload of work to do tomorrow, I probably will not be wandering downtown tomorrow night for the clinic. If I cannot go, the least I can do is send sagely and cryptic (but ultimately simple and obvious) numbers in my place. They should serve as a reasonable facsimilie. I bet they even have a similar effect scaring newbies. =) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]New Debian PPC user looking for SCSI drive
On Tue, Jan 06, 2004 at 07:06:24PM -0800, Max Lemieux wrote: > I'd also love to hear from anyone else who's running a *nix variant on > Apple hardware, as there are still a couple of glitches I need to work > out (Kensington trackball, etc) ;) Me. But I suspect you were hoping for a non-Apple brand of *nix? ;) The trackball would be nicer if it were USB simply because there'd be no confusion as to how to make use of the buttons. If you don't post that you have found the solution to the trackball setup by this weekend, I'll investigate mutli-button ADB pointing devices. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]MVLUG Talk Postponed
On Tue, Jan 06, 2004 at 10:16:51AM -0800, John Sechrest wrote: > The MVLUG talk by Hal Palmeranz has been canceled tonight. > OSU is closed and we are unable to travel safely to the > meeting. I wouldn't describe the walk to the driveway as "safe" in much of the valley today. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]PDA not working
On Sun, Jan 04, 2004 at 02:19:19PM -0500, Jamie wrote: > : will tell you if you are actually attempting to mount the correct > : device. > There are also /dev/sr* for scisi devices, my cdrw uses /dev/sr0 sd == scsi disk sr == scsi optical drive (cd/dvd/etc) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]superformat /dev/fd0
On Sun, Jan 04, 2004 at 09:45:45AM -0500, Jamie wrote: > Nyal, > the issue is you want to unmount it before formatting it. Since you used > konsole to check it, supermount had mounted it, and it was in use. You want > to unmount your drive, and first it has to not be in use to unmount it. So.. > cd to another location, and make sure nothing is using it (such as > konqueoror.) Right. Rarely do you need to actually do a full format of the device. The best tool for formatting it for dos is mformat, which takes parameters very similar to the dos format command. (I don't remember what they are..) Also, you might want to check the mke2fs manpage--some of the options to mke2fs which are more advanced are probably wise when formatting a flopy because they will reduce the space taken up as overhead. (more overhead is usually used with ext2 than FAT16, which has more than FAT12.) Once again, I have no memory offhand of what parameters you might want to use exactly--I'm just suggesting that you look if you're interested. ;) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]New year's resolution...
On Sat, Jan 03, 2004 at 03:19:44PM -0800, Bob Miller wrote: > OTOH, the 1920x1200 24" displays (WUXGA, I think) look awfully nice. > They're cheaper than a big screen TV, and they have a much higher > resolution than HDTV. It'd work well as a TV if you can sit within > ten or fifteen feet of it, and it'd work extremely well as a monitor. > For the money you saved on the plasma screen, you can add a low-noise, > high performance PC and a TV tuner card. Running Linux w/ > xine/mplayer and MythTV/Freevo, of course. I'd hold out for 2160 horizontal resolution since the scaling is perfect 2:1 for 1080i HDTV if you're looking at the 24" display as a TV. Lower resolutions are reasonably nicely interpolated if there any attempt to do so correctly simply because anything over 2:1 can be made to work without getting too blurry. Brightness may be a factor, but I expect that a 24" display is going to be backlit similar to Apple's 20 and 23" models--much brighter than any laptop you've ever seen. Ever. A factor for me is that the gamma ramp is different than it should be (TVs tend to be brighter and have color temperatures which approach blue.) Another is that interpolation tends to cause some things to be darker because they are small and lose some of their sharp edges if the interpolation isn't done right (it's not in this instance..) Getting it right is doable if they hardware knows how, but I'm using software which makes some assumptions. It's not been enough of a problem to warrant special efforts to correct--I just raise the brightness to full when watching TV. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]I'm back!
On Sat, Jan 03, 2004 at 10:37:48AM -0500, Linux Rocks! wrote: > : What??? I wasn't an Oregonian before, just because I didn't have a > : chainsaw? Had I known, I would have bought one in Ashland on the way > : in. > > If you dont have a chainsaw on your keyring, next to your bike lock key, and > your earth first key tag, then your just not an oregonian. sorry... Almost nobody outside of Eugene must be an Oregonian then. ;) > Seriously thougth... weird things like one calamaty after another is rather > normal in oregon... Im just surprised you didnt have any events when blasting > the tree peices out of your driveway! We have had some rather interesting weather of some sort or another every year in Oregon as far back as I can remember. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
[eug-lug]New year's resolution...
I was thinking hard about new years resolutions this week and have settled on one. My new year's resolution is 1680x1050. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]QMail Folders
On Mon, Dec 29, 2003 at 06:00:38PM +, Bob Crandell wrote:
> >...but I don't generally like it. I don't remember the .qmail filtering
> >syntax, but I remember that it wasn't as good as exim's and nowhere near
> >as good as procmail or similar. If your qmail setup doesn't already use
> >procmail if available |procmail with whatever args are appropriate is just
> >a line away in your .qmail file.. ;)
> >
> This is supposed to be system wide. What does your /etc/procmailrc file
> look like to create folders?
You're in luck! My .procmailrc and associated files are designed to be
self-teaching for Debian developers, as once my .muttrc was. Although the
way I use mutt has changed too greatly for my .muttrc to serve to help
pine users migrate (its original purpose), my .procmailrc is still largely
useful by people who need to filter their mail.
Please not the differences between what's "normal" and how I've set it up
here (to fit more in to the way NeXT and MacOS X do things):
$HOME/Mail -> $HOME/Library/Mail/Folders
$HOME/.procmail -> $HOME/Library/Mail/Procmail
$PMDIR/log -> "$PMDIR/Mail Log.txt"
You probably want to change these back. ;)
Another weirdness that predates MacOS X is the location of my primary
inbox. I don't have /var/mail/knghtbrd or ~/Mailbox or ~/Maildir or
anything of the sort. I have $MAILDIR/Inbox, which is honestly a hell of
a lot more logical. It was the one thing pine ever did right - your inbox
belongs with the rest of your folders. (In pine's case, it was only
virtually so, but I can see no reason why not to make it actually so.
Obviously, don't do this if you use pine because it gets confusing..)
Anyway, here are my .procmailrc and my rc.maillists:
# Knghtbrd's .procmailrc
#
# Long, long ago on an ISP far, far away there was this newsgroup called
# teleport.support.unix. In this newsgroup were many helpful people who
# taught me everything I ever needed to know about procmail. They rocked!
#
# Most of what they taught me was kinda teleport-specific - you can't just
# filter on the envelope the way I do in my spam filter unless you feed the
# right stuff to procmail. I'll mark any rules that require custom stuff
# and what custom stuff you need..
# Set on when debugging
VERBOSE=off
SHELL=/bin/bash
# Replace with your mail directory (Pine uses mail, Elm and mutt use Mail)
MAILDIR=$HOME/Library/Mail/Folders
# Directory for storing procmail log and rc files
# It would be "normal" to use $HOME/.procmail for this...
PMDIR=$HOME/Library/Mail/Procmail
# And $PMDIR/log is often used here..
LOGFILE="$PMDIR/Mail log.txt"
# ---custom stuff---
# IFF your copy of procmail is built with the Maildir patch (it should be!)
# and you want to use it to put mail into ~/Maildir, uncomment the next
# line. The trailing / is what tells Procmail to write to a maildir.
#DEFAULT=$HOME/Maildir/
# Or you can put all of your mail in mbox format, but in $HOME instead.
#DEFAULT=$HOME/mailbox
# Personally, I like all of my mail in $MAILDIR..
DEFAULT=Inbox/
# This just saves stuff to $MAILDIR/test if it has test in the subject
#INCLUDERC=$PMDIR/rc.testing
# The Mail Filtering FAQ chops .procmailrc up into pieces based on what
# sort of recipe is in the file. Not a bad idea really..
INCLUDERC=$PMDIR/rc.spam
INCLUDERC=$PMDIR/rc.maillists
INCLUDERC=$PMDIR/rc.idiot
INCLUDERC=$PMDIR/rc.special
INCLUDERC=$PMDIR/rc.dupe
Here's my rc.maillists file:
# Sorts the mailing lists
# SF, icculus, etc all have group- mailing lists. Let's match on an
# entire group's lists all at once! Note that we deliver with :0 and not
# :0: below. Don't do that with mbox. Maildir needs no locking.
:0
* ^Mailing-List:.*twilight-.*icculus\.org
* ^Mailing-List:.*twilight-\/[-A-Za-z0-9]*
"List:Twilight/"
# And actually, the above rule reflects a change in my mail viewing
# habits. Originally, the rule was intended to deliver each of a group's
# lists to a seperate folder. Note the \/ in the expression, that's
# procmail's own signal that what follows should be stored in the variable
# MATCH. The expression that follows needs to be directory safe for what
# we're doing, so we limit what is matched to -, letters, and numbers.
# Here's how it used to work:
#:0
#* ^Mailing-List:.*twilight-.*icculus\.org
#* ^Mailing-List:.*twilight-\/[-A-Za-z0-9]*
#"List:Twilight-${MATCH}/"
:0
* ^Mailing-List:.*neither-.*icculus\.org
* ^Mailing-List:.*neither-\/[-A-Za-z0-9]*
"List:Neither/"
:0
* ^List-Id:.*eug-lug\.efn\.org
"List:eug-lug/"
Use postfix or exim? Miss qmail's single line command to make a Maildir?
Well duh, write a script! Here's one:
#! /bin/sh
# In the absence of real error handling...
set -e
install -m 700 -d $1
install -m 700 -d $1/tmp
install -m 700 -d $1/cur
install -m 700 -d $1/new
___
EuG-LUG mailing list
[EMAIL PROTECTED]
http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Zero Wiping, partition table resurrection
On Mon, Dec 29, 2003 at 11:56:56AM -0500, Linux Rocks! wrote: > : A poorly kept secret is that HDs have things like bad sectors all the time > : and that the drives themselves have known for ages how to recognize > : sectors that are going bad before they do. When this happens, they > : silently reroute your bits to unused sectors they do not advertise as > : actually being available. This screws with things like interleaving, but > : nobody interleaves for a speed boost anymore. (Why? RAID finally really > : does mean inexpensive disks...) Anyway, you only see bad blocks when the > : drive runs out of spare sectors. > > True, those are called "reserve sectors, typicly drives have 1% reserve > sectors to "replace" bad sectors on the fly... > I dont understand how this effects interleaving... As i understand it, > Interelaving is the space between the tracks (or is that cylinders). > Anyway... An interleave is a re-ordering of sectors. Generally, it was assumed that most controllers can't write two adjacent sectors and you have to wait a whole revolution of the drive to write the next one. Adding an interleave of 2 or 3 (read every odd or every third sector) had the potential to really speed things up. Interleaves were drive, controller, and on VLB and PCI machines, bus-speed specific. Some systems could handle 1:1 just fine, others did better with 1:2 or 1:3 with the same drive. Having to seek to a reserved sector kinda breaks the smooth write process (and the smooth read for that matter), so that's a factor. I'm told that ESDI drives lack reserved sectors by design for this reason. No way to verify this last point though. > : Low-level formats of drives are scary undertakings. Usually when I've > : needed to do it, it was because the drive was damaged. Usually, the > : reformat doesn't help matters even a little bit. Just a word of warning. > > Ive done this many times... its really not so scary if you use the right > tools... I think Ive broken 1 out of 50 drives or so, and ive recoverd the > other 49 (ok, some of those drives only worked for a day or so, but many > lived happy lives for many years following...) Usually, having to break out the utility (or rather, go digging for it on google or the manufacturer's website) is a sign of impending doom. It's like installing Windows 98. As soon as you reach for the software, you know you're about to do something dangerous and undesirable. You just know you're going to have even bigger problems just as soon as you lay the thing aside and try to do anything useful with the hard drive. ;) > You can use older bios low level formatting if you know all the right > settings.. letting it guess can be risky though... I recommed using > manufacturers LLF utility over bios LLF utilities. I have only ever used a BIOS formatting with SCSI because SCSI controllers of old had different schemes for the format of the drive based on who made the controller. Some settings still exist that cause you to be suddenly unable to access a SCSI drive such as whether the SCSI BIOS should try to make the drive work in DOS or not. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]QMail Folders
On Fri, Dec 26, 2003 at 05:01:56PM +, Bob Crandell wrote: > How do I have QMail create folders in addition to cur, new and tmp in Maildir? I generally use procmail for the purpose.. [EMAIL PROTECTED]:~/Library/Mail/Folders$ ls Inbox/ List:Twilight/ Sent List:Neither/ List:eug-lug/ Spam/ [EMAIL PROTECTED]:~/Library/Mail/Folders$ It is also possible to nest Maildirs.. The tree for the above folders might look like... ~/Maildir/ ~/Maildir/Neither/ ~/Maildir/Neither/cur/ ~/Maildir/Neither/new/ ~/Maildir/Neither/tmp/ ~/Maildir/Twilight/ ~/Maildir/Twilight/cur/ ~/Maildir/Twilight/new/ ~/Maildir/Twilight/tmp/ ~/Maildir/cur/ ~/Maildir/eug-lug/ ~/Maildir/eug-lug/cur/ ~/Maildir/eug-lug/new/ ~/Maildir/eug-lug/tmp/ ~/Maildir/new/ ~/Maildir/tmp/ ...but I don't generally like it. I don't remember the .qmail filtering syntax, but I remember that it wasn't as good as exim's and nowhere near as good as procmail or similar. If your qmail setup doesn't already use procmail if available |procmail with whatever args are appropriate is just a line away in your .qmail file.. ;) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Zero Wiping, partition table resurrection
On Sun, Dec 28, 2003 at 11:47:31PM -0800, Larry Price wrote: > This is kind of a borderline question; > > A disk was intentionally zero'd out using > > dd if=/dev/zero of=/dev/hda > > however the DOS fdisk utility couldn't rebuild the partition table > afterwards. Generally speaking some fdisk programs really can't cope with that. I thought the DOS one could. I know that if the disk isn't screwy, cfdisk will be able to fix it. A zeroing should not affect the low-level format because from the perspective of zeroing the drive, you're operating at a logical level rather than a physical level. A poorly kept secret is that HDs have things like bad sectors all the time and that the drives themselves have known for ages how to recognize sectors that are going bad before they do. When this happens, they silently reroute your bits to unused sectors they do not advertise as actually being available. This screws with things like interleaving, but nobody interleaves for a speed boost anymore. (Why? RAID finally really does mean inexpensive disks...) Anyway, you only see bad blocks when the drive runs out of spare sectors. In order to pull this off transparently, the PC needs a very different picture of the drive than the drive electronics get. The PC sees a virtualized disk, much the way Linux applications believe that they have some 3 gigs of memory, regardless of how much RAM and swap you actually have. Writing to the Linux memory space, or to the drive's virtual space, has little or no real effect on the low-level representation of what's actually stored on the platters or in the DRAM latch matrices. Low-level formats of drives are scary undertakings. Usually when I've needed to do it, it was because the drive was damaged. Usually, the reformat doesn't help matters even a little bit. Just a word of warning. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]multimedia puzzle...
On Sat, Dec 27, 2003 at 04:47:05PM -0800, T. Joseph Carter wrote: > On Sat, Dec 27, 2003 at 01:06:44PM -0800, Ben Barrett wrote: > > I can suggest ncftp for a very good CLI FTP client. > > lftp is also rather nice, but it is not always perfect (it doesn't show > the MOTD for example..) > > > > AVI is a wrapper, AFAIK, for a wide variety of stuff. Even for divx, there > > are multiple versions. There are some alternative media players for 'bloze > > on sites such as cnet's download.com ... best wishes for a media-filled new > > year! > > AVI files are actually RIFF files given the AVI extension so you know that > they contain video. RIFF is also used by WAV files, with the WAV > extension meaning that the files contain PCM sound data with no or > lossless compression. > > Apple's equivalent to RIFF is called moov. The name of moov is taken from > its first tag which holds the header information which identifies the > file's contents. Both come from the Amiga IFF file format, which is > described by many Amiga fans as "binary XML". I do not know what the R > stands for in RIFF, but RIFF is a little-endian format. The Amiga and mac > both are big-endian. The structure of the file, regardless of whose > version you're using, is as sequential blocks prefixed by a block size and > type. I believe the size is a 32 bit number. The type is four bytes and > is intended to be ASCII for some measure of sanity when viewed in a hex > editor. ie, moov being the QuickTime format used originally for QT movies > only, the identifier makes perfect sense. > > A little more useless/random information, file(1) indicates that WAV files > are RIFF WAVE. You guessed it, WAVE is the name of the block which > contains the header information telling you that the file is PCM data, > what rate, how many channels, etc. > > > Caveat 1: Whether the block name or size comes first, I can't remember off > the top of my head. > > Caveat 2: Given how much uses the moov format these days, I have to > wonder if I'm not mistaken about its origins with QuickTime. > > Caveat 3: I'm not sure if moov is in fact just Amiga IFF and the only > thing special about it is the signature tag. > > Caveat 4: In order for IFF to be binary XML, one must see nesting. > Indeed, there is nesting, but not as much as you find with XML. At some > point in the file, you wind up with one or more huge bloxks of data in > some format specified either by the header or the name of the tag itself. > > Vaveat 5: There are too many caveats in this bit of useless information. <@Knghtbrd> LordHavoc: moov is basically IFF isn't it? <@LordHavoc> Knghtbrd: similar overall but not quite <@LordHavoc> Knghtbrd: 4 byte size, 8 byte name, then the data <@LordHavoc> Knghtbrd: IFF is 4 byte name, then 4 byte size, then the data <@LordHavoc> Knghtbrd: (same as RIFF and AIFF) <@LordHavoc> Knghtbrd: differences between IFF, RIFF, and AIFF are in name restrictions (IFF requires all names be uppercase, for example, RIFF and AIFF do not impose that restriction, and RIFF is little endian where as the other two are big endian) <@LordHavoc> Knghtbrd: and in completeness (IFF is a larger spec than RIFF and AIFF which do not support things like archives, catalogs, and certain other special constructs) <@LordHavoc> Knghtbrd: and of course in who runs the registration body <@LordHavoc> Knghtbrd: (all formats are supposed to be registered) moov begins with those four letters. After that is a Uint32 size, char name[8], and data. Nesting of tags is doable only because Apple has decided that it wants to do that - the data is a binary black box as far as the format is concerned. Interesting is that AIFC files on the mac are not in AIFF format, but actually moov format. That's enough useless information for one day. ;) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]multimedia puzzle...
On Sat, Dec 27, 2003 at 01:06:44PM -0800, Ben Barrett wrote: > I can suggest ncftp for a very good CLI FTP client. lftp is also rather nice, but it is not always perfect (it doesn't show the MOTD for example..) > AVI is a wrapper, AFAIK, for a wide variety of stuff. Even for divx, there > are multiple versions. There are some alternative media players for 'bloze > on sites such as cnet's download.com ... best wishes for a media-filled new > year! AVI files are actually RIFF files given the AVI extension so you know that they contain video. RIFF is also used by WAV files, with the WAV extension meaning that the files contain PCM sound data with no or lossless compression. Apple's equivalent to RIFF is called moov. The name of moov is taken from its first tag which holds the header information which identifies the file's contents. Both come from the Amiga IFF file format, which is described by many Amiga fans as "binary XML". I do not know what the R stands for in RIFF, but RIFF is a little-endian format. The Amiga and mac both are big-endian. The structure of the file, regardless of whose version you're using, is as sequential blocks prefixed by a block size and type. I believe the size is a 32 bit number. The type is four bytes and is intended to be ASCII for some measure of sanity when viewed in a hex editor. ie, moov being the QuickTime format used originally for QT movies only, the identifier makes perfect sense. A little more useless/random information, file(1) indicates that WAV files are RIFF WAVE. You guessed it, WAVE is the name of the block which contains the header information telling you that the file is PCM data, what rate, how many channels, etc. Caveat 1: Whether the block name or size comes first, I can't remember off the top of my head. Caveat 2: Given how much uses the moov format these days, I have to wonder if I'm not mistaken about its origins with QuickTime. Caveat 3: I'm not sure if moov is in fact just Amiga IFF and the only thing special about it is the signature tag. Caveat 4: In order for IFF to be binary XML, one must see nesting. Indeed, there is nesting, but not as much as you find with XML. At some point in the file, you wind up with one or more huge bloxks of data in some format specified either by the header or the name of the tag itself. Vaveat 5: There are too many caveats in this bit of useless information. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Associating Files...A Reply to T.
On Sat, Dec 27, 2003 at 12:32:47PM -0800, nyal wrote: > Dang T, gimme a chance to grab an extinguisher!!! I'm NEW to Linux so I don't > know it all yetand if you read my second post you'll see I figured out > what I wanted to do all by myself! I've gotten some excellent suggestions > from the list and hopefully I'll get more as I continue this journey off the > beaten path. Sorry if I came across as more disgruntled than I intended. Somehow my answer that more information was needed to answer your question got itself intermingled with a rant on how complex it is to do such a very simple thing in Linux because everyone and their dog has a "standard" way of doing it which is incompatible with everyone else's "standard". Just keep in mind that most of us aren't psychic when asking your questions. Last time I checked, John Edward was not a lug nut. But then again, I'm convinced I'm more psychic than he is (for that matter, so are my sister's dog and the pizzabox sitting in the dumpster behind my dorm building..) But that's another rant. ;) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Associating Files
On Sat, Dec 27, 2003 at 06:03:25AM -0800, nyal wrote: > Greetings All, > > I'm trying to figure out how to associate certain file types with certain > apps...like Kuickshow with .jpgs and XMMS with .pls files. I know it can't > be that hard, I'm just not looking in the right place. If anyone knows of a > site that explains how I'd be mucho appreciative. For what? You think this is integrated? The beauty of X11 is that you must configure each and every little thing seperately and not necessarily in a sane or logical manner. Gnome has one place. KDE another. XFCE another. ROX yet another. GNUStep still another. Mozilla has its own, and then there's the generic MIME associations, a square wheel used by oldschool applications which do not use the newer (but still square) wheels created by KDE, Gnome, ad nausiem. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Re: PDA's
On Mon, Dec 22, 2003 at 07:58:25PM -0800, Dirk Ouellette wrote: > I told my wife to get me this one from BestBuy as the 2 rebates add up > to $50 and that leaves it well within her price range at $84. It uses > Palm OS 4.1 software so I assumed it would work well. Thanks for all > advice. The one minor disadvantage of the Clie's is that they expand with memory sticks rather than SD/MMC. You probably won't be adding Bluetooth or WiFi in the memory slot with the Clie. This kinda limits the device to syncing with your PC and serving the intended purposes. Wireless networks add a whole lot of future possibilities which nobody at Palm dreamed anyone might do with the thing (similar to the incredibly cool if pointless hack to turn a Palm into a universal IR remote..) This may not be a factor for you though, given that pretty much all of the USB memory card readers are now quite inexpensive and nearly all of them work flawlessly in Linux. (It's odd, Bluetooth MMC is very cheap but WiFi MMC is currently woefully expensive. WiFi CF is relatively cheap, but Bluetooth CF is as expensive as WiFi is in MMC format! Palms use SD/MMC, PocketPCs use CF.) The major disadvantage that applies only to me is that Sony uses a different font than Palm does, and Sony's font lacks the boldness which makes the tiny Palm font so readable for me. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]pop3 -> imap
On Mon, Dec 22, 2003 at 06:30:07AM -0800, T. Joseph Carter wrote: > On Thu, Dec 18, 2003 at 04:46:33PM -0800, Rob Hudson wrote: > > I think I'll try the UW imap server. I'll have to figure out how to set > > this up using SSL similar to how I did it for POP, but I see that it has > > docs on this. > > > > Thanks all. > > Brain dump regarding Palm options: Obviously I need to look more closely at what I'm replying to when I am (still) awake at 6am. =) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]pop3 -> imap
On Thu, Dec 18, 2003 at 04:46:33PM -0800, Rob Hudson wrote: > I think I'll try the UW imap server. I'll have to figure out how to set > this up using SSL similar to how I did it for POP, but I see that it has > docs on this. > > Thanks all. Brain dump regarding Palm options: Most modern Palms have USB cradles. This is significant because it means you'll need a kernel driver to talk to it in Linux. The Handspring Visor was the first Palm to use a USB cradle and it just so happens that with a little tinkering the Visor cradle driver works with modern Palms as well. Dennis Soper recently got to play with this. I suspect that the driver in the kernel supports the standard Palm cradle by now, but he can tell you more I hope. Palm devices classify into two groups, PalmOS 4 and PalmOS 5. The PalmOS 4 devices all use 33MHz Dragonball (m68k) processors. They're not very powerful at all, but they are effective and they had all kinds of hacks available that just don't exist with the new PDAs. PalmOS 4 devices use Graffiti, the single stroke alphabet that goes back to the original Pilot. Graffiti2 is a download, but on a PalmOS 4 device you'll never be able to write your graffiti on the screen itself with such models--you're limited to the writing area because the CPU is too sucky to interpret the alphabet outside the writing area. Taps and drags are as good as it gets on the main screen. PalmOS 5 devices use ARM CPUs and I believe they start at 133MHz and go up from there. They feature Graffiti2 ONLY. The original alphabet had some patent issue applied to it or something. Graffiti2 is regarded as more natural. Usually I prefer the Graffiti2 letters. You can enable the feature to write anywhere on the screen with PalmOS 5, but it's not very accurate with my crappy penmanship and Palms without a writing area have a keypad anyway. In terms of the PalmOS 5 devices, the Zire 71 is hard to beat for toy factor. It can be had under $200 easily nowadays, and is a color screen. It's the model I carry with the built-in camera. It can also be a SD/MMC based mp3 player, play some Sega GameGear games, and is generally a nifty gadget. I can go a few days between charges, but should charge it nightly if I were smart. Only 16 megs of RAM, which is paltry compared to the Tungsten series. Much thinner and smaller is the Tungsten T, a PalmOS 4 device that can be had cheaper. It features Bluetooth, is thinner than the Zire 71, and the writing area crunches inside the case. It's also a color device, and is more affordable than the PalmOS 5 devices in the Tungsten series. The Tungsten T's are the smallest Palms with full-sized screens. You can get a smaller screen with the older m series, but they're almost as thick as my Zire and have a dinky screen to boot. The Tungstens have the real screen, but are thin. The Tungsten T2 is the PalmOS 5 version of the Tungsten T. A very nifty Palm, same crunchable writing area so it takes up less real estate when you're no using it. Color screen, battery is not quite as good as the Tungsten (which had much less CPU to power with it), but it's not too bad really if you charge it every night. It has some of the game and music options of the Zire 71, but it's missing the camera. You can do better for a camera with $30-50 though. The Tungsten T3 is an upgraded version of the T2. The writing area is gone, replaced with a longer screen. I _believe_ you can enable a virtual writing area without having to enable writing anywhere on screen. Or, you can have a taller or wider screen than normal. This is the PDA I'd buy today if I were to buy another. The Bluetooth is nice, though probably not so useful to a Linux user. These things run about $300. The last of the non-keypad Tungstens is the Tungsten E, a shiny silvery thing that looks a lot like the Zire 71. The camera is replaced with a bit more RAM, mainly. This is the budget Tungsten, but I believe it does still have 32 megs, which is more than I've needed for anything I've done. I bet it can be found for under $200 without much effort. So far excepting a jab at the m series, I've talked only of color models because I need to be able to read the thing and I can't read the sort without color and backlight. The Zire and Zire 21 sell as basically a school supply, and are priced sub-$100. No color, and the Zire is a PalmOS 4 device. I do not know about the 21, but I assume it may be as well. The 21 has more RAM. You won't get any game made by Sega on the low-end Zires and mp3 players aren't in the running. Their main claim to fame is that they're cheap and, if backed up regularly, almost disposable if they had to be. The Zire's have a silvery plastic hardcase available, I recommend it (one size fits all.) The Tungstens have metal hardcases available, and the one for the T series is really nice IMO. 3rd parties make cases for various Palms or generic PDAs. I mention cases because the ones that come with the Z
Re: [eug-lug]Adding/Detecting USB Devices with RH *+
On Sun, Dec 21, 2003 at 09:55:01AM -0800, Jim Darrough wrote: > I have RH 8.0 running (probably will upgrade to 9.x soon) and > don't understand how to make the system recognize USB devices. I want to > use a USB Hard Drive and a USB CD/RW drive. Can someone point me in the > right direction? I sure hope those are USB 2 devices... Both should be part of the usb-storage driver once you have the appropriate chipset driver for your USB controller (EHCI for USB 2, UHCI most likely, or possibly OHCI, for USB 1.1. You have two options for a UHCI driver usb-uhci and uhci (the JE driver). I can't recommend one over the other, but I suspect most likely a canned kernel in RH uses one or the other. I do not know if 8.0 even supports EHCI. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Sound Blaster Question......
On Sat, Dec 20, 2003 at 05:00:01PM -0800, Mr O wrote: > Lots of good advice so far. The last remaining thing that > affects how the proper drivers works is whether or not your card > is retail or OEM. There is an actual difference in the design of > the two that affects cost and performance. Generally speaking, > the cards with all gold connectors are the easiest working. > Colored connectors are OEM cards and can be more of a hassle. > So, armed with all that knowledge... you're off to the trenches. This really only affects the analog/digital switch on a certain class of pre-5.1 live cards, AFAIK. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Sound Blaster Question......
On Sat, Dec 20, 2003 at 07:16:23AM -0500, Nyal wrote:
> Greetings all,
>
> I'm new to Linux, currently using Mandrake 9.2 and I have a question about a
> soundblaster card that's in this box I just bought.
>
> I'm trying to get it to work but so far no luckMandrake recognizes the
> card and has snd-emu10k1 (EMU10K1) as the driver
You've got a Live or Audigy card. What's more , it looks like you have
ALSA drivers for it. Before I get into many details about Linux, there
are some interesting things you need to know about the emu10k[12] and AC97
combination. The description below is based on a Live card, but it
applies to Audigy's as well so far as I know:
Windows tries hard to hide this from you, but you basically have two sound
cards. One of them is analog, the other is digital. All of your analog
connections on the card go through your AC97 chip. Your analog speakers
also run through the AC97. The AC97 has an aux output that goes to the
EMU10k1 (or 10k2 for Audigy) chip. This card controls digital speakers
and has the effects processor and whatnot. It handles wavetable (only
matters with ALSA, OSS drivers don't support this function) if you're
using the card as a MIDI synth, PCM sound (playing wav or mp3 files), etc.
It also controls the {Live,Audigy}Drive, if you have one of those.
The reason why this matters is that, the emu10k1 also routes back into the
AC97 chip via its PCM input. This means it's possible to create feedback
loops and the like when fiddling with the various mixer settings, etc.
Also, what you hear out of analog speakers may not match what you hear out
of digital ones. Doing it this way means it's a little less flexible than
it could/should be, but it allowed Creative to use an off-the-shelf AC97
chip and save some green.
To configure this beast, you want alsamixer. And you want to hit the
emu10k1 ALSA wiki page, though you'll have to google for that since I do
not remember where it is now. alsamixer has LOTS of settings, and most of
them you will be able to change there, but should not. Their values make
no sense anyway, they're actually switches and things you'd set with
another program. The wiki page will have details.
> I've played with AuMix and KMix but no luck.looked at the Soundblaster
> website but no luck (or drivers) there.
You won't find it there. You want to go to either emu10k1.sf.net (OSS) or
somewhere off of alsa-project.org (ALSA, what you're using..)
> The motherboard I've got (ASUS K7M) has Onboard AC'97 audio that I can use if
> the SB card is a no-go.
>
> Am I missing something or is Creative just Linux Unfriendly? I'd like to get
> sound up and running so's I could listen to CDs (or 'net Radio) while I
> browse the www.
You're missing a lot of somethings, mostly in that the driver is far more
complex than you bargained for because, although ALSA tries to simplify
the setup a little bit, its mixer ends up complicating things. The OSS
drivers can do more (eg, use the effects processor), but you have to
control all of the emu10k1 functions with seperate programs that use
neither OSS nor ALSA control mechanisms. In OSS at least, the OSS mixer
controls are tied to the card's AC97 controls, so unless you have digital
speakers or so you're probably good to go.
The OSS drivers can do one thing for people with digital speakers - turn
off the AC97's PCM and set the AC97 capture to igain (that is, all analog
devices) which is good if you want line-in and cdrom and everything else
to just work without thinking about it for digital speakers as they do for
analog..
The ALSA drivers have only one interface for configuring them, the mixer
(and sfxload if you use the wavetable feature, but that's a different
story all together..) This winds up simpler in the long run because you
can set it once and forget it unless you need to change capture devices
and set up ALSA to automatically save and restore your mixer settings on
reboot. (It should already be so configured for you..) Learning what all
those damned controls are for and how to set the ones that aren't sliders
if you need to (you shouldn't need to) is a bit of work.
Windows is so much simpler. But, the Linux drivers are both so much more
powerful. ;)
> Any help is mucho appreciated and look for other newbie questions from me in
> the future!
Hope that helps at least a little bit.
___
EuG-LUG mailing list
[EMAIL PROTECTED]
http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]pop3 -> imap
On Fri, Dec 19, 2003 at 11:40:36AM -0800, Rob Hudson wrote: > Is there a way to set up imap so that it uses a different password than > my user password? I'm cautious of sending passwords in plaintext and > set up imap on port 993 so it uses SSL. But if I'm logging into > Squirrelmail not using https, what's the point? I supposed I should > look into compiling apache to use SSL as the preferred option. > > Any advice? Yes, compile mod_ssl and make squirrelmail REFUSE to allow non-SSL connections. ;) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]pop3 -> imap
On Fri, Dec 19, 2003 at 08:58:17AM -0800, John Sechrest wrote: > % > We are using Cyrus on PEAK. And we have seen it perform better than > % > UWash IMAP, which we used before. However, it has broken some > % > of our older tools like MUTT and MH. > > % At least mutt should be easily fixed. I can't speak to mh, but mutt is a > % program I know inside and out. ;) > > Yes, after a bit we got mutt working again. > > but MH seems to have no concept for IMAP. > > It is a problem for me, since I use MH extenstively. > And I use procmail extensively. > > So the Cyrus switch has broken two of my main tools on those systems. Procmail is the other program I know inside and out. ;) How did your setup change exactly, in terms of how mail was stored? I may be able to help you fix both, though MH fixing may well be a side-effect since I do not know anything about it. ;) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]new graphical browser winner for me.
On Thu, Dec 18, 2003 at 06:29:31PM -0800, Mr O wrote: > Have these tests all been run at various times during the day > and in various order to prove your results? :) I know that if > Mozilla is already open and I bring up a new tab I can have > euglug.org loaded before I get to blink. Of course if you're > counting loading time for the browser too then I'm at a loss. Indeed, it makes most sense if you download the page to disk and then dump the copy from disk in a loop about 1000 times and report the time of the result. Benchmarks should be as free of randomness as possible, which means a network connection is out unless that is what you're benchmarking. The loops guarantee that the disc caches have as little effect as possible. Take the resulting times and / 1000 them for the effective performance differences. You could do it only 10 times, but I suspect that would give you inconcievably small numbers to work with. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]pop3 -> imap
On Thu, Dec 18, 2003 at 03:52:09PM -0800, John Sechrest wrote: > We are using Cyrus on PEAK. And we have seen it perform better than > UWash IMAP, which we used before. However, it has broken some > of our older tools like MUTT and MH. At least mutt should be easily fixed. I can't speak to mh, but mutt is a program I know inside and out. ;) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]pop3 -> imap
On Thu, Dec 18, 2003 at 01:59:25PM -0800, Rob Hudson wrote: > I've got teapop set up on my FreeBSD server that handles my mail. I > heard good things about IMAP and would like to also install > SquirrelMail, which requires an IMAP server. > > Is there anything special I need to know to move from POP to IMAP? > What is a good choice for an IMAP server? Courier is nice for IMAP, but I think it uses Maildir internally (that would have been a feature when I tried it..) Courier provides a complete integrated Mail Thingy, but of the suite, I found that I did not like it's SMTP server. It just wasn't as good as exim IMO, let alone postfix. It happens that both of these speak Maildir themselves, so I was fine to use it with Postfix. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Larry in the news!
On Thu, Dec 18, 2003 at 12:34:58PM -0800, Patrick R. Wade wrote: > > I attempted to add one today via USB using a device I thought was > >supported. It's not. > > Not supported where; at the TiVo or at the other end? I've used the > same model i gave you, with Debian on my VAIO. IF you're having > trouble with your Mac, you could install Linux thereupon and try it. The mac end. And I'm not putting Linux on my G5. I have a damned fast, damned quiet machine here. The TiVo makes far more noise than it does. The ability to throttle down the fans and control the CPU speed is in the PMU, for which Linux has no driver yet. This means the CPU runs at half speed, and I don't recall what it does with the seven fans. As it happens, my PS2 supports the thing in games which recognize a broadband adapter. Someone's working on a mac driver for this thing too apparently, but the project's only just started. The company provides full specs and there are both BSD and Linux drivers for the thing to refer to. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Larry in the news!
On Wed, Dec 17, 2003 at 06:42:26PM -0800, Ben Barrett wrote: > Post divx to list... 8*> I would except ... In order to be able to talk to the tivo without exposing it to the world wide hostile internet, I need another ethernet port. I attempted to add one today via USB using a device I thought was supported. It's not. The 10baseT version of the chipset had a driver made by ADMtek, but the version used by the Linksys device has no driver I can find anywhere. I'm not the only one who has looked. Porting the driver would involve knowing the kernel internals of both FreeBSD and Xnu, and that's going to take more time than I have. My best bet at this point is to either buy an Airport Extreme card and make my network connection over that, freeing my Ethernet port for communication with the tivo or to find a compatible PCI card. While many are compatible, the one I happen to have (a DP83815) is not. Plus, this uses a valuable PCI-X slot for a single ethernet port... Ugh. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
[eug-lug]Larry in the news!
Our own Larry was in the news tonight saying that not all commercial email is spam. He was promptly lynched. =) No, actually, he was talking about the anti-spam law and how it's a beginning and all. I caught most of the segment at 5pm on channel 16 here in Eugene (3 on cable). I am watching for a repeat of the segment in their future newscasts today to see if I can catch the first 10-15 seconds of it. If I can get it, I'll try to capture it off the TiVo and bring it to a clinic or something. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]bash command line - loop over a range
On Mon, Dec 15, 2003 at 02:14:57PM -0800, Cory Petkovsek wrote: > On Mon, Dec 15, 2003 at 12:41:09PM -0800, Bob Miller wrote: > > Rob Hudson wrote: > > > > > But is there a way to do something like this?: > > > for i in [1-20] ; do cp graphic_$1.gif graphic_text_$1.gif ; done > > > > > > It doesn't work but it seems like there should be an easy way to set up > > > a range like that. > > > > #!/bin/sh > > > > for ((i = 7; i < 14; i++)) > > do > > echo $i > > done > > This doesn't work on all bashes. Specifically bash on my solaris 8 box. > Why not? Because it's new, even as bashisms go. for i in `seq 1 9`; do ... That should be portable. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Those HP Servers
On Fri, Dec 12, 2003 at 05:21:54PM -0800, [EMAIL PROTECTED] wrote: > > Or would society get the most benefit from recycling these junkers > > responsibly, then holding a bake sale to buy Headstart a new Celeron > > box? > > > > Brainstorming time. Throw out some ideas. > > Mini fridge. With an ice maker. Pull one of the SCA drive release levers for mini icecubes. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]FileSystem.mount()
On Fri, Dec 12, 2003 at 05:26:22PM -0800, Bob Miller wrote: > > Doesn't this open you up to the world of evil that is NFS security? > > Not if you restricted it to localhost. NFS uses a priveleged > port, and ordinary users can't sniff the lo interface, so > I don't see a way offhand to simulate a client or eavesdrop > on other clients from localhost. > > Maybe I'm missing something... You missed the network transparency is all. ;) As that was not part of the specification of lufs, it'd work fine to emulate NFS. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]power fluctuations
On Fri, Dec 12, 2003 at 02:16:44PM -0800, Ben Barrett wrote: > We're getting some power fluctuations here at work, by the river > roughly across from the UO. > > Anyone else? Here on the UO campus. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]FileSystem.mount()
On Fri, Dec 12, 2003 at 02:11:53PM -0800, Bob Miller wrote: > > The whole idea of a filesystem in userspace is pretty hackish, so I won't > > ask what's wrong with it. Instead, I'll ask: What would be a better way > > to do it? > > Give your filesystem an NFS server interface. An NFS server listens > for SunRPC/UDP packets and replies with more of the same. You can't > get better isolation from the kernel than that. Better, it uses the > very portable Berkeley Socket interface, so the same code could run on > Linux, HP/UX and Windows with a few tweaks. It can be network > transparent if you want, or it can be restricted to localhost. Doesn't this open you up to the world of evil that is NFS security? ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]FileSystem.mount()
On Fri, Dec 12, 2003 at 07:41:15AM -0800, Brad Davidson wrote: > Maybe 6 months or so? I don't recall exactly. I'm sure it's improved > since then, but it had a 'cruft' flavor to it that I don't feel from a > lot of startup projects. The whole idea of a filesystem in userspace is pretty hackish, so I won't ask what's wrong with it. Instead, I'll ask: What would be a better way to do it? ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]FileSystem.mount()
On Thu, Dec 11, 2003 at 11:33:45AM -0800, Brad Davidson wrote: > I gave LUFS/SSHFS a try a while ago. The project as a whole looked very > hackish, and it crashed constantly on my laptop. > > Now my laptop is PPC so there may be some endianness bugs, but even so - > it seemed very amateur to me. How long ago was this? ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]SMT soldering
On Wed, Dec 10, 2003 at 09:55:30PM -0500, Jamie wrote: > : http://www.9thtee.com/tivomemory.htm > thats some of the worst soldering ive seen ! (except when i was doing > training...) I saw some stuff... some people just shouldnt be allowed to play > with hot surfaces... > Seriously though, it looks like someone that was maybe an engineer or novice > did the soldering. In production, solder joints should all look consistant, > shiney, and have nice filets. (and no bridging). I know what good soldering looks like and wasn't even considering what the soldering they'd done looked like. I was thinking pasting it to give you an idea of the size of the pins (microscopic to my eyes..) > I suspect you could do as good as the photo (might want a big magnifying > glass in your case...) FWIW, I can't solder a plain DIP of the TTL variety. I lack the steady hands to avoid bridging in that sort of situation. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]SMT soldering
On Wed, Dec 10, 2003 at 11:43:47AM -0500, Jamie wrote: > I do, but Im not there right now :( Well, I was kinda trying to avoid sending it somewhere.. > As far as what they say, Id say they have things backwards. You almost never > want a low power iron, you want a hot iron (but a very fine tip (and one with > aa slight flat surface works best). You gotta be good, and work fast. > Ive done a lot of repair from people that were not good.. it really sucks, do > it right, do it fast, and dont burn it. They suggest tinning the tip and just using it to press the lead in place. That's all it takes for the tiny leads.. > A low watt iron gets everything hot before it melts the solder, and riskes > damage, a hot iron gets just the specific area hot, but risks burning (if > your not fast), and thermoshock. > Is it adding chips to empty locations? or is it piggybacking chips? > What is the lead/land size? Empty spaces on the board next to the existing memory. Here's the link to the upgrade for my model, including a reprint of the soldering instructions which appear on the forum: http://www.9thtee.com/tivomemory.htm > Pretty much all flux is liquid (unless your sweatting pipes, or doing stained > glass. Fluxes come in 2 catagories, waterbased and resin based. resin based > were phased out over 10 years ago because cleaning required solvents (we used > a lot of freon(sp?) back in the day... I did some testing and found that dawn > dishwashing liquid was about the best for stuff you could find around the > house (dont laugh, it really works!) The idea behind flux is to remove > oxides. Most of the solder you find will be resin core solder, and requires > solvents to clean (alcohol works, burned flux may need a little help with an > acid brush...), really burned flux may require an orange stick (or tooth > pick...) I believe it is still common to see solder use resin core, but a liquid flux is recommended for this rather tan relying on the stuff in the solder. > ESD/EOS is a big consideration too... gotta be safe and not volt your > circuitry, or you will end up with dead tivo (or worse... psycho-tivo!) Yeah, I'm a bit concerned that they don't talk about this. DRAM is notably picky about ESD. > Soldering really isnt that hard to do... Ive worked with a lot of people that > can do it in their sleep... but there are a lot of people that say they can, > but dont really know what they are doing... The only think I can suggest, is > watch them work on stuff before you let them loose on your tivo! It was mainly something I was considering doing if I am going to add more space to the thing, which is something I'm planning to do actually. > : Another option, apparently, is a new device that takes a PC133 512M DIMM > : and read-caches the entire database in RAM. That alone would probably be > : a decent speed boost, but it's a more expensive upgrade and would only > : affect the database. > > uhh pc133 ram is fairly cheap... check ebay, you may find one for $20... > (prolly spend more like $50...) True enough, but the upgrade card is not. ;) > : > : > : (For anyone wondering, I did attempt to find out and as far as I know, > : nobody has tried to overclock the 50MHz series 1 TiVo..) > hmmm... seems to me that if your CPU is 50 mhz, then why do you need pc-133 > ram? isnt pc-133 for 133mhz bus? how do you get a 133mhz bus on a 50mhz cpu? I suspect that any DIMM I could find would work, including PC66 and slower. The bus is 25MHz after all. I think they're just assuming that at this point it's pretty much all PC133 for SDR memory and that anyone who would know differently knows what they can use. > Is that a motorola chip? The series 1 is, yes. Series 2 machines are ARM chips, 4x the speed, and twice the memory (isn't that right Bob?) The Series 2 machines are not very hackable since the ROM verifies that the ramdisk is signed by TiVo before it runs anything and the ramdisk replaces any modified boot files with fresh copies. If you defeat the ROM (that is, dump it, modify it, make a new ROM chip, amd replace the existing ROM on the TiVo which is a SMT chip), you can then modify your ramdisk not to blow away your changes. Once your changes are not blown away, you can tweak your TiVo to your heart's content, until a new software upgrade replaces the ramdisk with one that blows away your changes and hopefully doesn't turn your TiVo into a brick when a ROM verification fails, leaving you without a TiVo till you pop the cover, put the drive in your PC, boot a custom Linux dist, and tweak everything until it works again.. Needless to say, despite how slow they are, we rather like series 1 TiVo. You can open it up and add a network card (series 2 boxes have USB ports for this purpose) and stream data right off your TiVo (pay extra to do this on a series 2 box), even copy shows from one TiVo to another (not possible with series 2) or extract video across the network,
Re: [eug-lug]Wanted
On Wed, Dec 10, 2003 at 11:36:10AM -0800, Rob Hudson wrote: > I thought it would be more, but you're right... I can get a lower end > Athlon with motherboard and memory for about $150 at mwave.com. mwave warning, they have been caught at times selling refurb goods without marking them as such. I landed one such component - an IBM hard drive. Who in their right mind would buy a refurbished hard drive, especially given that some models of IBM hard drives were at the time known to be flakier than most pastries. The price was a great deal, but not that great that it should have been a refurbished part with only a 30 day warranty (discovered 6 months later!) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
[eug-lug]SMT soldering
Does anyone here do surface mount soldering or know anyone who does? I'm considering a memory upgrade for my TiVo. It comes with only 16 megs and adding another 16 involves adding two SMT chips and reconfiguring the device to use them. I'm told it can be done without special tools if you have a low-powered iron, tiny solder, liquid flux, etc, but the other prerequisites are sharp eyes (or good optics) and a steady hand. I lack these. ;) Another option, apparently, is a new device that takes a PC133 512M DIMM and read-caches the entire database in RAM. That alone would probably be a decent speed boost, but it's a more expensive upgrade and would only affect the database. (For anyone wondering, I did attempt to find out and as far as I know, nobody has tried to overclock the 50MHz series 1 TiVo..) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]FW: Debian Investigation Report after Server Compromises
On Wed, Dec 03, 2003 at 11:30:48PM -0800, Ken Barber wrote: > Sorta like asking a PhD in psychology if he's ever looked at the > DSM3 It's quite probable that a PhD in Psychology may never have seen the DSM3. The DSM4 has been the standard for awhile now. The differences between the two are mostly interesting to intro undergrad students, apparently. ;) (We now return you to your regularly scheduled discussion, debate, and flamewars..) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]FW: Debian Investigation Report after Server Compromises
On Wed, Dec 03, 2003 at 12:34:23PM -0800, Ben Barrett wrote: > It came to my attention, after reviewing the Debian report, that there are > many mail systems out there, which use userland accounts for POP mail (not > secure, but plaintext) that also have SSH logins enabled. I was guessing > that this might've been how they got in with a "sniffed password". In fact, efn is one such place. efn does not currently offer any sort of encrypted email access and I'm sure there are at least three points between me and imap.efn.org where my password can be sniffed by anyone desiring access to efn servers. I get by only by using a passwd for efn that I use nowhere else. I actually tried to fix this at one point on the BSD/OS box which was serving up imap at the time, but for some reason stunnel just didn't work on that machine. Never did isolate a cause. The only option available in these cases is to ensure that mail is pulled only from unpriveleged accounts. However, privelege-elevating exploits exist for most OSes in some fashion or another, so that's no magic bullet either. The only real solution is to encrypt what you can, firewall like mad, and to be religious about security patches (in all senses of the word, I suppose..) ;) > I don't know how they could get a keyboard sniffer on a developer's machine > without first compromising that machine, in a similar fashion; so I'm > assuming that something like a shared [plaintext] password was > packet-sniffed initially... which still begs the question of where the > packets were sniffed. Was an ISP compromised or some insider helping out? > Maybe a developer was working via wifi, without considering the > implications? The details of how arronl's box was compromised are not being discussed openly. It was indeed vulnerable to attack, but a different attack. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Debian compromise solved.
On Tue, Dec 02, 2003 at 10:35:34AM -0800, Cory Petkovsek wrote: > > > Beat you to it. However read the additional info link I posted. The > > > patch is only in the 2.4.23 kernel. Do you keep right up with the > > > bleeding edge kernels? The debian servers were on 2.4.21 and 22. > > > > In the stable branch? You bet I do. > > > > I don't deploy them widely until I've tested them, but still. > > The 2.4.23 kernel was not available when debian was cracked. It wasn't > released until 6 days later on the 28th. That's true, but the patch was available before then. The patch was available in September. I remember discussing the implications when the problem was found. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Debian compromise solved.
On Mon, Dec 01, 2003 at 01:44:21PM -0800, Cory Petkovsek wrote: > > This Debian Security Advisory finishes the story. > > > > Let this be a warning to us all to keep all our software fully up to > > date. I don't want to read in the news about any exploited Linux > > boxes in Eugene. > > Beat you to it. However read the additional info link I posted. The > patch is only in the 2.4.23 kernel. Do you keep right up with the > bleeding edge kernels? The debian servers were on 2.4.21 and 22. In the stable branch? You bet I do. I don't deploy them widely until I've tested them, but still. It's amusing to note that Debian STILL does not support ATA133 with its standard install media, two years after I protested the need for such support. But then, that's Debian for you. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]printing via cups
On Tue, Nov 25, 2003 at 09:49:13PM -0600, Timothy Bolz wrote: > Ralph > Thank you. I edited /etc/cups/cupsd.conf and edited > > > Order Deny,Allow > Allow From 127.0.0.1 > Allow From 192.168.0.2 > #Deny From All > > > > AuthType Basic > AuthClass System > Order Deny,Allow > Allow From 127.0.0.1 > Allow From 192.168.0.2 > #Deny From All > > > I didn't know if #Deny From All should be hashed out or not. It should be commented unless you change the order. Why is obvious after a moment's thought. =) > If I want to add any other DHCP computers would I just add the range > 192.168.0.2 -192.168.100? You should be able to specify a CIDR subnet. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]spam filtering options and reccomendations
On Tue, Nov 25, 2003 at 03:01:11PM +, Bob Crandell wrote: > I like SpamAssassin. It's easy and tweakable. I get no false > positives, now, and only about 10 spam a day. Others here don't like > it. I think they are trying to achieve zero spam and I don't think > that's possible. Anyway, there will be more opions soon. SpamAssasin is good if you've got a small box and not many users using it. When you have thousands of users and tossing as much mail back and forth as efn or UO has to (I think efn may have the higher mail volume of the two), you start to see it's limitations. First, it's perl, so it's not as fast or as resource efficient as a C program would be. Second, it is perl that wasn't designed in any meaningful manner; it was merely hacked, tweaked, and appended as necessary to block the latest things people are doing. You need a pretty serious machine to keep up with that at significant volume. Anything bigger than efn or UO would probably need a cluster of SpamAssasin boxes to keep up, along with a set of things running to monitor the processes to kill and resurrect them when it cracks under pressure. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Serial ATA and Linux?
On Mon, Nov 24, 2003 at 06:22:45PM -0500, Linux Rocks! wrote: > What about MCA Let's leave that one dead, shall we? > we wont mention the oddball ones like std-32 or pc104 PC-104 was an ISA bus in a pinheader format. If you can imagine the 3.5 drive adapters designed to connect a 3.5 drive to a 5.25 floppy cable, you can imagine the thing which is done by PC-104. Basically, you get a pin header instead of a card edge socket. If memory serves, you could just literally crimp a ribbon cable to connect a normal ISA card to a PC-104 connector. It was small and easily embeddable, but not proprietary. PC-104+ required a slightly more complex interface because you need to route the PCI channels as necessary (which any FlexATX PCI riser can do) and also connect to a miniPCI connector, which is far more compact than standard PCI. It's also generally too fast for a ribbon cable. But again, the connector is not exactly proprietary, so it's doable. Note that miniPCI the PC104+/embedded system connector probably has very little in common with the miniPCI card form factor used for optional 802.11 in many modern laptops. The miniPCI connector in question is probably too thick for use in a laptop and is four-pin-wide. IIRC, the laptop connectors are similar to PCMCIA in their pin size. Do not know about the other standard. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]Serial ATA and Linux?
On Mon, Nov 24, 2003 at 11:06:02AM -0800, Bob Miller wrote: > My P4P800 motherboard has two SATA connectors. Does that mean > I could hook up a max of two SATA drives to it? Yes, just two drives. You've got a hybrid system with the typical limitations of hybrid systems: the new technology is supported, but you will find ourself wanting an upgrade when the new technology becomes the standard one. Seven years ago I bought my first system with PCI slots. There were three common standards of the time for improving ISA: EISA, VLB, and PCI. VLB's time had since passed and, and it was common still to see boards with two ISA, two EISA or VLB, and two PCI. The PCI on these boards was so-so, and there were only two. PCI-only boards were unheard of (and technically speaking still are for ia32 machines which still provide an ISA bridge for AT keyboards, PS/2 mice, and floppy drives..) But still, ISA implemented as a bridge to PCI allowed PCI to be the dominant system architecture and its performance has been good enough until now. At that time, I had the foresight to buy a board with only two ISA slots and four PCI. They were rare in those days. Very rare. Two years later it was still common to see three ISA slots, though it was no longer uncommon to have only two. It didn't matter, though, because the PCI was not constrained by ISA. SATA on your board is constrained both by IDE and by PCI. There;s no help for it I'm afraid. Look for PCI-X systems with SATA in the near future to provide more ports and more throughput than either IDE or PCI. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]sco: novell-suse breaks sco contract
On Tue, Nov 18, 2003 at 05:10:36PM -0800, Larry Price wrote: > Oooh, I'd be glad to give him a free pie !! Motion seconded. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]cameras, gphoto2 and USB (standards and lack thereof)
On Tue, Nov 18, 2003 at 02:51:03PM -0800, Bob Miller wrote: > > BTW, Larry, you forgot to remind folks not to unplug a USB mass storage > > device before unmounting it = ) > > Does mounting it "-o ro" help in that respect? No. Kernels get damned confused when mounted devices disappear. In the case of a floppy, you can put it back and it will be in the same spot, so the kernel will deal. In the case of a USB device, at least with 2.4.x, there is no promise the device will not appear in a new location with a new device name and thereby really fark things up. If you can access your camera with mtools, I suggest doing so. Too bad there is no way to tie mtools into the Linux VFS (good or otherwise..) Actually, this has got me thinking about how to better handle things like swapping disks, root filesystems on removable CDs, etc. It seems that the -bind mount option and / being completely virtual (ie, in the kernel) would be the right way to go about it, but I haven't come up with a good way to implement it. What's clear is that what I envision is outside the realm of automounters and traditional UNIX filesystem methodology. (Which means it'll never happen in Linux..) Ah well. =) ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
Re: [eug-lug]IRC
On Tue, Nov 18, 2003 at 12:49:25AM -0500, Linux Rocks! wrote: > I would ofcourse suggest using irc.freenode.net for an IRC service, > there are many #linux's out there... I'd counter that OFTC is a better network and is free of lilo's method of running a network, which seems to be alienating anyone who is compitent and then trying to bring in people who are not in order to prove that he is the only one who can be trusted to run things. His need for absolute control is somewhat reminiscent of other tales we have all heard or experienced, and the man lost all respect in my mind (which at one point was quite a lot actually) when he started begging for donations to help pay his bills so he could provide for his family on a part time job and spend the rest of his time basically sitting on his ass in front of an irc window all day. irc.oftc.net is run by many of the people who were responsible for LISC and then OpenProjects. (OpenProjects became Freenode at some point when lilo decided that his goal was to transcend irc and throw RFC 1459 totally out the window in an incompatible manner.. Most irc clients happen to work with it still just like most web browsers handle broken HTML..) irc.oftc.net is a very small net, but it has been growing somewhat slowly for a couple of years now. I've known the people running it today as long as I have known lilo - they're good people and they know what they're doing. ___ EuG-LUG mailing list [EMAIL PROTECTED] http://mailman.efn.org/cgi-bin/listinfo/eug-lug
