Re: Ethernet cabling - no horror
Peter Sealy wrote: On 24/01/2006, at 3:29 PM, Robert Howells wrote: OK ! With all that knowledge about ethernet , can anybody say what is the maximum distance is for a single hop before a repeater would need to be employed ? 100m Yep, for Category 5/5e/6 UTP, which is what usually gets used. STP has different rules (I don't know them), and fibre goes a *lot* further than 100m. I've seen 300m Cat5e runs in real-world use, but I wouldn't want to do it myself. -- Craig Ringer
Re: The horrors of Ethernet cables
Michael Parker wrote: I've made my share of Ethernet cables, and my opinion that it varies on ease depending on cables and tools that you have. But to be sure, this is not easy the first time you do it, and you are likely to mess up many times initially. Quality gear: - Good cable - Good connectors that match the type of cable, ie solid-copper (cross-connect) cable connectors or woven-copper (patch lead) cable connectors - A GOOD QUALITY CRIMPER If you only need a few cables and they are set distances that you can buy, you are better off buying them. Typically purchased cables are more reliable than cables hand made because people screw up (stating the obvious). Most of the time, this screw up, involves (assuming the cable works) not forcing the entire cable far enough inside the connector so that the individual cables are prone to yanking out over time and becoming flaky, with certain wires only making contact some of the time. Yep. Using the wrong type of connector for the cable, or using shoddy connectors and/or a crappy crimper is also a great way to ensure you have problems. It sounds like your missing some basic tools (like a cable tester). That can only make this tougher, but it's still possible. Note that most "cable testers" are actually just continuity testers. Sometimes they also check that the wiring order is correct. A real cable tester that can detect *flakey* cables (the worst kind) will cost you a very large sum of money from someone like Fluke. I used to make my own cable, but have since decided it's not worth it. For small batches I buy it in, for larger batches I get a cable tech who has good quality (and expensive) equipment to make up a bunch. I still make my own cable at home, but it's less important there. Remember that Ethernet cable doesn't just work or fail. Bad cable can cause truly bizarre issues - sometimes ones that don't even appear to be related to the computer in question. The most fun one I ever had was due to a cable that'd been damaged by rats - the Mac OS 7 box connected to it would take forever to log in to the network, and AppleTalk worked but TCP/IP didn't. The issues went away when I moved it to use a different under-floor cable, and came back if I put it back on the first one. Moral of the story: Don't waste your time and money making cable unless you can test it properly; it'll cost you more time and frustration than the purchase price down the road. -- Craig Ringer
Re: Appletalk problems
Jim Cummins wrote: Hi Craig: I tried to post this to the WAMUG list on wamug@wamug.org.au but it bounced (what's the address now for mailing? I just get digests). However, I saw in the archive that you answered a similar query a couple of weeks ago so maybe you can help. We run off our teaching material on a HP2200 (version 4.0) USB shared printer using an iMac as a server on Murdoch's LAN. It's started locking up on files with large graphics (particularly TIFF), even though we have printed versions of these before (originally created in Office 2000X but now running in Word for Office 2004) . I tried moving the file to the iMac: apart from launching Office 2004 for the first time and installing a few fonts it printed off with no problems. However, when I went back to my own terminal and tried printing the file, again the printer stopped the job and showed a red light-I guess indicating buffer overload. I've tried re-installing the printer, re-booting the iMac server, rebuilding the permissions etc. but it doesn't seem to make much difference. This seems to be an Appletalk problem. Any ideas? In all honesty, I know little about shared USB printers. The sharing can be done in two ways: - By a custom driver on both the client and printer server - By a driver on the printer server that presents the printer as a network PostScript printer and acts as a RIP for incoming PS data. Cheap printers usually use the former. Pro printers sometimes use the latter, though often you buy the driver as a separate product. Which it is can be important - if it's acting as a RIP, you need to make sure the RIP isn't exceeding any preset memory or disk scratch space limits, for example. I wouldn't assume this is an AppleTalk problem at all. First, it's not clear that AppleTalk is being used - if the printer is being discovered over ZeroConf (rendezvous) it could just as easily be using TCP/IP. Second, it's not clear that the problem is with the network communication rather than with the sending or more likely receiving software. I'd start by looking through the system logs (see Console.app) for anything informative. Also see if you can print to the printer using the network client driver from the print server (ie "network print" to the same computer). Some more details about what method of sharing is used would be helpful. What drivers did you need to install on the client? How do you add / select the network printer in the client? What does the printer configuration utility say about the printer on the client? And on the server? -- Craig Ringer
Re: MACROMEDIA Dreamweaver updated
Peter Hinchliffe wrote: I find it interesting that VersionTracker today lists Macromedia Dreamweaver 8.0.1 as an update: interesting because it's not called ADOBE Dreamweaver, and because development is obviously continuing, at least in the short term. A sure sign that it will not meet the same fate as Windows Media Player? Most developers don't tend to do name changes immediately after buying a product. Frankly, it's too hard, and it also confuses existing users. It's typical to do it next time they do a major update/release, when they've had time to bring the project in to use their tools, rename everything, tweak the user interface style to fit the rest of their products better, and do a bit of integration work with their other offerings. Can we really expect Adobe to continue to sell two Web Site design tools (Dreamweaver and GoLive)? Seems to me one of them will have to go... Adobe seem sensible enough to keep them going until they've managed to merge most of the best functionality of the two products. We're not dealing with Computer Associates here, thankfully. Still, one never does know. -- Craig Ringer
Re: Software Update, finding the package?
From: Peter N Lewis [mailto:[EMAIL PROTECTED] Sent: Thursday, 12 January 2006 11:30 AM To: MacGeeks List Subject: Software Update, finding the package? Perhaps someone has an idea for this. My dad has a shiny new iMac (PPC of course), but unfortunately he ignored my advise and went with Telstra as a ADSL provider, and so he has a measly 200MG/month limit. Software Updates probably exceed that each month all by themselves! I find it incomprehensible that providers who provide such low-cap services don't offer local unmetered mirrors for updates to major operating systems, games, and popular software packages. Even hen they do, they seem to go to no effort to ensure that vendors are aware of this and include them in the automatic mirror selection. Why not just host an Akamai node? Oh, yeah ... they wouldn't be able to rip off customers who don't understand that 200MB is barely anything in this world of 100MB OS updates and multi-megabyte web pages. -- Craig Ringer
Re: macbooks firewire 400 only?
Mark Secker wrote: I'm unfamiliar with the "express card bus" specks ExpressCard is a dual interface that supports a card that communicates to the computer over either USB-2 or PCI Express. The former is great for card readers etc; the latter would support FireWire 800 with no problems at all. -- Craig Ringer
Re: Eudora-Mozilla
Mervyn & Giuliana Bond wrote: After a long marriage with Eudora and Internet Explorer I am afraid they must go their separate ways. IE no longer works on some sites that I wish to access. Safari does most but Mozilla seems to manage all that are in my area of interest. However, I wish to continue to use Eudora, not the Mail associated with Mozilla. Problem - when a web site is highlighted in an email in Eudora how do I click on it and get Mozilla instead of IE? I'd recommend grabbing Firefox instead (available from mozilla.org); it's just the browser part of Mozilla, and integrates better into the operating system. It'll do things like use your internet preferences to see what mail client you prefer. -- Craig Ringer
Re: USB failure
Edward Arrowsmith wrote: Good morning Is it possible that a USB port can suddenly stop working? My PowerBook has two USB ports, one has decided to stop working but the other is fine. Any clues please? Yes, it can happen. If they're both connected to the same USB root hub it'd be most likely caused by a faulty connection on one port (eg the solder attaching the connector to the motherboard failing). If they're on different root hubs it could also easily be a failed or damaged chip. The issue could also easily arise due to software problems. If the Mac OS X boot CD supports USB, I'd test it out from there. Try using system profiler if it can be run from the CD. Of course, you should also follow all the usual "shotgun" Mac diagnostics - repairing permissions, checking the file system, and of course rebooting the machine. Speaking of system profiler, have a look there and see if the device appears at all when attached to the failed port. System profiler looks "deeper" into the system, and might see the device even if some layer above is having a problem. -- Craig Ringer
Re: Virtual PC and Mac - security
Hi,
I admin a bunch of WinXP systems at work, and this answer is made from
that experience:
Paul wrote:
My question relates to security. I'm pretty confident about Mac
security, but with Windows running within it, I'm interested in what
security precautions to take.
For example, should the Windows firewall be on?
Unless you need to disable it for some specific reason, yes. More layers
of protection rarely hurt.
Do I need to install the regular Windows OS updates.
You absolutely must keep it up to date. E-mail is not the only exploit
vector. For example, the recent WMF security hole could probably attack
you through a Visio file with an embedded WMF image.
Note I don't run any Windows email, so consider anti-virus to be
unnecessary.
That's not entirely true, but you'll probably be fine. I'd recommend
installing and using AVG from grisoft.com anyway. If you're confident
you can avoid getting the system infected, I wouldn't bother letting it
remain resident, instead just run scans occasionally. Mostly though it's
safer to let it run resident ("continuous scanning" essentially).
I have only had to use Explorer once, to access a
trusted site that doesn't support Safari/Firefox - so I also consider
anti-spyware to be unnecessary also.
You're probably OK there. I wouldn't want to use any persistent/resident
spyware scanner (many of which cause more problems than they solve), but
do consider using something like AdAware from lavasoft.de to do scans
every few months.
Note I'm connected to the Internet via ADSL with a NAT router. So
I'll raise the other niggling question - without Virtual PC running,
should the Mac firewall be on or not? I've heard conflicting
positions on whether the NAT router is adequate protection or not.
A NAT router makes it difficult to initiate a connection to your systems
from the outside world. The extremely basic firewalls in Mac OS X and
Windows XP are intended to do much the same thing. Things that can make
it through NAT (e-mail borne nasties, IRC / instant messenger attacks,
browser exploits, etc) will most often also make it through the XP and
Mac OS X firewalls.
That said, unless you specifically need to disable the firewall for some
reason, leave it enabled.
I personally wouldn't let windows near the net without AVG anti-virus
and ZoneAlarm firewall.
I agree with regards to AVG, though if you're a careful and informed
user you can usually get away without running it in resident mode. I
think ZoneAlarm is unncessary for most users with XP SP2, and it can
cause more problems than it solves.
I've had to fix more Windows boxes that broke due to ZoneAlarm
(conflicts, bad upgrades, broken uninstalls, etc) than due to viruses.
I cant emphasise the iffy security of IE enough.
This is especially true right now, where any website can take total
control of your computer through an image. That means that (eg) some web
forums will permit another poster to put up an image that'll control
your computer.
Given what you have said, Windows OS updates should only be necessary
*if* the normal operation of the OS and programs specifically require
it. You can, given enough patience, install just required updates.
Noo! Just turn on automatic updates. It's low fuss (though sometimes
really annoying about bugging you to reboot), works well, and won't
break things.
Also it may improve your mileage if you turn off:
Active desktop
Screen saver
Auto shut-down/sleep
Messenger service
Auto updates
I'd recommend leaving automatic updates enabled.
Off-line files
System recovery
By disabling system restore, you turn off the ability to repair your
system after a bad program/driver install. This can cost you a lot of
time and frustration, since while it's turned off it doesn't collect any
of the information needed to do recovery.
It does have a small but detectable performance impact. Despite that I'd
recommend leaving it on unless you keep backups of your Virtual PC disk
images somewhere, or are willing to risk the possibility of having to
erase and start again.
Menu fading/transition, window animation(?) (plus a few other items on
the same list in Display panel, this is from memory as I don't have
access to XP at this juncture; HOORAY!)
These are important, they cause surprisingly large slowdowns especially
under emulators. Switching the Windows XP theme back to the older
Windows 2000 style also puts a real rocket behind the OS. IIRC you can
do this in the Display control panel, or right click on the desktop
background and chose properties. I'm not 100% sure as I'd have to reboot
to check, and I'm using a real OS right now.
--
Craig Ringer
Re: networking for idiots(me!)
Neil Houghton wrote: Nope, no external routers involved. The two iMacs are connected by an ethernet cable, the old G4 flat-panel has the modem so is connected to the phone-line (as I mentioned we are talking dial-up internet, not broadband). I just set the old iMac to share its internet connection and told the new iMac to connect to the internet via ethernet and it seemed to just find the connection (ie how a Mac should work!) The old G4 mac is acting as a router and DHCP server the same way an ADSL or Cable modem/router might. Yes, in my case there is only the mac itself but typing its network address into the "connect to server" panel is how I did successfully connect. Good to hear it. H... not sure how this works in my case. Presumably the G4 iMac is providing a software routing solution for the internet sharing - possibly this is why its address is different when it is connected and sharing the connection? When connected the address was of the form 144.139.xxx.xxx That's a Telstra address range, so it was assigned from your Internet service provider. but when unconnected the address was of the form steve-jobs-computer.local Odd that it shows a name when not connected, but an address when connected. Perhaps it can't find out the name associated with the address Telstra hands you. While I get the concept of machines having addresses I don't really understand the distinction between Network address The network address isn't something you need to worry about usually. Your computer can almost always figure it out from your IP address and subnet mask. IP Address The IP address is the primary address you need to worry about. It's used by TCP/IP, AppleTalk/IP, and most other modern p, Ethernet ID rotocols. A DNS name like "something.com" is looked up in a directory to obtain the IP address to connect to when you ask to view something.com . Subnet Mask The subnet mask is used to chop a network into smaller groups of computers which are only allowed to talk to each other through a router. Subnets are network administrator voodoo, you just don't need to care so long as your computers's subnet mask matches that of all the other computers on the same part of the network. Router Address (when it's not another machine) When you try to talk to a computer that's not on the immediately local network, the router address specifies where it should address the data to if it wants it forwarded along to the real destination. It's sometimes called a default gateway address. Ethernet Address The Ethernet address is a unique address used for lower-level communication between computers directly on the local network. TCP/IP and other protocols travel over Ethernet when you're on an Ethernet network. The only time you're ever likely to care about this address is if you're trying to configure a DSL/cable router to hand out fixed addresses to your computers. IPv6 Address - all of which my machine(s) apparently have! IPv6 is the "new version" of IP (the lower level part of TCP/IP). It's designed to solve a bunch of problems with the current Internet, but nobody really uses it much even though it's been around for ages, because nobody else uses it much. Chicken and egg. You're extremely unlikely to need to care about IPv6. -- Craig Ringer
Re: LCD vs Plasma TVs
Reg Whitely wrote: Hi Chris Look closely at screen resolution. From what I've seen lcd has much better resolutions, similar to your Mac screen. Plasma has much lower res, some nearly only 640x480. On a big screen that would look terrible if hooked up to your Mac. It also depends on what you need to do. Plasmas have burn-in problems that mean they're unsuitable for use with many things, such as computer applications, that leave bright elements on the screen for long periods. Remember old macs like the Mac Plus that you'd see with a menu bar burned in across the top of the screen? I have idea why on earth Apple made the menu bar WHITE in those days, but anyway you'll suffer from similar problems with a plasma display. -- Craig Ringer
Re: National Australia Bank
Mark Secker wrote: I've literally had hundreds of them coming via dozens dead or dormant e-mail accounts of forwards that I have. never ever EVER EVER open ANYTHING like this EVER EVER even if it's from your own IT department Further more I have NEVER EVER seen a legitimate e-mail of this type. if they are legitimate they will tell you to ring their service center. and even then you look that up in the white/yellow pages rather than use any phone number they give you A legitimate bank email will never ask you for your PIN number, net banking details, etc - if it does, it's a scam, and should be reported to the bank. You should never follow a link in a message that appears to come from a bank (or, really, anybody else for anything important for security). Instead, use a bookmark or type in the address you know they have. For similar reasons, if they provide a phone number to call or address to send something to, do not use it - look up the bank's details in the phone book instead. It is extremely important to understand that an e-mail can appear to come from any address of the sender's choice. If I can have your permission I'll demonstrate this shortly by sending a message to the WAMUG list that appears to come from you. Because sender addresses are so trivially faked, you can not trust that a message is from the person it appears to be from, and should generally be suspicious of any message, no matter who it's from, that asks for security details, personal details, or asks you to take actions like open an attachment, visit a website, or perform tasks on your computer. Sucks, doesn't it? -- Craig Ringer
Re: Remote Desktop on a VPN
Robert Howells wrote: Preamble My daughter is home from San Diego for the week and we cannot get her PC laptop to work through a VPN to the Desktop on her PC at work. Windows XP, or some older version? What VPN software? What VPN technology does her company use (Citrix ICA, Microsoft RDP, IPSec, PPTP, "SSL VPN", etc)? I've made a guess from what's below, but confirmation would be nice. Procedure to get to remote Desktop : Click on VPN connect icon on Laptop it asks for a Password Password is entered and the Laptop should access the San Diego site. A " connected "indication shows OK CLICK On Laptop Remote DesKtop CONNECTION ICON From this I'm assuming you're using Microsoft Remote Desktop with the RDP protocol. Confirmation/correction would be nice. which should send data into the server at SD and cause the daughter's PC Desktop screen to appear with a Login screen We get nothing ... just a blank screen ... ! Some VPN technologies require the "server" to be able to connect back to the client, which won't work through a NAT router (like most DSL and cable routers) without special settings. Microsoft Remote Desktop Protocol doesn't seem to be one of them, though... so right now it's hard to say what's wrong. Laptop is OK, as it will access the remote desktop from my next door neighbour port on 4 port Router/modem , also from a PC shop ethernet and from a Macdonalds wireless connection. OK, so it's working through NAT elsewhere, and the server is also working fine. Your comment below makes sense to me, making this all the stranger. So problem should be due to ISP (ISP has fobbed us off to their specialist staff who are available Wednesday !) OR settings in my ADSL modem. If ArachNet offers a web-based control panel for your service, try visiting it and looking for "blocked ports". If there are any, turn them off, and see if that helps. If turning them off does not help, then I strongly suggest turning them back on, since port-blocking by the ISP will help make it harder for certain types of attack over the Internet to succeed. I personally wouldn't expect this to do anything, but it's the only way I can think of that your ISP could be affecting the connection. QUESTION :Does anybody know if we need special settings in the ADSL modem to run the VPN and remote Desktop. Some quick searches suggest that it should not usually be necessary. However, if your DSL router has a firewall built-in, you might want to see if it has any settings for "VPN", "Microsoft Remote Desktop," "port 3389" or "RDP". Also look for any logs the router may keep, and see if there's anything useful in them. Another thing you can try is, on the Windows laptop, run: Start->run type "cmd" and hit enter In the resulting window, type "telnet name-of-vpn-server 3389" . and see what happens. If you post the resulting messages here, it might be informative. To quit once it finishes, try just closing the window. -- Craig Ringer
Re: Dell 20" & 24" LCD Screens Going Cheap
Brett Carboni wrote: I have one of the 24" ones and they are fantastic. I use it with my 17" powerbook. Agreed, they're great displays. My only real issue with them is that they're too bright for a lot of colour work - even at minimum brightness. At max, they're feirce ... you'd have a really hard time working on one without sunglasses ;-) They also don't seem to support notifying the operating system when they're rotated, so you have to do display rotation manually using a control panel / keyboard shortcut. No big deal, really, especially if you don't rotate the display often. With this one you probably won't want to as the cables can get in the way. Very easy to set up. Just plug in and go. Had it for about 5 months & no problems. Paid ~ 1375. Can almost have 2 A4 pages side by side at 100% in Quark. Yep. I was more delighted by being able to have a full A3 portrait page on-screen at once, at full size. Wow. Alas, I don't get to use it for my own work (mostly programming and sysadmin with some web site development etc thrown in), since the display is at work and connected to one of the G4s. I have the feeling, though, that I could fit four or five nice large terminals on-screen at once, which would just be bliss. -- Craig Ringer
Re: Delay in Sys 9.2 startup
David Noel wrote: I've got a problem in restarting on my 450 Mhz iMac running system 9.2. Restarting is normal up to the stage where the Desktop and menu bar appear, with the Desktop showing devices and file icons correctly. However in the last month or so, folders in use do not open, and the system appears frozen and unresponsive. After waiting for about 2 minutes, however, in-use folders do appear, and I can proceed. Most of the macs at work do this. As far as I can tell it's related to network access - mostly, I think, connecting to the servers they're set to use on boot, and so on. I haven't found a workaround yet, which is very annoying when you need to reboot five times a day. -- Craig Ringer
WTB: Mac OS 9 compatible colour meter
Hi folks I'm seeking a Mac OS 9 compatible display colour meter such as an older Spyder. Unfortunately I'm having real trouble finding one new, and I'd be very glad if one of you folks who's been lucky enough to leave Mac OS 9 behind might be able to sell me one. I imagine somebody will have an old one lying around. If interested, I may be able to arrange advertising in trade, though obviously I could also just pay cash. -- Craig Ringer IT Manager POST Newspapers
Re: terminal App slow
Rob Phillips wrote: Over the last few days, it has been taking terminal a minute or more to respond to each command I type. Any ideas what causes this? Is it a 10.4.3 thing? Does it respond normally as you type, and only have a delay before executing the command and displaying a result? Now, here's a bit of a barrage of questions and things to try. Please reply with the results; it should help indicate what might be wrong or at least narrow it down a bit. Is there a delay if you run the command " : " (just type a colon, and hit enter) ? That command says "do nothing", but it's a shell builtin - it doesn't involve creating another program like most commands do. The same thing with a separate program is "/bin/true" - is there a delay when you run that one? What about "id" ? What's the output of the command "time id" ? What does the "hostname" command say (er... I think Mac OS X has a "hostname" command; if not, try "uname -n" instead) and is there a delay when you run it? A delay of a minute or so could easily be a timeout, probably a network timeout. Are you using NetInfo for distributed settings and user management? What about an Active Directory (Windows) Domain, NIS, or LDAP? I'm assuming that you've already checked to make sure the system isn't otherwise busy, and that it's only the terminal (or, at least, POSIX applications) that're slow. If you're feeling brave, you could try using a tool like truss or strace to watch what a program is doing, and see if you can tell why it pauses. If you like I can explain how to do that, but please try the tests I suggested above first. -- Craig Ringer
Re: dvd regions
Rob Davies wrote:
Return DVD and inform store of such an issue, although these days it
does not make much difference as most DVD players are not so
restricted. But they may have a region 4 copy of your disc, and yes
technically the shop is breaching some sort of copyright or something
similar.
While they might be breaching an agreement with their suppliers, I
really don't see how they'd be doing anything legally wrong. Not that
I'm a lawyer or anything. DVD regioning is a technical measure to create
artificial market segmentation. It's a business tool, and I'd be amazed
if there's anything illegal about selling "out of region" DVDs.
The scheme is enforced by contracts and licenses with DVD player makers
("To get a CSS decoding key, you must also suppport regioning to our
specifications") and vendors ("if you want us to sell you our stock, you
can't sell any out-of-region DVDs from any other supplier"). In some
countries it's backed up by piggy-backing on legislation that protects
"technical anti-copying measures" by relying on the fact that CSS is
both a copy protection technology AND a market segmentation device.
Sneaky, aren't they?
The upshot of this is that, AFAIK, it's quite legal to buy and sell
out-of-region discs, supplier agreements permitting. You can also play
them if you can find a player that'll do it, and if the local laws don't
consider the player a copyright circumvention device (which they most
likely won't - the player isn't what does the copying). Your challenge
isn't legal, but rather getting around the rather neat locked in
agreements the studios and the DVD CCA have with equipment vendors and
disc retailers. All I can say there is "Thank-you, China!".
--
Craig Ringer
Re: OT: M$ Access to MySQL
Paul wrote: Your other alternative might be to use the ODBC interface . potentially slow, unreliable Or as I wrote, " ... for Access's internal data store to poke into the Access database. ... " That snipped part changes the meaning rather dramatically. Most heartening indeed. I had a gut feeling that ODBC or even MyODBC connections might cause me premature ageing. Either approach will involve ODBC. The way you use MySql / PostgreSQL as a backend for access is through the ODBC driver for the database. With the other approach, your program uses the Access ODBC driver to interact with the Access database file. The main difference is that the MySQL / PostgreSQL ODBC drivers are made for fast, reliable, shared access either locally or over a network. Most of the real work is done on a proper database server, and the drivers just "translate" from the ODBC interface used by many applications to and from the native protocols the database servers speak with their clients. By contrast, my understanding is that the ODBC drivers used to interact with Access databases aren't really made for such robust use. For example, I don't know what the situation is with network access to Access databases, or if it's even possible from non-Windows machines. As far as I know, the ODBC driver essentially contains the program that reads the database file, potentially having to share access with (eg) a running copy of Access. Shared databases _can_ work well without a central server, but usually tend to scale poorly and can have reliability issues. The issues you'd run into sharing your Access database would be the same sort of thing as people who use MYOB over a network between several clients encounter - the more users, the slower and less reliable it gets. When you consider that a web based application could easily have several concurrent users... -- Craig Ringer
Re: OT: M$ Access to MySQL
Paul wrote: Hi all I have a pending project with the company I work for, if funding comes through. They have an internal Access db and wish to make this available to customers online for the purposes of ordering. My initial thought was to use PHP/MySQL. The problem is how to link or convert the data efficiently. Any ideas or thoughts? Have you done this perhaps? The only DB migrations I've done have been from an old FileMaker Pro version to MySQL, and later through to the considerably superior (and also free) PostgreSQL. This was a full migration, and involved re-building the user interface as a web application, so it's probably not what you're after. In this case, though, you might be interested to know that PostgreSQL (and I think MySQL) can be used as data "engine" back-ends for Access. Rather than using Microsoft Access's internal database engine (JET?), you use an ODBC interface to store the data in a real relational database system (MySQL, PostgreSQL, MS-SQL, etc) and just use Access for the user interface. The actual database server need not even be on the same computer (you could, for example, run it on your main server). With your data in a real RDBMS, it's stored in a superior database engine that's much more suitable for multi-user access, more easily backed up, and probably faster and more reliable. You can still use Access for the user interface, so as I understand it users who aren't actively modifying the database schema (creating, modifying, and altering tables and views) won't even know much has changed. On top of that, you can do things like have your web server talk to the database to expose that same data through a web application, use 3rd party and open source reporting tools to analyse your data, and more easily write quick SQL queries against the database (using tools like the `mysql' and `psql' command-line clients) if you just need to do some one-off analysis. In your case, being able to share the same data between Access and your web server would no doubt be the most attractive aspect of the configuration I've described. I haven't done this with Access myself yet (I have with other database tools), but probably should have a play with it since I have a copy of Access lying around. Here are some articles (straight off Google) on the topic: http://techdocs.postgresql.org/v2/Guides/Using%20Microsoft%20Access%20with%20PostgreSQL/view http://www.ucl.ac.uk/is/mysql/access/ http://66.102.7.104/search?q=cache:rZdNUWo1Wc4J:www.washington.edu/computing/web/publishing/mysql-access.html+%22Microsoft+Access%22+MySQL&hl=en&client=firefox-a Your other alternative might be to use the ODBC interface for Access's internal data store to poke into the Access database. I suspect that approach won't be too attractive though - potentially slow, unreliable, and almost certainly much less capable of dealing with larger numbers of users. I've seen similar schemes work very poorly indeed. -- Craig Ringer
Re: G5 slowing down
Rob Findlay wrote: General automatic maintenance is carried out by the various scripts which are controlled by Cron and if your mac is left turned on 24/7 it all gets done at the right times. I simply don't get this. Mac OS X is at its fourth major revision, and Apple still haven't put a tool like anacron in place to run scripts after the machine next starts up if it misses a timed job. This is _really_ not difficult, and it's also not hard to make it wait until the system load is low if they want to avoid disrupting the user. Does Mac OS X actually do this now, with the "cron isn't getting run" stuff being a holdover from older versions, or does it still not know how to run a missed cron job? If the latter, it's time for lots of Mac users to start filing bugs in Apple's reqest system for "use anacron or a similar tool to manage missed cron jobs". -- Craig Ringer
Re: Word to TeX
Pat Scott wrote: Hello, Does anyone out there use LaTeX? My husband need some advice. He is a newbie with a Macintosh - has had his 15" G4 Powerbook, OS 10.4.2, for only 6 months. He is writing a technical textbook, composing it in LaTeX as he goes along. He needs to be able to import images and mathematical formulas from Word documents. He knows there is Word-to-TeX software for pcs, but is there any for the Mac? Your best option for images might be to try to extract them from the Word doc where possible. I don't know if Word for Mac has anything as useful as the ability to simply drag an image from a Word doc to the desktop to extract it (I don't think the win32 versions do - they're what I use when I must use Word). You might be able to copy and paste images from a Word doc into an image program, then save the image. Alternately, I usually use OpenOffice to open the Word doc, then "save as HTML" - a horrific hack, but one that gets the original images out, usually without recompressing them and thus losing quality. Formulae are more difficult. I'd probably want to try to get the formula as an EPS that I could reference in the LaTeX document. If you're really lucky, the formula editor can save an EPS file or some format that can be converted to an EPS. Failing that, you might be able to print to PDF and either use the PDF directly, or crop it down and save it as an EPS in Acrobat Professional. I don't have access to Word right now, and I have no access to Word for Mac at any time, so I can't do any more. Hopefully those are a few ideas that might at least prove worth investigating. -- Craig Ringer
Re: Website download protection
Phillip McGree wrote: One option that you can do is to have the image in a table cell, but as a background image instead. Someone can still look through the HTML code, work out the exact address for the image and download it direct, but at least people won't be able to drag & drop the image to their own computer. That will stop most people. Indeed. However, since they're low-res and watermarked, a farly significant question might be "Why bother?". It's not impossible that any "protection" will annoy potential customers more than it helps you. For example, it's entirely possible that the customer is saving the images so they can print a contact sheet for someone who wants to pick which ones to buy. That's what I usually need to do at work, for example. Stopping them from doing that only harms you, since it reduces your chances of making a sale. The user-hostility aspect might not be too bad with simple means like table backgrounds, but it's certainly an issue with sites that resort to using Java or Flash "image display applets". When I've been looking for artwork in the past I've often just left those in disgust, as they've been slow, frustrating, and often sufficiently unpleasant to use as not to be worth bothering with. Even if you're just using the table background trick you still need to consider what you're actually gaining by stopping people saving the images. It's also important to remember that every time a user views the image, that's a "download". There's not necessarily any way to distinguish saving an image to disk vs displaying it in the page unless the user-agent explicitly chooses to send some indication in the headers. Moreover, most browsers will simply use the local disk cache if a user saves an image to somewhere, so most likely that won't even appear as a hit on your web logs. That's certainly the case with my tests (Apache 2, no cache control headers set, Mozilla Firefox). -- Craig Ringer
Re: PowerMac Specs and Pricing
Daniel Kerr wrote: Two dual-core 2.5GHz PowerPC G5 processors 512MB memory (533MHz DDR2 SDRAM) NVIDIA GeForce 6600 with 256MB of GDDR SDRAM $5299 I've never quite understood why Apple do this. They build a machine with fairly droolworthy CPU power, then give it less RAM than Dell ships with some of their bottom-of-the-range PCs these days. And Mac OS X undeniably wants a LOT more RAM than WinXP for good performance. The video card at least makes more sense. It's a lower middle-of-the-range card, but if you assume that most users won't need 3D and that that card's sufficient for maximum GUI performance in Mac OS X that's alright. The RAM, however, I just can't understand. Going to 1GB RAM as standard couldn't cost Apple more than $70 per machine (a _single_ stick of quality 1GB DDR2 is ~AU$150 retail; two 512MB modules come out much the same), and I'd be amazed if it didn't add 1/2 again to performance in lots of common uses. -- Craig Ringer
Re: 200GB iPod Nano - And now for something completely useless....
On Thu, 2005-10-06 at 10:17 +0800, Mark Secker wrote: > curiosity may be mankind's greatest tool ...but... does this not > defeat the purpose of a iPod Nano? > > <http://uncyclopedia.org/wiki/Ipod_Nano_200gb_Instructions> I imagine that's the point. Uncyclopedia is not what you might call a dead serious site. Have a browse around some of the other articles to get a feel for the site ;-) -- Craig Ringer
Re: Mac v PC
On Fri, 2005-09-30 at 14:03 +0800, Martin Hill wrote: > Some people argue that any Windows user not running AV software only has > themselves to blame if they get infected from a file passed on by a Mac > user, as here at my institution it only takes 7 minutes for an unprotected > PC to be compromised once connected to the network. However, it is a > courteous thing to do for a Mac user to run AV software, just not essential. Almost. I'd still be concerned about Word macro viruses, myself. Or is Office/X sufficiently incompatible as to make that a non-issue? -- Craig Ringer
Re: Mac v PC
On Tue, 2005-09-27 at 18:47 +0800, Rick Armstrong wrote: > Dear Mac Users, > I have a frustrating debate with a colleague who swears that a PC can do > anything that a Mac can do but cheaper and faster. Well, each platform can do the vast majority of things the other can. There are specialist domains where one beats the pants off the other, but they tend to vary as time goes by. As with most things, it depends on what you need. Somebody who insists that there's a single better option all the time, for everybody, is usually not actually considering that others have needs different to their own. I tend to side with your colleague on "cheaper and faster" in the hardware department. Software is less clear-cut. Anyway, it depends on the user. For someone who doesn't know how to lock down a Windows box and keep it secure, they'd better count their wasted time and repair costs in that price. Do note that BOTH OSes break in bizarre ways at times, and there's not much to be done about that. > Could someone give me a > short reply (PC owners can't understand anything more then a paragraph). If you're looking for reasons why you can't bring this fellow over to your point of view, you're staring at it. If you won't respect his views at all, why should he in turn listen to you? > I am in the graphics industry and use all Adobe software and QuarkXPress > software. Ditto. I'm also a system administrator, UNIX geek, and C++ & Python programmer. > I have no ambition to compare a Mac with a PC and the mention of > processing speed not clock speed doesn't get through. ... which is odd, since it's just as important for x86 systems these days. A 2GHz P4 is somewhat lame ; a 2GHz Opteron is very much not, and a 2GHz Pentium M is also pretty darn respectable. Anyway, for many tasks a decently fast disk and a whopping amount of RAM is more important than CPU power anyway. It's important to note that while clock speed isn't an especially useful measure of performance by its self, that doesn't necessarily mean that the lower clocked CPU is faster. All things being equal, the higher clocked one will still be faster. Clock is important, it's just not overly useful without knowing the real clock-for-clock performance of the CPUs in question. All evidence I've seen to date, for example, suggests that Pentium M is both faster clock-for-clock, and higher clocked on average, than the mobile G4 used in PowerBooks. A rather more useful measure for most people, anyway, is performance per dollar. Even then, who cares if all you really want is a nice small, portable laptop for word processing and 'net access, and don't actually care very much if it happens to get 20 or 30 frames per second in some program you'll never use. *drools over 12" iBook G4*. > I am starting to doubt the power of Macs now. From a Mac user forever. Well, don't stress - you'll get to use your favourite OS on what I suspect are cheaper, more powerful CPUs soon. Whether that translates to lower hardware prices remains up in the air, but that's not really why you buy them anyway, is it? It's not about "belief" or "doubt", it's about using what you prefer, and what suits your needs better. -- Craig Ringer
Re: Cheap colour laser printers
On Fri, 2005-09-23 at 19:36 +0800, Peter Curtis wrote: > Hi all > I'm getting really peed off with the cost of cartridges and ink for > the inkjet printers! > Can anyone recommend the inexpensive colour laser printers? Perhaps > give some information on brand, cost etc. Be careful. Many laser printer manufacturers are now "chipping" cartridges so they can't be refilled, and trying to shut down companies that work around those restrictions to refill them, just like inkjet makers are. This is especially true in their lower-end laser printers. All that means is that you can't assume a laser printer will be that much cheaper to run - you really need to check the actual consumable costs. Also remember to check the cost of service packs (new fuser, etc) and how frequently they're required. -- Craig Ringer
Re: ADSL modem
On Fri, 2005-09-23 at 08:21 +0800, Peter Hinchliffe wrote: > On 22/09/2005, at 10:52 AM, KEVIN Lock wrote: > > > A friend is connected to the net with Plusweb in Freo. He took > > their broadband setup despite being told they cannot help Mac users > > (what the!). > > They have supplied him with a 'Billion' USB modem and I have loaded > > the software for the modem, but cannot see how to change the mode > > of connection from Ethernet to USB connection. Anyone direct me on > > doing this or are Macs not USB compatible? > > > > 350mhz slot loader iMac plenty of memory. > > > > > > Just use Ethernet. There's nothing to change. The modem manufacturers > rarely provide Macintosh USB drivers for their modems, and in my view > it's just a waste of a USB port. The Ethernet port is not doing > anything else but connecting to a network, so it's the only logical > choice IMHO. Just enter the IP address of the modem into your browser > and you're away. Works every time... Additionally, using Ethernet means that you can avoid installing any extra software on your Mac. That means nothing to break when you upgrade your OS, no shoddy drivers to cause your mac to crash or your connection to work unreliably, etc. If you already use Ethernet for a home network, you can put the modem in there - but for security reasons, it's often better to get an add-in network card to keep your Internet connection on a separate port that you can firewall. You can get an extra NIC for as little as $12 from somewhere like Austin, and most cheap NICs are RealTek 8139 based, which I'm pretty sure Mac OS X supports without add-in drivers. If not, the driver is available on the RealTek site. If you aren't using your Mac's Ethernet port already, then you don't need to worry about that of course, and can just use it for the modem. -- Craig Ringer
Re: CMS for web site
On Mon, 2005-09-05 at 22:56 +0800, Shay Telfer wrote: > Free content management systems that are around are > MySource: <http://mysource.squiz.net/> > Zope: <http://www.zope.org/> > PostNuke: > <http://www.postnuke.com/modules.php?op=modload&name=Navigation&file=index> I also know of Plone (based on Zope), Mambo, and Drupal. There are probably many more. Note that depending on what you want, a web CMS may be overkill. It also introduces more security concerns than a static site does. Have you considered approaches like building the site from more basic content using a script? It's hard to know if a CMS is actually the right answer, or if your web designer just wants to hit every problem with the CMS hammer, without knowing your needs. -- Craig Ringer
Re: Conversion of Quark 4.1 file
On Thu, 2005-09-01 at 08:44 +0800, Mike Murray wrote: > Hi all > > I've received a small file from a colleague ( abusiness card) said to > be created in Quark 4.1. the file has the extension .dat Let me guess ... winmail.dat ? If so, your colleague most likely uses MS Outlook and has it incorrectly set up, so that it sends Microsoft's messed up attachment format to people who don't also use Outlook. The format is MS-TNEF . A quick search on VersionTracker reveals: TNEF's Enough - 2.0 http://www.versiontracker.com/dyn/moreinfo/macosx/8415 ... which should do the trick. InDesign CS should then import the resulting .qxd Quark file. I'd want to check the result, as InDesign's Quark import can be a bit patchy, but it should be alright. Failing that, if you send it to me I can make a PDF of it for you when I can get access to one of the Quark machines at work - probably early next week. -- Craig Ringer
Re: FS: MIDI and ADB/usb & graphics hardware etc
On Mon, 2005-08-29 at 18:00 +0800, Neil Houghton wrote: > > 1x A5 Wacom ADB graphics tablet with 1x iMate ADB to USB converter > > $30 (Will not split) > > Graphics tablet works under classic OS's only (will function under > > OSX but no pen pressure sensing). iMate can be used to connect old > > ADB devices such as software dongrals keys > > Just a word of caution here - do not assume that an ADB software dongle will > work through an ADB/USB adaptor. I had such a dongle to use Ashlar Vellum 3D > (CAD software) and bought such an adaptor (belkin from memory) - no go It is, however, well worth a try. I use an iMate ADB->USB adapter to handle the hardware dongle for some software the POST uses. It's entirely fuss free - it even works under Mac OS X (in Classic) with the new drivers, despite the original software being of Mac OS 7.1 vintage. Now, what I'd prefer to do is find the person responsible for deciding that a hardware dongle is a good idea and beat them with the dongle. Alas, in this case that's impractical. As for why I have such a problem with them: It's an artificial single point of failure in what is usually a key business system. They're a major cause of incompatibilities, faults, and other problems in otherwise fine systems and software, and I just can't imagine what sort of thought process leads to shipping them. After all, if you actually want to illegally copy the software you'll just tweak it so it "forgets" to look for the dongle - this is, after all, not all that hard. A hardware key is increasingly a strong factor in purchasing decisions here. I don't like being treated like a criminal who'll do the wrong thing if given the chance, and I don't like the extra point of failure hardare dongles introduce. I've already had to temporarily "fix" software until I can get a replacement dongle - it's tiresome, and something I'd much rather avoid. > > This is an incompatibility. The USB ADB adapter only works with ADB mice > > and > > ADB keyboards. It will not work with other ADB devices unfortunately. The same is true of the iMate out of the box, but Griffin provide a driver that lets it support more (presumably by providing a virtual ADB bus). Maybe it's worth a try? I think I have a spare here - if I can find it I might be able to loan it to you. -- Craig Ringer
Re: KVM switches any recomendations
On Thu, 2005-08-25 at 19:20 +0800, Craig Ringer wrote: > On Thu, 2005-08-25 at 18:26 +0800, Paul Sparrow wrote: > > any one using KVM switches, any recommendations? > > Not Belkin. Not unless they've fixed some *major* problems with their > mouse PS/2 emulation recently. Of course, since this a Mac list, if you're only using Mac gear you're rather unlikely to care about that. If you don't need PS/2 mouse support, then otherwise I've found them just fine. They also work fine unless you want to support more than two-button mice (think scroll wheel), so long as you're only using Windows systems for the PS/2 mouse. Cables were rather pricey last time I looked, but that may have improved. That'll teach me to forget which list I'm replying to ;-) -- Craig Ringer
Re: KVM switches any recomendations
On Thu, 2005-08-25 at 18:26 +0800, Paul Sparrow wrote: > any one using KVM switches, any recommendations? Not Belkin. Not unless they've fixed some *major* problems with their mouse PS/2 emulation recently. -- Craig Ringer
Re: Ozmac Mail
Dark1 wrote: Looks like my last email didn't go through Just reformatted my hard drive and did a clean install but I forgot to save a copy of the incoming mail server and outgoing mail server details for my ozmac email. If someone could give me the details I'd really appreciate it: Incoming mail server Outgoing mail server At a guess: mail.ozmac.com for both. Why: A little testing reveals that *.ozmac.com all point to 203.33.71.75, which is running services for POP3, IMAP, SMTP, and probably more (I didn't scan it, just checked the services I expected you'd need). The server claims to be 'mail.ozmac.com' in its SSL cert (IMAPs and POP3s, according to openssl s_client, and from the web page) so you'll need to use that for IMAPs / POP3s to avoid having your mail client complain at you. Sorry for the hasty response earlier, and I hope this is helpful. -- Craig Ringer
Re: Ozmac Mail
Dark1 wrote: Looks like my last email didn't go through Just reformatted my hard drive and did a clean install but I forgot to save a copy of the incoming mail server and outgoing mail server details for my ozmac email. If someone could give me the details I'd really appreciate it: Incoming mail server Outgoing mail server Actually, Google says that I'm an idiot who can't read. Sorry. -- Craig Ringer
Re: Ozmac Mail
Dark1 wrote: Looks like my last email didn't go through Just reformatted my hard drive and did a clean install but I forgot to save a copy of the incoming mail server and outgoing mail server details for my ozmac email. If someone could give me the details I'd really appreciate it: Incoming mail server Outgoing mail server Google says: http://broadband2.ozemail.com.au/things_you_should_know/support/ -- Craig Ringer
Re: DVD regional codes
On Mon, 2005-08-08 at 11:06 +0800, Lloyd White wrote: > I am helping a newbie Mac user and have come up against a problem I have not > encountered before. He was given a couple of Chinese DVDs with different > zone codes and copied them to his hard disk and got a warning that he had > only 2 uses more and he would lose the use of his super drive. He is now > worried. > > Is this use of a different zone recorded somewhere or is it in the firmware? It's in the firmware. > In other words can doing a clean re-install of the OS wipe out that record. No. > Or can we delve into the depths of the system and remove that record? Generally not, though it depends on the drive. What you *CAN* do, for some drives, is re-flash them with firmware that doesn't care about region codes. This isn't entirely safe, and you'd need to find firmware for the drive. There are also programs that run at the OS level that manage to make the drive overlook these issues. -- Craig Ringer
RE: iWorks
On Fri, 2005-08-05 at 12:40 +0800, Michael Schmidt wrote: > Hi. I've heard that Publisher docs can be opened in Photoshop. Anyone > know anything about that? News to me, but it'd be nice if it was possible. I suspect you might be thinking of PDF/TIFF/EPS files /exported/ by Publisher, rather than native Publisher documents. -- Craig Ringer
Re: Fwd: Exporting/Converting Pages Doc to CMYK
On Fri, 2005-08-05 at 08:11 +0800, Jon Davison wrote: > Hi Rod. You must convert your RGB images to CMYK in Photoshop. 'Image - > Mode - CMYK'. > Then go; 'Image - Adjustments - Selective Colour' to bring back some of > the colours that have > changed. CMYK will increase the file size as well. > > This is because images seen on computer monitors and TV screens using > RGB have a far greater range, or gamut of colours that ink on paper > can ever have. My understanding is that that's not necessarily true. Some truly awe-inspiring results can be achieved by things like die-sub printers, or high-res hexachrome printing on gloss stock. The issue is that there are parts of both the CMYK (and most other print colour format) and RGB gamuts that /do not overlap/. The diagram on this page might help: http://dx.sheridan.com/advisor/cmyk_color.html Thus, there are some RGB colors that can not be represented in CMYK - and vice versa. Since we usually design in RGB, it's the RGB ones that can't be represented in CMYK that generally bother people. Accurate monitor AND PRINTER profiles, plus tools that understand ICC and can provide out-of-gamut warnings will help a lot in avoiding unexpected problems when going to print. Remember that the gamut of your specific printer might well be different to that of the CMYK colourspace, too - that's part of the reason why accurate output profiles are crucial. > So Photoshop or another image editor is what you must use. Also make > sure you have 'bleed' > around your image if it is going to be trimmed. Your printer specifications will provide the required information for this. -- Craig Ringer
Re: Exporting/Converting Pages Doc to CMYK
On Fri, 2005-08-05 at 09:14 +0800, Rob Davies wrote: > > Hi Rod, > > > > I'm not sure about this, but I don't think Pages can output in > > CMYK and I don't think Preview can convert RGB to CMYK. Photoshop > > I think can. > > Problem here is that Photoshop will usually modify layout and fonts/ > type?? Nope. If the PDF was created correctly - with all fonts embedded - Photoshop produces a fantastically accurate and rather high quality rendering. As noted in my other post, however, it /is/ a raster rendering, with the associated potential problems. > The other option is to convert your images, logo's etc to CMYK and > adjust for loss, in Photoshop or similar type of program > (GraphicConverterX, Painter, Illustrator, Freehand). That presumes that Pages will preserve the CMYK images untouched. It's not impossible that it uses Mac OS image loading functions to load them in its preferred colourspace, then exports those converted versions. I'd be surprised, but I wouldn't make any assumptions until I'd checked the resulting PDF with a tool like PitStop. > Problem you will > have here is does Page allow you to re-import as CMYK? Then to Export? It darn well should. You might need to try different image formats though - for example, it's possible it'll handle CMYK TIFF but not CMYK JPEG. The only way to be sure is to test, and then preflight the resulting PDF. > Yes output as PDFX-3, Be aware that PDF/X-3 permits RGB. Additionally, most printers' tools don't understand ICC profiles, often discarding them entirely. This is bad if you're relying on them to achieve the desired colour accuracy. If they want a CMYK PDF, your best option is to give them a basic CMYK PDF. If you can force ColourSync to apply your profiles then output a plan CMYK PDF document, that'd be the ideal option. > This I am not sure will create what the printer is actually after, > usually requiring actual separated files I'd be stunned to have any printer ask me for explicit separations these days. That largely went out when CTP arrived (thankfully) though it's sometimes still necessary for hexachrome and other advanced printing. Even CMYK+spot can be handled easily in a single integrated PDF. If they want explicit PDF separations, you need a printer that's noticed we're not in the early 90s anymore. Their preflight, imposition, or RIP tools should do this for them to their specifications with no fuss at all. > If not specifically ask him what he requires. ... and send them a file to test BEFORE DEADLINE and ask them to preflight it and tell you if it's OK. If they say "huh, pre-what?" then find a new printer. -- Craig Ringer
Re: Exporting/Converting Pages Doc to CMYK
On Fri, 2005-08-05 at 07:12 +0800, Ronda Brown wrote: > I'm not sure about this, but I don't think Pages can output in CMYK Ouch. That'd make it almost useless for commercial printing. Surely there must be /some/ way to do it... > and I don't think Preview can convert RGB to CMYK. Photoshop I think > can. Be aware that Photoshop will rasterise your job. This will result in a **MUCH** larger file, and will lose certain information. In particular, the printer's RIP may be unable to optimise text and line art for clean and sharp edges. Depending on your job this may or may not be an issue. -- Craig Ringer
Re: iWorks
On Fri, 2005-08-05 at 06:58 +0800, Ronda Brown wrote: > On 04/08/2005, at 8:46 PM, Edward Arrowsmith wrote: > > > Is it possible to open Publisher files using the Pages component of > > iWorks please? > > > > If not, is there a program that will open publisher files in OS X > > please? > > > > thanks and Best wishes > > edward > > Hi Edward, > > No program on the PC or Mac, not even Microsoft Word, can open MS > Publisher files except Publisher. Your best hope is to save them as > Word using Publisher on the PC first. Even then, Publisher needs the original fonts and probably images. It might even mangle the document unless you use the same printer setup - unless they've finally fixed that. > I believe Publisher can export documents in a MS Word format ... badly, especially if the recipient doesn't have the fonts used in the document ... > or each > page as a single image. Pages can import both of these file formats > to varying degrees so you might wish to have the sender resend the > Publisher file in these formats. The best way to handle weird and broken formats like Publisher is convert them to PDF. It's essentially read-only, but PDF with embedded fonts is the only thing that'll reliably preserve the document layout, colours, and content. -- Craig Ringer
Re: MIGHTY MOUSE!
On Tue, 2005-08-02 at 21:46 +0800, Malcolm Burtenshaw wrote: > Apple have released their Might Mouse - http://www.apple.com/au/mightymouse/ If the next iBook models come with similar functionality, I might have to make good my long-standing claim that I'll buy a mac laptop the day that they have a decent built-in mouse. -- Craig Ringer
Re: drawing/grafix for panther/tiger
On Tue, 2005-08-02 at 17:17 +0800, Craig Ringer wrote: > On Tue, 2005-08-02 at 16:42 +0800, James / Hans Kunz wrote: > > hi guys > > panther/tiger are fast animalsbut is there a useful > > grafix/drawing > > package available > > (eg corel draw for os9 & osx.2) especially that can import corel draw > > doc's > > corel (& many others) stopped writing software for mac because apple > > makes to many changes when updating/upgrading the system The lower part of this page might also be interesting, though it's not specifically Mac oriented: http://www.maa.org/editorial/mathgames/mathgames_08_01_05.html You might well find that some of the apps listed there have Mac versions. -- Craig Ringer
Re: drawing/grafix for panther/tiger
On Tue, 2005-08-02 at 16:42 +0800, James / Hans Kunz wrote: > hi guys > panther/tiger are fast animalsbut is there a useful > grafix/drawing > package available > (eg corel draw for os9 & osx.2) especially that can import corel draw > doc's > corel (& many others) stopped writing software for mac because apple > makes to many changes when updating/upgrading the system Adobe Illustrator, for one. Not sure if/how it handles Corel Draw though. I'd also want to try out the SVG editor and general vector graphics tool Inkscape, which has a Mac OS X port: http://www.inkscape.org/download.php I haven't used it on Mac OS myself, but under Linux it's an impressive tool. The GIMP, a bitmap graphics tool, is also worth checking out: http://www.gimp.org/macintosh/ GIMP for mac appears to rely on X11. I'm not sure whether or not Inkscape does, but given that it's built on GTK+ my guess would be "yes". -- Craig Ringer
Re: .hqx
On Mon, 2005-08-01 at 17:52 +0800, Malcolm J McCallum wrote: > Can someone tell me please what I need to read a .hqx file. :-) .hqx is usually used to identify a BinHex encoded file. That's an old and inefficient way of encoding binary files into text so that they can be transmitted through stupid programs that convert line-endings of files even when they shouldn't. The most common example is email programs. IIRC it also preserves Mac metadata such as type/creator codes, and preserves resource forks. These days, most mail clients automatically encode attachments in base64/AppleSingle/AppleDouble, so you don't generally need to care about that. You can decode the BinHex file into its original file using stuffit expander, as noted by others here. You'll still need the program to read the original file, though - if it's .sit.hqx then it's a stuffit archive and Stuffit will take care of it. Otherwise, you'll still need to identify what program to use to open the decoded file. -- Craig Ringer
Re: Zip files and Zip files?
On Wed, 2005-07-27 at 21:00 +0800, Andrew W. Hill wrote: > I haven't quite figured out the problem yet, but I have also noticed > it. Additionally, I've found that some zip methods totally screw up > the file, even changing the final file size. A different final file size doesn't necessarily mean something is wrong. There are different compression "strengths" with zip, usually a trade-off between CPU time and file size. Additionally, there is more than one option for compression method, and different apps might have different defaults. > Three main methods are > using "Archive" from OS X, Stuffit's zipping, and gzip from the > terminal. These aren't actually the same thing. Gzip doesn't produce a zip file (confusing, I know) but actually a single file that's been compressed using (I think) the `deflate' encoding that's the most common one used in zip files. It's the simplest, but it's not a actually a zip file and some tools that know about zip files might not handle it. -- Craig Ringer
Re: USB phone
On Tue, 2005-07-26 at 20:38 +0800, Lloyd White wrote: > I have an iMac with OSX 10.4.2 and have been given a Planet UP-100 USB > Phone. > > I downloaded Skype and can make calls from the desk-top. However when I plug > in the USB phone I get no reaction. The dial keys do not do anything > although the volume control on the phone changes the Mac volume. I expect you'll need drivers. Google says: http://www.planet.com.tw/product/product_dm.php?product_id=363&menu_id=3 including: "Operating system: Windows 98 SE, Millennium, 2000, XP and Mac OSX" Unfortunately, there don't seem to be any manuals or drivers, at least not trivially found (looking for UP-100 on the downloads page returns zero results). -- Craig Ringer
Re: I Can't Access Apple Australia Or Apple US Or Apple UK
On Sat, 2005-07-23 at 13:02 +0930, Onno Benschop wrote: > Use traceroute and see what it says. I don't suppose Mac OS X comes with `mtr' or an equivalent in the Network tool? That'd be awfully nice. -- Craig Ringer
Re: CD won't eject !
On Wed, 2005-07-20 at 18:06 +0800, Craig Ringer wrote: > If it won't eject after restart, reboot again and hold down the mouse > button during boot. The mac should eject the CD. That's what I get for not reading all the backlog before posting. Sorry for the repetition. I keep on forgetting about mail clients that break threading. -- Craig Ringer
Re: CD won't eject !
On Wed, 2005-07-20 at 12:52 +0800, Stephen Chape wrote: > Hi folks, > > I have a CD that will not eject. > Tried eject key on keyboard ... nothing ! > Tried control click and get message "the disk is in use and cannot be > ejected". > No programs or applications are open. Maybe there are fonts on it that've been loaded or something. If it won't eject after restart, reboot again and hold down the mouse button during boot. The mac should eject the CD. Yep, that's intuitive design, folks :-P -- Craig Ringer
Re: Laser printer-fax-copier?
On Wed, 2005-07-20 at 12:03 +0800, Shay Telfer wrote: > IMHO, get a scanner and a laser printer. That way you won't have to > replace both of them when one breaks, or be without one while the > other's being repaired. Yep. You can also upgrade to a better version of one or the other more easily. > I think Harvey Norman had Lexmark lasers reasonably cheap last week. > No idea what they're like though. > > As for fax, there are numerous on-line fax options. Or pick up a > cheap fax machine. Or use ... *gasp* a fax modem. Yes, it's true - there still is a use for MODEMs. -- Craig Ringer
Re: Laser printer-fax-copier?
On Wed, 2005-07-20 at 11:26 +0800, Lara wrote: > Hi alll, I see inkjet multifunction centres have been discussed here > recently, but I can't see a discussion of laser printers. Looking for a > B&W laser-fax-copier. We have two AlBooks and an Airport Extreme base > station, can plug the powerbooks in directly to the printer when needed, > but one that will talk to the base station would be even better if this > doesn't cost vastly more. > > Price is a key point as is ease of use. The only one I've used fails on the "price" critera by about $10,000. It's a *fantastic* device though - it'll even scan a batch of documents and save them in a network file share for you. Like you, I've only seen inkjet multifunctionals on the lower end. I was utterly floored by how well one of them worked recently (a friend has one), though, and how quick and high quality it was. I'm a hardened inkjet hater - and if it can't tell me *in* *words* what's wrong with it on a status display, I don't buy it - but I was still pretty impressed. -- Craig Ringer
Re: Tiger on a iMac SE 400
On Fri, 2005-07-15 at 10:23 +0800, Bart Raffaele wrote: > Hi all > > I think I asked this question a while ago. not to sure > > Has anyone ran Tiger on a iMac SE 400 the translucent graphite model. > I've got one at the moment with it's ram of 128 13gig hd > Of course I'll be cranking the ram up as well maybe 1 gig. That much PC-133 SDRAM will be rather expensive. Combined with the external DVD-ROM (and probably a new hard disk - remember that in addition to space, the hard disk is crucial for performance), you might almost be better off thinking about getting an iBook. -- Craig Ringer
PowerBook G3 dead to the world
Hi folks I have a PowerBook G3 (Bonze Keyboard, FireWire) borrowed from work that I was hoping to drop Mac OS X onto and use for a slow-but-working Mac OS X build tester for the development work I'm doing. Unfortunately, the dratted thing won't boot. Even when the AC power is plugged in, it is entirely dead to the world. The battery status button does nothing, the power button does nothing. I've tried resetting the PMU as per the Apple knowledge base (this machine has a simple button on the back to do it) and have had no luck. It's been sitting idle for a long time - over 9 months. If the backup battery has run down, does that mean the system needs servicing before it'll even boot? (If so, permit me to grumble at Apple). Ideas? -- Craig Ringer
Re: Saving streamed M3U files
Paul wrote: Antony N. Lord wrote: I've been listening to a series of radio episodes at work with OS X and XP machines using Quicktime / iTunes or maybe even WinAmp (PC). One of these apps let me save the entire show (after it was downloaded) - I can't for the life of me remember which one! I suspected an update somewhere along the line has "fixed" this so you can't do it (so it always streamed and can't be saved). Anyone got any ideas? Failing that there's always WireTap or similar... Cheers, Antony. I know that old 2.X versions of winamp (the only other mp3 player there has ever been! ;) did let you 'pipe' the audio straight to disk instead of playing it, in other words it ripped mp3s. I cant remember if I tried an mp3 stream though. IIRC it wrote WAV data. Useful for temporary storage, not so useful for archival due to its size. There are tools out there that'll do that, and there are tools out there that'll write the raw MP3 stream to disk too. I can't remember names right now. Google might help, especially with the term streamripping (and variants thereof). -- Craig Ringer
Re: Shock and horror
Aurora wrote: It isn't the workers fault I know but they can't/or are not allowed to answer a question that isn't on their script. I don't mind spelling things or repeating answers but I would like them to be able to answer simple questions such as if I'll get a renewal notice. The whole "scripted support" thing amazes me. It's entirely beyond me what the people who put these policies in place think they're achieveing, other than angering their customers and ensuring their support staff will never gain skills so as to become more useful to the company. I can see that they might want to place clear boundaries on support, for example "no, we're an ISP, we won't help you fix your computer because it has eight million viruses on it. Here are a few companies you can call..." . They go a lot too far with this though, to the point that it becomes simply bizarre. I've called one big OEM, who I shall leave nameless, and been told "I can't help you with that, it's not on our script" then asked the same question a different way and got an answer. Hilarious. It's for just this reason that I've stayed with WestNet even though I can't get an 8Mbit service from them yet (and would like one). Their support aren't that great in terms of technical knowledge, but they will try to help you out or get you the information you need. As someone who generally calls for information and can handle the technical stuff himself, that's incredibly valuable. If WestNet imposed some sort of scripted support plan, their most of their value to me as an ISP would be gone instantly. Does anybody here know why companies put these scripts in place? I've never seen the logic behind it. Let's just hope Apple doesn't go down that path, since their support when I've had to deal with them has been adequate (if inferior in technical knowledge at least when I've called) and thus vastly better than any other OEM I've had the misfortune of dealing with. I've never heard them tell anyone to use a provided "system restore CD" that reformats the computer and erases all data without even prompting the user, for example - unlike one big OEM with a 3-letter name. -- Craig Ringer
Re: Shock and horror
Greg Sharp wrote: On 5/7/05 2:51 PM, "Rod" <[EMAIL PROTECTED]> wrote: And what happened to the jobs of those at the Australian call centre? Apple did the same thing to TAC (who service techs call for help). There were some great guys in there who were very knowledgeable and helpful :-( Its great that some Indians have work, but hopefully those who lost their jobs when all these call centres closed here have found work again. From what I have read, this is also happening in the US. And not only call centre work - programming and general tech jobs too. All to make that bottom line look better for the next day's trading on the NASDAQ ;-) Let's hope those who did lose their jobs found work again. Unfortunately most don't. The other big looser is the customer. I'm running a poll on our site on this very issue and although only about a half dozen votes have been submitted so far, everyone who has voted felt the quality of service from India is nowhere near as good as the Australian based support. I suspect that rather than anything inherent, this may be because companies who prioritize cost over service are moving to outsourced support first. Also, if the company is seeking to cut costs, the chances are that if they couldn't outsource to India they'd hire untrained 14 year olds here. (Of course, with computers an untrained 14 year old has a good chance of solving your problem, but anyway...). That doesn't make it any better for the user, of course. I am sure that it's possible to outsource support and still provide good service ... but that'd cost a lot more, especially since you need to hire people with actual skill in the language you're providing support in, or train them extensively in it. Blame the cost-cutting mentality, not India. -- Craig Ringer
Re: iBook 1.33GHz G4 Performance
On Thu, 2005-06-30 at 14:14 +0800, Craig Ringer wrote: > On Thu, 2005-06-30 at 13:07 +0800, Andrew Schox wrote: > > Hi all, > > > > Can anyone give me a subjective (or objective for that matter) > > opinion on the sort of performance boost I'd get from going from an > > 800MHz G3 iBook to a 1.33GHz G4? There's 500MB RAM in the old one, > > and I'd probably do the same in the new one. > > In a word - lots, though how much and what will depend on what you do. > I'm hardly a Mac expert, but I do use them a bit at work and I have > quite a lot of experience with hardware changes in general. OK. I didn't notice that your new machine was going to be an iBook - it was only mentioned in the subject. Oops. You can pretty much disregard my comments on disk performance - there might be a slightly better disk, but it's hard to say. RAM is probably easier to come by for the new iBooks than older G4s, too. The rest still stands, sans the comment about not being sure how cut down the mobile CPUs are (they're both mobile versions so it probably doesn't matter much). Re video hardware, do be aware that the iBooks come with lower end video hardware than the PowerBooks. I'd check to make sure it's even better than what your current one came with (though I sure hope it is). Ah well ... that'll teach me to double check the subject line. -- Craig Ringer
Re: iBook 1.33GHz G4 Performance
On Thu, 2005-06-30 at 13:07 +0800, Andrew Schox wrote: > Hi all, > > Can anyone give me a subjective (or objective for that matter) > opinion on the sort of performance boost I'd get from going from an > 800MHz G3 iBook to a 1.33GHz G4? There's 500MB RAM in the old one, > and I'd probably do the same in the new one. In a word - lots, though how much and what will depend on what you do. I'm hardly a Mac expert, but I do use them a bit at work and I have quite a lot of experience with hardware changes in general. >From my experience with mac and non-mac gear, your biggest gains will probably be from disk. The 7200 RPM 3.5" hard disk in the G4 should be a *lot* faster than your PowerBook's 5400RPM or 4200RPM disk. Disk performance makes a big difference for many tasks, and seems to be a major factor - perhaps the biggest - in boot speed and the speed with which many apps load. Consider upgrading the disk in the G4 to a nice modern disk (they're cheap, too) for more capacity and even more speed. Unfortunately, G4s take older PC100/PC133 RAM that's getting expensive now. 512MB should be OK, but I'd want more for Mac OS X personally. A fast disk will help offset slower RAM by making small amounts of swapping and a smaller disk cache slow things down less. I don't know how much the CPU will affect things. I wouldn't expect orders of magnitude, but 2x is entirely possible for CPU intensive tasks. It depends on how cut down the mobile G4s are. How much this affects you in real world use depends on what your workload is like. The other big factor to consider is your video hardware. With Mac OS X, video hardware is important. This is actually true of all OSes, it's just more obvious with Mac OS X. Anyway, if your new G4 has significantly superior video hardware you may see considerable gains in UI snappyness, etc. P.S: If that powerbook's for sale... I actually am looking for an affordable Mac PowerBook or iBook that'll run Mac OS X. I need a box to test Scribus on under Mac OS X. I'm not buying a new one just for development work when they're going to Intel soon anyway - I'll hopefully be able to use Mac-on-Linux on my desktop then, buy a developer edition of OS/X if they release one, or something like that. -- Craig Ringer
Re: Local Stores
On Fri, 2005-06-24 at 20:30 -0700, Andrew W. Hill wrote: > All, > I've been out of town for about 2 years and am looking for a place to > buy some PC100 SO-DIMMs for my iBook. Where's a reputable yet > affordable place in Perth? For generic items like memory that aren't Mac specific, I tend to use Austin (http://www.austin.net/) in Osborne Park. They still carry PC133 (you just won't find PC100 anywhere, but PC133 is compatible). http://www.austin.net.au/priceguide.asp?category=RAM Note that I'm in no way associated with Austin, except for being a happy customer. -- Craig Ringer
Re: Macusers out in the cold again!!
On Fri, 2005-06-24 at 10:52 +0800, Neil Houghton wrote: > Very interesting, I think! I agree. I've been hoping someone would take an interest in porting/improving wine/winelib for Mac OS X on Intel. > Sort of like VPC but completely different? (Wine is not an emulator) Completely different. VMWare and as far as I know VPC emulate a whole "virtual" computer, including CPU, disk, etc. They run the Windows OS in its entirety on that emulated hardware, then run programs with that OS. Under PowerPC they must emulate the CPU, but when running on Intel they'll be able to virtualize it instead (WAY faster, but still quite a bit slower than not emulating/virtualizing). Wine, on the other hand, is a "compatibility layer" that sort of translates Windows program formats and API calls into something that the host OS (eg Linux, and soon Mac OS X) can understand and run. No emulation involved, so it's way faster, but it can't run all the programs VPC can. Technically, wine: - Provides facilties to understand and load the Windows binary format (Windows doesn't use Mach-O or ELF, but its own format). - Implements enough of the Windows API (the "language" programs use to ask the system to do things) for a majority of programs to run. This includes the Windows graphical user interface libraries, file system access, etc. - Tries to make its implementation of the Windows API "translate" to similar native calls where possible. Good examples here are file read/write calls, which translate very well. The GUI doesn't translate as well, so Wine has to implement a lot of the Windows GUI layer. - Makes everything look to the program as if it's still running on Windows. > Does that mean you get to use your Windows applications without having to > pay Microsoft for either VPC OR any incarnation of Windows? Yes, since Wine doesn't use any parts of Windows. This is, however, only *if* the application will run under wine. Some won't, because they require kernel extensions / drivers to be installed (games with copy protection are a big one for that, but support for them is being added currently) or they rely on parts of the Windows API that Wine doesn't know how to translate. In general, best results will be achieved if the company that makes the program does a small amount of work to tweak their program so it'll run well under Wine, or a bit more work to convert it to use WineLib (permitting it to run stand-alone). Alternately, a program can often run well if the Wine developers have spent some time tweaking Wine to work for that particular program. Lots of the things Wine works well with now are things Mac users won't care as much about - MS Office for Windows, Adobe Acrobat, Adobe Photoshop, etc. However, I can see MYOB and Quicken support being really popular requests (I don't know - Wine might be able to run them already). > Does that make it better/worse than (or just different to) the VPC/windows > solution - which I have found to not be a real solution for me - sometimes > things that work after a fashion but are too slow to be actually usable - > other times I just can't get them to work! Different. Some things that work with VPC won't with Wine (VPC provides a *full* Windows API and can load drivers) and a few things that don't work with VPC will with wine. The main advantage of Wine is that you just run the program - you don't have to load up a whole emulated Windows OS first. It's much more memory efficient, and you don't have to worry about having essentially two OSes running. You also don't have to pay for a copy of Windows, directly or indirectly. It should even be possible to let you just double click on the .exe, or or even make the Windows app into a .app bundle. The speed issue should be partially solved by the move to Intel anyway. With that move, VPC won't need to emulate the Intel CPU anymore - instead, it can let the emulated OS run on the real CPU, and only has to emulate some instructions. That's part of what's called virtualization - you'll be hearing that term more and more I suspect. Despite what some folks have said, this does NOT mean that the emulated OS will run at full speed - even VMWare takes something like a 30% peformance hit, and I doubt Virtual PC will be able to do even that well. It's a heck of a lot faster than full emulation like VPC currently has to do, though. Wine will not have to take that 30% performance hit. There is some performance cost from the additional "wrapper" the program runs within, but not all that much. Programs do tend to be slower to start up, though. To give you an idea of performance under Wine, modern games that run under Wine are generally quite playable, only suffering a small performance hit frequently estimated at less than 10%. -- Craig Ringer
Re: Macusers out in the cold again!!
On Thu, 2005-06-23 at 16:29 +0800, Matthew Healey wrote: > > Yeah. Unfortunately, there a few reasons developers use DirectX that > > aren't easily written off: > > > >(a) hardware manufacturers tend to do a better job of their DirectX > >drivers and support on their consumer cards; Quality OpenGL is > >relegated to pro models. > > > > Does that mean that all this time, Quake 3 and Doom 3 based games > actually haven't been working on the PC? No, of course not. It *does* mean that Id had to spend time working around and/or fixing OpenGL issues, identifying bad drivers, etc. Also, remember that Id has the market power to go to a graphics card manufacturer and say "if you want your card to be listed as supported by Doom 3, fix this bug" - and have the card manufacturer jump to comply. Id have always stuck with OpenGL. They're also one of the prime movers behind OpenGL 2, and possibly the biggest factor in finally getting it out the door as a finished standard. I strongly suspect they're what's kept it alive on Windows for consumer cards, and I wouldn't be surprised if they were one of the main factors in growing manufacturer support for OpenGL 2, either. > >(b) Until recently, OpenGL sucked compared to recent DirectX > >versions. Only with OpenGL 2.0, which is still only now seeing > >some adoption, is this rectified. > > > > Ahh how things change. Not long ago it was DirectX that sucked like a > vacuum. Don't get me wrong, it still has its problems. Lots of them. However, when developers can do something more quickly and efficiently using new DirectX shader magic, that's likely to be an attractive option. Sure, they could probably use an OpenGL extension, but only if they can confirm it's widely - and correctly - supported by cards and drivers. With OpenGL 2, things should hopefully be on more of a level regarding technical capabilities. > Well the newly released NVidia 7800 supports OGL2 out of the box. I noticed that - and it makes me rather happy. -- Craig Ringer
Re: Macusers out in the cold again!!
On Thu, 2005-06-23 at 14:37 +0800, Matthew Healey wrote: > On 23/06/2005, at 12:56 PM, Craig Ringer wrote: > > > That doesn't mean you could run DirectX apps native, but it does mean > > that it may be possible to use the WINE layer to run DirectX windows > > apps. Developers may even be able to use WINElib to compile tweaked > > versions of their Windows DirectX apps into something that runs out of > > the box on Mac OS X. They're potentially much more likely to do this > > than port, since it'd be MASSIVELY less work. > > Alternatively, developers could get a clue and just use OpenGL to > begin with and then not have to worry about these sorts of issues. Yeah. Unfortunately, there a few reasons developers use DirectX that aren't easily written off: (a) hardware manufacturers tend to do a better job of their DirectX drivers and support on their consumer cards; Quality OpenGL is relegated to pro models. (b) Until recently, OpenGL sucked compared to recent DirectX versions. Only with OpenGL 2.0, which is still only now seeing some adoption, is this rectified. Hopefully if OpenGL 2 starts seeing good driver support etc we'll see a move back toward OpenGL. Especially if it's not left to stagnate for five years again. -- Craig Ringer
Re: Macusers out in the cold again!!
On Thu, 2005-06-23 at 12:13 +0800, Gordon Campbell wrote: > I doubt we'll ever see WorldWind on Mac since it makes very heavy use of > DirectX, which is Microsoft's way of providing better hardware access for > these sort of apps. Yep. That makes a true native implementation unlikely. > We're never going to see DirectX for MacOS since it's a > totally Windows thing Now that, it's not safe to assume. I've played DirectX games under Linux ;-) . WINE has technology to support DirectX, and it's improving all the time. There's also a commercial wine fork called Cedaga with even better directX support. The implementation uses OpenGL to render the DirectX requests, and is likely to be ported to Mac OS X if the rest of WINE is. That doesn't mean you could run DirectX apps native, but it does mean that it may be possible to use the WINE layer to run DirectX windows apps. Developers may even be able to use WINElib to compile tweaked versions of their Windows DirectX apps into something that runs out of the box on Mac OS X. They're potentially much more likely to do this than port, since it'd be MASSIVELY less work. > so even after the move to Intel, we're no more likely > to get WorldWind or other apps. Getting WorldWind to work would involve > almost totally re-writing the app to use OpenGL or Core Image instead of > DirectX. Not necessarily, as above. > I think it's highly unlikely that the move to Intel, in and of itself, will > draw more developers etc. to the platform. I tend to agree there, though I expect the number of quick'n'dirty ports using things like WINELib may well increase. If Apple decides to release developer builds of Mac OS X for stock x86 hardware (or perhaps an emulator like VMWare with a bundled Mac OS X deal) that might help. I'm not holding my breath. -- Craig Ringer
Re: Macusers out in the cold again!!
On Thu, 2005-06-23 at 11:51 +0800, Neil Houghton wrote: > Read an article on NASA's Worldwind > <http://worldwind.arc.nasa.gov/> > - it sounded great until I visited the site to find: > > > System Requirements > > > > *Windows 95, 98, ME, 2000, or XP > > *Intel Pentium 3, 1 GHz, or AMD Athlon or higher > > *256 MB of RAM > > *3D Graphics Card > > *nVidia GeForce 2 Ultra > > *ATI Radeon 7500 > > *Intel Extreme Graphics 2 > > > > > > *DSL / Cable connection or faster > > *2 GB of disk space > > > Will Apple moving to Intel do away with this discrimination No*. > or will we still > have to use Windows PC to access things like this (and the DLI Skyview WA > and the Commsec Advanced Trading Program, and, and ) Yes you will*. * Unless the move to x86 encourages improvements to tools like WINE etc to the point where you can transparently run Windows programs much of the time. This is a big maybe. The problem is, and will continue to be, the Mac OS X API. The CPU architecture doesn't matter much. The problem isn't getting the program for PowerPC, it's getting a Mac OS X port of the program. This will not change, though porting might get easier if tools like WineLib are improved. -- Craig Ringer
Re: Directory
On Mon, 2005-06-20 at 17:20 +0800, Skehan Adrian wrote: > Can anyone who knows of an application that one can use to print a > directory/list of all file and folders on a hard drive please. If you don't mind the format being pretty simple, you can use: find / -ls or find / -print (depending on what formatting you want) from the Terminal application. To save the output to a file (for later printing from a text editor) you can use: find / -ls > all_files_on_disk.txt -- Craig Ringer
Re: Copying from .dmg files
On Sun, 2005-06-19 at 17:03 +0800, Severin Crisp wrote: > I find that when copying from Disk Image (.dmg) packages the transfer > is initially fast but after about 3MB it slows right down. As it is > essentially a disk to disk copy I can see no reason for this, after > all the disk image has been decompressed, opened and mounted by that > stage - or has it? I'm not sure it has; it may well be decompressed on the fly, as it's read. Anyway, a possible explanation for the behaviour you're seeing is disk cache. Your OS and/or disk may buffer and cache writes, so that the hard disk may not actually start writing for the first few megabytes of the copy. This permits uninterrupted, and thus much faster reads. Once it starts writing, it has to write a chunk, go and read another chunk, write it, and so on. This is just a guess, but matches my experience on other platforms with modern disk subsystems. -- Craig Ringer
Re: Mac Unix Books
On Wed, 2005-06-15 at 09:40 +0800, Shay Telfer wrote: > >>Good morning group, > >>Is there any specific reference books for Mac Unix > >>that could be recommended for a novice? > > Don't forget The Unix Haters' Handbook: > > <http://research.microsoft.com/~daniel/unix-haters.html> > > Follow the link to the free download. It's really not funny unless you've used UNIX for a while and really got to hate some of its quirks. It's terrifying, however, how much of that will strike a chord with a user of a "modern" UNIX system even now, many years later. It's also amusing to see how far we've come. -- Craig Ringer
Re: World of War Craft
On Tue, 2005-06-14 at 17:38 +0800, Toby Oldham wrote: > P.S. Craig, you can say the game is frustrating - but to call it > boring assumes that the only joy derived from the game involves > killing stuff (which incidentally, is only 80% of the gameplay *heh > heh*). Nope, actually I count that as part of the boring in World of Warcraft. Just different ideas of what's interesting, I guess. Anyway, this is waaay OT now... -- Craig Ringer
Re: World of War Craft
On Tue, 2005-06-14 at 14:00 +0800, Dark1 wrote: > You guys are scaring me. Makes me wonder if I should open pandora's > (Blizzards) box. Don't do it man. It's boring, it's ugly, it's frustrating (going by the constant stream of expletives coming from my on-gaming-shift housemate), and it's somehow totally, hopelessly addictive. You know one of the older games in this genre, EverQuest, is frequently referred to as "evercrack," right? -- Craig Ringer
Re: WoW addiction Re: Worrying too
On Mon, 2005-06-13 at 08:35 +0800, Tobes wrote: > [raises hand somewhat shamefully] > > I'm on the Icecrown server ... I generally play fairly intensively > for 2 weeks, then leave it for 2 weeks. Then you're doing better than my two housemates, who play in shifts for 8-12 hours every day. Terrifying. -- Craig Ringer
Re: WORRYING
On Sun, 2005-06-12 at 12:14 +0800, thefrogs wrote: > I am running 10.4 with broad band. > I scanned some images but couldn't find them the vue scan program had > also to be reactivated from my serial number. > Then BTV Pro icon on the dock was lost. > I searched for my images and they were saved as aha.tiff in a folder > called Greek Holiday in my users folder as a new user. Other programs > have had to have their serials reinstalled. I feel I have been > hacked. That sounds considerably more likely to be a problem with your install of MacOS/X, your disk, or your filesystem. MacOS X, by default, won't even accept incoming connections from the outside world, and a hole in its firewall would be reasonably big news. Unless you have intentionally disabled that firewall, or opened up some ports, I would think it unlikely that an attacker could reach you. Even then, they'd have to find an exploit, or successfully determine your ssh password (if you opened ssh) to get in. That's not to say that your theory is impossible, just that I'd consider it unlikely. A badly behaved / badly written program, or just MacOS X going a bit insane, is way more likely. -- Craig Ringer
Re: Another viewpoint on why Apple went Intel....
On Fri, 2005-06-10 at 20:39 +0800, Rod wrote: > <http://www.pbs.org/cringely/pulpit/pulpit20050609.html> > > Probably malarky, but he does have an interesting point. That man is a genius. He should be in comedy, because he's evidently good at it. If he's actually serious, he should probably stick to Slashdot. I haven't seen quite such an out-there conclusion drawn from so little evidence - some of it wrong - and so many assumptions in quite some time. Great stuff ;-) -- Craig Ringer
Re: Interesting article on cpu switch
On Fri, 2005-06-10 at 18:36 +0800, Stephen Chape wrote: > How did you manage that Kev ? > I have advised every person that has asked me to buy a Mac. > So far every one of them has bought a Windoze PC, then come to me for > assistance. That's very interesting. I usually just ask them a few things about what they do and need, and suggest what I think sounds most sensible based on their needs and technical knowledge (ability to deal with viruses and security issues, mostly). That's often an eMac, iMac, iBook, or Mac Mini, but sometimes an XP box if they need some software that's Windows only, etc. Most folks end up buying what I suggest, and I usually get good feedback afterwards. When they don't, it's often because they ended up buying a mac laptop instead of a Windows laptop and finding an alternative to the software they needed. Maybe the fact that many of these people are workmates who I support day-to-day, or family who see me as the "family geek," may have something to do with it. I also don't try to give them a "sell" on a particular product, but rather evaluate what might be better for them. That probably helps - I'm not seen as pushing my own preference. Rather the opposite, actually ;-) . > However my brother has > already told her that a Mac would be much too difficult for her, "It's > a specialist machine". Of all the potential arguments against getting a Mac, I wouldn't have thought that to be one of them. WTH? That's *never* been true, unless maybe you count the old A/UX systems. -- Craig Ringer
Re: FTP through Tiger Server Firewall problem
On Fri, 2005-06-10 at 16:56 +0800, Martin Hill wrote: > Apples docs at this detail are pretty poor :-( Yeah. The FTP protocol is, frankly, pretty poor too. It's just not made for the modern Internet, with its hideous mess of intentionally non- standards-compliant software (firewalls) and bizarre routing like NAT. That said, I agree it'd be nice to have some better documentation on how to get it working properly. -- Craig Ringer
Re: Changing to IMAP
On Fri, 2005-06-10 at 09:34 +0800, Antony N. Lord wrote: > Anyone got some recommendations / comments before I do the switch? IMAP is bliss and you won't look back, though you'll find you need to pick your email client more carefully. One of the more reliable ones I've seen is Mozilla Thunderbird, though I personally use Evolution (not for Mac). I'm unimpressed with Mail.app's IMAP support on 10.2 but haven't tested it on 10.3 or 10.4. You do need to find an IMAP server that will give you a large quota, preferably one that also keeps good backups if you care about your mail. With IMAP your mail is actually kept on the server, you see, though you an usually set your client to download a complete copy. This is problem for me as I run an IMAP server for work, but this can make your ISP's email service unattractive both in capacity and backup quality. One alternative is to use POP3 with "leave mail on server". This works OK, but you don't get to share your read and replied status flags, message labels, etc. You also can't have the server filter your email into mailboxes before you even download it like you can with IMAP - for me that's a MASSIVE problem. Note that not all IMAP services will let you do server-side filtering, though. -- Craig Ringer
Re: Interesting article on cpu switch
On Fri, 2005-06-10 at 20:51 +0800, KEVIN Lock wrote: > In my experience, people about to get into a computer for the first > time look to a friend to ask if that friend can be their techie when > they purchase.In my case, I have advised a few friends that if > they want my help then they should by a Mac. They did and I am. Same here. The specs don't mean much to most folks. I often give them a few helpful tips for coping with the specs. In many cases, though, I just advise them to get a Mac. Not liking them much myself doesn't mean I don't think they're way better for most users, especially the less technically skilled who won't know how to keep themselves virus free if they use Windows, etc. -- Craig Ringer
Re: 2 main reasons against Mac
On Fri, 2005-06-10 at 09:28 +0800, Martin Hill wrote: > > 1. "Macs are too expensive", > > Indeed, that's what I put down as my 3rd point and with the advent of the > Mac Mini and the potential for cheaper intel components (some more > expensive, some cheaper), this also is being better addressed by Apple. The mac mini sure does help in that regard. I *really* wish it had gigabit ethernet (it doesn't cost a noticeable amount more for gigabit NIC chips now), though. > >or "Mac prices would be OK but MacOS is waaayy too expensive." > > I don't get that one much. Are all your friends linux users? ;-) That, and Windows users. OEM XP Pro = ~$260 ... and nobody in their right mind buys anything else. You won't be buying annual upgrades, either. I think what bothers me personally, though, is the lack of upgrade pricing, and the fact that the site license is really pretty unattractive for <10 machines. Of course, you do get a significantly better system for your extra money with MacOS/X. These days it's almost fair to add the price of a virus scanner (and update subscription) to your Windows license cost, too. I guess I just miss the days of $100 MacOS upgrades every couple of years. > Well it depends if you'd prefer to not get a new OS with lots of nice new > features every year and a bit or would rather wait 5 years for such new > capabilities and then pay more. Well, personally I prefer to get one every six months and pay nothing, but I'm a Linux user. I do "pay" quite a bit in donated time spent doing development work, but that's fun. You also "pay" in time spent stuffing about, but IMO not more than any other OS for many tasks now. For the systems at work, I'd be much happier getting a new OS every three to five years and paying much less on average over that period, as I do for the win32 systems, yes. Annual upgrades are frustrating, especially when every single one breaks something. I do appreciate the fact that you can skip MacOS/X releases. Going 10.2 -> 10.4 etc makes things much more reasonable, though IMO still a bit steep. > Personally I prefer the former. Because of > Apple's 10.1, 10.2, 10.3 naming scheme and Microsoft's glacial pace bringing > out Longhorn, too many PC users get confused thinking Apple's new OS > releases are equivalent to service packs on the PC. They aren't, I know. Sometimes I almost wish they were a bit more like service packs, though. Also, new OS or no, if you just want the basics working and stable it's not really attractive. I usually get each release and wonder "what will this one break?" not "what goodies do I get in this one?". That said, from a developer's PoV the rapid releases are actually wonderful. Apple *do* fix things, especially in their development environment, and this is very helpful. I'd be sooo annoyed to be stuck with an incomplete set of CUPS libs and headers on 10.3 for the next 3 years... > > There's no "OEM escape" like with Windows, > > Do you mean Windows being bundled with a new computer? That helps given how rarely it's updated. It's also generally possible to buy windows "with a new computer" and stretch the definition of "new computer" quite a bit ;-) . Potential future Mactel users who might want to dual boot (eeew, I know ...) or emulate Windows may wish to remember this. > >> 1. Large software choices - Macs running Windows software at around native > >> speeds [...] > > > > Be careful with this. [snip] > Come on now - you would surely have to agree that it *will* get > significantly better than the current slow emulation on PPC situation?!! Yep, if nothing else it's extremely likely there's going to be some improved emulator/virtualizer like VMWare out there. It's not entirely safe to assume WINE will make it over, though I think it's pretty likely. All I'm saying it that right now it's probably wiser not to tell people these things *will* happen, unless someone who's going to do it steps out and says so. -- Craig Ringer
Re: 2 main reasons against Mac
On Fri, 2005-06-10 at 00:39 +0800, Martin Hill wrote: > What are the two main reasons PC people state over and over why the Mac > isn't a good option? In my experience, they boil down to: > > 1. Not enough software for the Mac > 2. Macs are too slow Really? The top two I've had mentioned: 1. "Macs are too expensive", or "Mac prices would be OK but MacOS is waaayy too expensive." 2. "I can't run [blah program I need] on MacOS"; or "To run [blah] on MacOS I'd have to emulate Windows, so why bother." I happen to feel the first one rather strongly myself. I don't mind paying for Mac hardware, but paying regularly for MacOS galls me. I know, Apple's always done that, and there's some protection offered by AppleCare, but I still find it very annoying. There's no "OEM escape" like with Windows, and the Apple "service packs" are full and expensive OS releases. Anyway, enough of that particular grumble for now... > 1. Large software choices - Macs running Windows software at around native > speeds [...] Be careful with this. So far there is no solid evidence it'll be possible to run any given Windows software package without running or emulating Windows as well. Yes, you probably get near-native speed depending on the app, but you'd still need VMWare or similar and a Windows license, or need to dual-boot. We don't even know absolutely for certain that you'll be able to dual-boot, either. It's possible that better options like a cleaned up and ported WINE will emerge, but it's not safe to assume that. For now, speaking with a clear "might" or "may" in there is probably wise. > [...] as well as X11, unix and ported linux software and of course Mac > software.(1) > > 2. Macs running processors with the same number of GHz as the competition at > last.(2) If Apple go for Pentium M, then I'm afraid that will only be the case if the P4 never makes it above 3.6GHz . Pentium M is clocked slower than the P4, but does much more work - kinda like a PowerPC CPU, or an AMD Athlon XP. This whole issue matters less now that Intel uses "model numbers" not GHz ratings, and now that AMD has been helping get the GHz message out too lately. Even Intel is behind the marketing push about clock speed not being important now. [snip] > (2) It'll be interesting to see if there is a significant performance > overhead which results in OS X running slower than Windows on the same GHz > Pentium chip. With all the Quartz eye-candy this is a possibility, but of > course Longhorn with it's own eye candy will be out sometime along in that > timescale so it may end up being a wash. We'll just have to wait and see. Given that people complain about Linux GUIs being slow, I wouldn't be surprised if MacOS took the same flak. MacOS GUIs tend to be smoothly updated, but IMO somewhat slow and more importantly sometimes VERY unresponsive to user input. Safari and iTunes come to mind regarding the last point. Of course, it probably wouldn't take Apple all that much work to fix that responsiveness problem, at least for their bundled apps. The GUI as a whole is fantastically responsive, it's just the UI within applications that's a problem. All it takes is for a button to go down then up again when clicked or the app to show some other response while it's working away, even if the actual requested action doesn't happen instantly. There's nothing worse than hitting stop in Safari, wondering if it's actually going to do it, and hitting stop again when you begin to think it won't - only to find that it then stops the page, changes the stop button to reload, treats the second click as reload, and begins loading the page from the start again. *FOAM*. -- Craig Ringer
Re: Interesting article on cpu switch
On Thu, 2005-06-09 at 20:00 +0800, J Philippe Chaperon wrote: > Well, it could also be like a BMW with an Excel or Kia engine!! How about a BMW (complete with expensive and incompatible parts *grin*) with a powerful but rather unpredictable and very beat up looking rocket motor strapped to the top ;-) -- Craig Ringer
Re: More ppc-intel queries
On Wed, 2005-06-08 at 13:23 +0800, Brett Carboni wrote: > > - It sounds like we're up for a(nother) software transition for > >sociological reasons: > > I think it may be more like economic reasons. > > Steve's vision for Apple in the future may be to be an opposition to > WIndows. Maybe... he has to worry about the fact that Apple *needs* Microsoft for Office, though. At least for now. > That way his share of OS sales will go from 3% to perhaps 48%, a > 1600% increase. Imagine being on-line buying a Dell and clicking a > Mac OSX Cheetah check box :-) It's logical to suppose that Windows > does need an opposition. Yep. The risk is that if they encroach too far on MS's territory they might be seen as a threat. That's probably not wise. > I'm not au fait with Carbon, Blue/Red/Yellow box, Fat/Thin/On-a-diet > binaries etc but I'm sure it eventually could be done. The only > question I would have is "can it be pirated?" And I'm sure there > would be other problems, but not insurmountable. I sincerely hope, for Apple's future, that it *can* be pirated. Why? Well, what better way to gain mindshare and increase the user base buying applications than to make it officially impossible - and against the license - to run MacOS/X on standard x86 boxes, but ensure that hacking it to work is never too hard. You keep your business sales and "joe avergage" home user sales, but more and more exclusively win32 users are exposed to MacOS/X. My personal suspicion is that they won't be able to /stop/ people running MacOS/X on standard hardware once they release the x86 port. It might have to be done via a virtualization layer akin to Mac-on-Linux, but I bet it'll happen. -- Craig Ringer
Re: More ppc-intel queries
On Wed, 2005-06-08 at 12:11 +0800, Craig Ringer wrote: > On Wed, 2005-06-08 at 11:52 +0800, James Devenish wrote: > > > - I had a look at the Anantech article and Apple's migration guide. > >Gems include: Mactel will not use OpenFirmware; Rosetta has a whole > >heap of limitations in its support. Great, what are the prospects > >that current Netboot, Office 2004, Adobe CS, and all the smaller apps > >and utilities, are going to work in future? > > For Photoshop CS and Office 2004 - pretty good. I don't see anything in > that list that precludes them working. As for netboot - nfi. Some more info: http://www.adobe.com/products/photoshop/systemreqs.html says: Macintosh * PowerPC® G3, G4, or G5 processor * Mac OS X v.10.2.8 through v.10.4 (10.3.4 through 10.4 recommended) * 320MB of RAM (384MB recommended) * 750MB of available hard-disk space * 1,024x768 monitor resolution with 16-bit video card * CD-ROM drive * Internet or phone connection required for product activation *
Re: More ppc-intel queries
On Wed, 2005-06-08 at 11:52 +0800, James Devenish wrote: > Also, Martin Hill mentioned some reactions being "emotional", rather > than "logical". That's entirely true, but my own reactions are rooted in > logical problems, too. The contrast between the time I spend with > PC-type hardware vs other hardware (BIOS being the obvious show-stopper) > means my emotional reactions are based on bitter (perhaps overly) > experience. Yeah... since it looks like Apple may use the PC BIOS, that's a worry. Modern PC BIOSes are much better, but nowhere near ENOUGH better. Personally, I was hoping they'd stick to OpenFirmware. If they're going to make their hardware incompatible (at least in that MacOS/X will require a mac), they may as well do it properly ;-) I guess they must want the option of dual-booting windows, or may want to reduce dev costs. I'd still be happier with OpenFirmware plus a BIOS compatibility layer, like what Intel did with EFI. I've inserted my comments below, in case my blather may be of some interest. > - My first question is: Do you think Mac OS X 10.4 and 10.5 will be >installable as `fat' operating systems (i.e. a single disk image of >OS & Applications will work regardless of which platform it's used >with)? Likewise, will installers and updaters always give you the >option of installing `fat' (or will they always install fat)? I can't >remember the 68k/PPC transition clearly. If the only choice wasn't fat binary, you were usually offered "Fat binary" or "m68k only" then later "fat binary" or "ppc only". I don't recall running into many installers that didn't let you install a fat binary. A "fat OS" would definitely be interesting. You might want to say that louder, and in Apple's direction ;-) > - If we were to buy a set of Macs for a new deployment that would need >to last four years minimum, what prospects do we have that if a >computer breaks down in two years, we could actually find a drop-in >replacement? It is already hard enough to get three-month-old Mac >software to work with three-day-old Mac hardware. Aah, welcome to my personal hell. > - Will users' profiles and files works seamlessly regardless of whether >the user sits down in front of a PowerPC and Mactel box? In >*practice*, I mean. I.e. what is the level of risk that there's a >gap between Steve's theory and the real world. (Shock, horror.) I'd be very surprised if applications didn't make foolish endian assumptions about user profile data - "I won't need to byteswap this, it'll be the same endianness as the host CPU". I imagine some of that will get ironed out once users actually start migrating across, but the typical developer answer for "non-critical" data will probably be "delete your settings." > - I had a look at the Anantech article and Apple's migration guide. >Gems include: Mactel will not use OpenFirmware; Rosetta has a whole >heap of limitations in its support. Great, what are the prospects >that current Netboot, Office 2004, Adobe CS, and all the smaller apps >and utilities, are going to work in future? For Photoshop CS and Office 2004 - pretty good. I don't see anything in that list that precludes them working. As for netboot - nfi. >What is the likelihood >that publishers will release cost-free patches for their old apps to >enable them to run on Mactel if Rosetta cannot do the job? If it's like the m68k -> PPC days, some will do it for free, and some will require multi-thousand-dollar upgrades. Not that I'm looking at a particular DTP company with a name beginning with Q or anything. >Won't we >just get stuck in two year's time when we can't buy PowerPC? I doubt it, personally. m68k support hung on for a long while. I imagine support will remain until the user PPC base stops buying enough software ;-) . Even then, there's a decent chance some enterprising developer will come up with a Rosetta-lookalike based on something like QEmu to do x86->PPC. If Apple doesn't beat them to it, that is. >Why not >use OpenFirmware so that admins can use their existing skills and >tools? ... and not have to deal with the pain that is the PC BIOS. I guess they might use EFI ;-) as they have only said they're *not* using OpenFirmware so far. That doesn't answer the "existing skills and tools" part, though. I agree that this is unfortunate, though I guess they must want to be more compatible at the expense of some initial pain. I still wonder why they didn't go down the EFI route and provide a modified OpenFirmware with a PC BIOS compatibility layer. -- Craig Ringer
Re: ppc-intel query
On Wed, 2005-06-08 at 11:32 +0800, Martin Hill wrote: > > From: Warren Jones <[EMAIL PROTECTED]> > > If the Mac will be able to run win software well, will this be a > > double-edged sword? > > - will that encourage companies to only write software for Win, knowing > > it will probably work on the new macs good enough? > > That is a possibility. However, I reckon it is more likely that the fact > that porting to the Mac will then be so much easier and cheaper I don't think it *will* be that much easier and cheaper. The difficulty has always been adapting your application to use MacOS/X's APIs, not the CPU - endianness and instruction set. Having SSE instead of AlitVec won't hurt, but for most developers the compiler takes care of that if they use it at all. Getting rid of the endian difference will be nice, too, but again probably not too major in comparison to the work required to do a proper port to MacOS/X. Porting from Windows to Linux probably involves a similar porting effort to what will be required to go to x86 MacOS/X. That said, there *are* tools to make that easier. There's a company called MainSoft that ports bits of Windows to UNIX to help Windows developers port their apps, for example. WINElib is another one - it lets your application "carry around" the Windows APIs it needs, giving app developers a sort of half-way-between porting target. All these work just as well on PPC though, and they haven't led to great floods of quick-'n-dirty ports. I wouldn't hold my breath. I doubt such ports would be received with much enthusiasm, anyway. Scribus started life on Linux, and even from there the port to MacOS/X isn't especially fun. Endian problems weren't a particularly big deal, but switching over to the "Mac way" of doing fonts, window management, etc is a lot of work. That's *despite* the fact that the app is built on a library called Qt that hides most of the basic differences - it actually ran on MacOS/X the first time I compiled it, it just needs fixes and polish. Given how much work that has been so far, and the "easy start" Scribus has thanks to Qt and it's UNIX heritage, I can assure you that there is a *lot* of work involved in a good MacOS/X port. -- Craig Ringer
Re: ppc-intel query
On Wed, 2005-06-08 at 11:14 +0800, Warren Jones wrote: > If the Mac will be able to run win software well, will this be a > double-edged sword? > - will that encourage companies to only write software for Win, knowing > it will probably work on the new macs good enough? It sounds like current options for running win32 on the new macs might be: (a) dual boot (install and run windows as well as MacOS/X) - best, but most intrusive; can't use both Mac and Windows apps at once. Requires Windows license. (b) Windows in a virtualized environment - fairly good, but slow for some things such as video, may not support Direct3D/OpenGL, etc. Requires Windows license. (c) API translators such as WINE. Least reliable, will only run some Windows apps without specific effort by WINE or the app developers to be compatible. No windows license required. I don't know about you, but none of those sound too fantasic. Given how much Mac users whine about X11 applications*, I wouldn't be worrying too much about win32 becoming dominant on MacOS. * Apple did a really half-hearted job of their X11 integration. As a developer of an X11 application this annoys me immensely - I've seen nothing to indicate it'd be that hard for them to provide access to the dock, Mac virtual filesystem, etc. They don't even ship the X server by default :-( -- Craig Ringer
Re: ppc-intel query
On Wed, 2005-06-08 at 08:50 +0800, James Devenish wrote: > In message <[EMAIL PROTECTED]> > on Wed, Jun 08, 2005 at 07:23:45AM +0800, Rod wrote: > > Steve says the soul of the Mac has always been the OS > > And the hardware its heart. > > Also, while there's been attention given to the fact that old PowerPC > apps will work on Mactel thanks to dynamic translation in Rosetta, what > happens the other way around? There are plenty of people who can't > replace all their hardware every three years. It's continual pain for me > that for most software, development for Mac OS 9 stopped several years > ago. So, the backward compatibility of Mactel software (i.e. ability to > run newer software on PowerPC Macs) is very much a concern for me. I share your pain on that one, actually. I'm *still* maintaining seven MacOS 9 macs I can't get new software for and now have to buy 2/h hardware to replace. Fun, fun, fun. It looks like I'll be stuck with them for a while now, too. I would expect that it'll take much longer for PPC users to get "left behind" this way, compared to how fast MacOS 9 was dropped, though. It'll be MUCH easier for app developers to build for PPC and x86, where MacOS 9 required constant pain and compromises to maintain support for. > But to me, x86 architecture seems like a relic and is reminiscent of > some of Apple's 'cheap', disappointing hardware choices -- need I > mention the backward steps Apple took with its flaky CRT-style iMacs?! Yeah, the innards of the instruction set aren't too pretty. The rest of the architecture is fine these days though - modern x86 chipsets and CPUs have resolved most of the long standing problems. Even the instruction set has been helped over the worst of its uglyness. The only "major ugly" remaining is the BIOS, and we can hopefully be rid of that with EFI at some point - or, if we're really, really lucky, with OpenFirmware. Hmm... Apple entering the x86 market might - just might, it's along shot - help improve the x86 architecture as well. If they do use and make good quality, well tested BIOS/firmwares, motherboards, chipsets and drivers, they'll still be able to maintain their "quality edge". SUN are shipping x86 gear (Opteron based) after all, and they're not exactly known for selling cheap and nasty garbage. > I guess Apple must have an assurance from Intel to design and deliver a > leading 64-bit chip in time for Mactel. My bet for the high end macs' CPUs is still AMD with the Opteron. For the lower end ones, it sounds like they'll use the new Pentium M so far. > But whatever CPU gets used in > Mactel, it seems to be given that it will be a standard Intel part -- so > that still leaves questions about DRM. Intel has said "there is no unannounced DRM" in their products. Yay. It doesn't sound like there's anything particularly really nasty in there from what's been dug up so far, but I share your general concerns. I'd be way more worried about DRM enabling chipsets than CPUs, though. The chipset really is the heart of the system, and if you're going to try to lock it down, that's where to do it. Intel already plans to release "with all new DRM!" video drivers. Yay. Somehow I don't expect Apple to use the onboard Intel Extreme[ly slow] video in their systems though. -- Craig Ringer
Re: ppc-intel query
On Tue, 2005-06-07 at 23:02 +0800, J Philippe Chaperon wrote: > What a real nightmare Steve Jobs has sprung on some of us!! I'm still trying > to fathom what all this will mean for the 'believers'. It'll be very hard > right now for the evangelists to evangelize the heathen to the Apple > religion Why? Most users' have always had the /real/ problem with Windows, not x86. If you really feel the need to view things in religious terms, I'm sure you can still manage to do so :-P > One thing I am certain of is that my dual G5, purchased to last me through > many upgrades, both software and hardware, or so I thought some months back, > is now like a lame duck sitting in the very busy Intel highway. Hardly. According to the announcement it'll be about two years, probably more, until Apple even release x86 workstations. They're starting with the lower end stuff - probably mac mini, iBook, etc. I wouldn't stress too much, myself. Sure, I'd be less than thrilled too since it probably _will_ eventually limit your software options, but it's not the end of the world. There's going to be at least OS/X 10.5 and a full suite of apps for that coming for PPC yet. > I now have absolutely no incentive to update to Tiger, I was planning to > place an order for one today!! My current 10.3.9 will do until it dies out, I imagine you'll probably want 10.4 or 10.5 eventually - I'd be surprised if developers didn't start requiring one of them eventually. That said, it sounds like it's safer to stick with 10.3 right now anyway... I'm not trying to say "you're wrong" or anything, just pointing out that this is unlikely to bring about the abrupt end of the Mac/PPC world. There's a lot of installed base out there. Did developers instantly drop 10.2 support when 10.3 came out? No - way too many people stayed with 10.2 and they wanted to be able to sell to them too. -- Craig Ringer
Re: RIP PowerPC
On Tue, 2005-06-07 at 16:51 +0800, Stewart Woods wrote: > On 07/06/2005, at 3:27 PM, Rod wrote: > > > As much as you would hate to think it, we do have a use for some > > Windows software out there :-) > > and don't get me started on games > > Boot into windows instead of waiting 2 years for a clumsy port - Where > do I sign? There might be even better than that if TransGaming (http://www.transgaming.com) move their tech over to work on MacOS/X too. -- Craig Ringer
Re: ppc-intel query
On Tue, 2005-06-07 at 14:42 +0800, James Devenish wrote: > In message <[EMAIL PROTECTED]> > on Tue, Jun 07, 2005 at 01:39:03PM +0800, Craig Ringer wrote: > > > [...] and putting a knife through developers' prospects. I imagine > > he's done a good job of doing exactly that, though. > > Really? > > Porting doesn't seem that hard. > > Is it expected that developers will just be able to cross-compile all > their apps and blindly hope for the best? If developers should be > testing and/or profiling their apps on the target platform, that doubles > the testing and doubles the hardware required. I haven't read the > porting document, but given past experiences (again, Solaris comes to > mind), it does seem a bit naive to assume that porting is no problem > (even if Apple's high-level interfaces "should" make it painless). That's true. It will increase testing requirements, especially at first. Again, well organized developers with good test suites etc will suffer less. > Again, all the same problems as transitions from 68k to PPC, OS9 to OSX, > etc. The transitions are entirely plausible, but still painful, and > Apple makes the pain keep on coming. For example: Apple's been rather > "stupid" recently and been selling new Macs that only run Tiger, i.e. > not Panther. Oooh, great. I'm still smarting from the macs that won't run OS9 anymore (but then I'm stuck in a legacy nightmare here at the POST). You're quite right in that another big change now won't be popular. We've had: System 6 -> MacOS 7 m68k -> PPC Old World -> New World (doesn't affect most app devs) MacOS 9 -> Clasic & MacOS/X and now: MacOS/PPC -> MacOS/x86 Yeah... it doesn't help developers view it as a nice stable platform to develop for. Even so, I don't see this as likely to be a particularly bad blow - and the potential gains are IMO significant. -- Craig Ringer
Re: ppc-intel query
On Tue, 2005-06-07 at 12:02 +0800, James Devenish wrote: > On Tuesday, Jun 7, 2005, at 11:11 Australia/Perth, Stewart Woods wrote: > > Why shouldn't I buy a new mac until next year? > > Because your bank doesn't like you making withdrawals? ;) > > I presume that Steve Jobs' intention was not to halt all Mac sales for > the indefinite future, disgruntle all recent switchers, spook anyone who > was considering switching or upgrading in the near future [...] I'd be surprised if he managed that. The initial response has been way more negative than I would've expected. Surely people will realize that these will still be "Macs" despite the different CPU, and that developers will maintain compatibility for a fair while yet (there being a big PPC installed base and all)? The main piss-off will be having to buy upgrades to all your software to get it to run at full performance on the Intel-based machines. Oh, and drivers. That *will* cause some fury. Hmmm... maybe you're right :S > [...] and putting a knife through developers' prospects. I imagine he's done a good job of doing exactly that, though. Really? Porting doesn't seem that hard. I've read the porting document, and most of it deals with the two areas I expected - endian issues, and vector instructions. In a well written application the vector code will already be in well-separated platform-specific parts of the code, preferably with a portable equivalent already done. That just shouldn't be too bad even if they do need to write some new SSE-based vector ASM for x86. If the AltiVec code is scattered through their codebase then someone needs to show them what software engineering is. Endian issues will be more of a pain for Mac developers who don't also support any little-endian platforms already, but shouldn't be *that* big a deal. They didn't cause too much trouble when porting Scribus over to MacOS/X for PPC (not finished, but hardly anybody has time to work on it right now). > Will be interesting to see the Intel > equivalents of the AltiVec and Xserve products, esp. given the > inevitable Mac-vs-Windows-vs-Linux performance comparisons (eg. Adobe > Photoshop for workstations, Apache for servers, etc.). My bet: Opteron . That'd provide SSE 3 and 3DNow pro. Admittedly not really the same as AltiVec, but not too bad either. > I wonder how Rosetta will perform, too. I'd expect a pretty serious performance hit. Reading between the lines, he keynote suggests that it'll be fine for apps which are often bottlenecked at the disk, or that are usually waiting for user input, but not so hot for apps that are CPU heavy. If it's less than an order of magnitude I'll be fairly impressed. > Presumably Apple will try to do a better job > than Sun has done with its years of messing around with x86 vs SPARC. One can only hope. -- Craig Ringer
Re: RIP PowerPC
On Tue, 2005-06-07 at 06:13 +0800, Rob Davies wrote: > I feel sorry for all the Apple retailers who have to re-evaluate > future whilst surviving the transition, knowing they are going to be > competing with our x86 brethren after the dust settles. They already are. I don't see that too much changes here, frankly, unless Apple decide to let MacOS/X run on standard PCs. Do you think that's likely, 'cos I sure don't. Thus, the competition is much the same as it was, though Apple might have a bit of an advantage if they let users dual-boot, or sell bundled versions of VMWare with Windows or something. Of course, I won't be surprised if the Mac-on-Linux crowd have cracked out a version that lets you run MacOS/X on a standard Linux PC within a few weeks of getting their hands on a preview build. That's geek territory really - I don't see your average user buying MacOS/X then installing Linux to run in in a virtualized environment. You have no idea how happy the idea of being able to run (probably virtualized) MacOS/X on my nice AMD desktop makes me, as a software developer and tester, though. I wonder how much more interest Apple might get from software developers if they provide a legit way to run MacOS/X on developers' existing Windows workstations? Marketing to the developers and improving the availability of the target platform, all in one. > Obviously Apple did not learn from the last fore-ray into sharing > chips and technology. Maybe they won't shoot themselves in the foot quite as hard this time ;-) > As Shay mentioned earlier PowerPC offered security x86 definitely > does not That's BS IMO. The incompatible OS APIs offered "security" - and still will offer the same level of it now. Just like before, there won't be viruses unless someone decides to target MacOS/X with them. The CPU doesn't matter. > oh well what's another day on a weekend checking for > rootkits, spyware, malware and this weeks new virus or rebuilding > OS. Dunno. Do you plan on running Windows on yours? If not, I wouldn't worry. Not unless you worry about viruses on your current mac. Just like always, you'll want to be aware of the possibility, but won't need to expect every file to be wrapped in viruses. > At Least $M Gates will be smiling another 12 months to get his > Longhorn $M stable, and out the door. Whilst knowing he can control > another part of the Apple, it's future development. Er... how so? Apple can go to any x86 chip maker - currently AMD and Intel for the high end, VIA and Intel for the low end. Microsoft can pressure Intel, but not as hard as they once could. I wouldn't think Apple has opened themselves to anything they weren't already subject to because of their platform's dependence on MS Office. Microsoft can't afford to pressure Apple right now, either. They have legal problems and need to look like well behaved, honest above-board folks. > One must ask if PowerPC has no future roadmap or growth why have all > the leading games console manufacturers switched to them for the next > generation. Well, they /are/ using very specialized chips. Not desktop PC material at all, really. IBM /has/ been having trouble ramping desktop/workstation chips to very high speeds, and remember that there's the Opteron to consider in the workstation market now. -- Craig Ringer
Re: He was dressed in all black.
On Tue, 2005-06-07 at 03:12 +0800, Doug Wilson wrote: > I just know I'm going to have nightmares of Intel chips chasing me across > Apple motherboards. Aah, but there you have it. If they're not *Apple* motherboards, probably with OpenFirmware, I'll be shocked. It sounds a lot like they've chosen Intel for Pentium M (note Jobs' comments on power efficiency) - but that doesn't mean they'll want to use unmodified Intel/Via/SiS chipsets and boards. I'll be very surprised if they'll make it possible to run MacOS/X on standard PCs or Windows on macs. As for what's ahead, my money - if I had any to spare - would be on Opteron for the G5 replacements, or failing that EM64-T Xeon. Current users of Virtual PC might be interested to note that it'll almost certainly get a massive speed boost on the new systems, as it'd only need to virtualize the CPU, not emulate a different CPU. I'd put money on VMWare and QEmu ports for MacOS/X in a hurry, too. I wonder how long the WINE (http://www.winehq.com/) port is going to take - imagine being able to run some of those Windows-only apps without emulating all of Windows? Interesting times head. If I can run MacOS/X as a guest OS under Xen (and they put two trackpad buttons on their new iBooks), I might even buy a Mac laptop for my development work. I have a half-finished MacOS/X port of Scribus to help finish, and this news makes it all sound much more fun :-) -- Craig Ringer
Re: What shall we make / Intel for Mac OSX
On Mon, 2005-06-06 at 20:17 +0800, Rod wrote: > > All I know is that the story was first floated on CNet, which in the > past has been a traditional Mac hater. They love spreading stories > like this one, as it really gets the mac zealots fired up :-) > > And as for Intel making PPC chips, I think they would have to swallow > a lot of pride to do that (or be thrown oodles of cash). There is another interesting possibility though. Pentium M. It's fast, cool, efficient, and getting cheaper in a hurry. I can see that being a seriously attractive CPU for Apple's lower-end offerings - and they might well be pondering moving the higher end gear to Opteron later. I'm still dubious about the whole thing, but if there's any truth in this jumping ship stuff, I expect Pentium M may be part of it. > My money is on a new digital hub hardware device, much like the > EyeHome device from El Gato. [snip] Me too. Intel's StrongARM is very popular for this sort of thing. -- Craig Ringer
Re: what shall we make of this then?
On Mon, 2005-06-06 at 13:14 +0800, Dark1 wrote: > the inquirers says NO > http://www.theinquirer.net/?article=23708 > > As far as I could tell from that report Intel hasn't actually said > that the chipsets don't have DRM technologies. All they've said is > that the chipsets don't have "unannounced" DRM technologies. > > Does anyone remember DRM for the apple DVDs. That was heaps of fun. > I love using Apple products but the day they start incorporating DRM > into their machines is the day I switch to being a PC user. What, so you can have DRM incorprorated into the product? iTunes uses DRM. DVD players - all of them (well, legit ones) - use DRM. Windows Media Player uses DRM. Some x86 PCs (particularly IBM notebooks) already have technology from the TCPA embedded that can, among other more useful things, help enable DRM. Microsoft has been talking about extensive DRM in the next version of Windows, though it's apparently been scaled back a bit now. Apple's doing quite a bit better on the DRM scale overall, but still not great. The only way to be rid of it entirely is to use a "minority" or free OS - Linux, BSD, another *nix, BeOS, etc. Then, of course, you lose some functionality. > Until everyone else is doing it I think it would be a very big > mistake to start incorporating DRM into chipsets. IBM's already doing it, so Apple wouldn't be the first. > If I'm not the > only person with this point of few on DRM then Apple could stand to > lose a lot of loyal customers. Most people won't know or care. They'll probably only be aware of the extra things the movie and music industries will let them do because of it, without realising what they now *can't* do. -- Craig Ringer
Re: what shall we make of this then?
On Mon, 2005-06-06 at 08:21 +0800, Mark Secker wrote: > ComputerWorld says the new PentiumD chips are shipping with DRM (copy > protection hardware) embedded > http://www.computerworld.com.au/index.php?id=580672002 > > > the inquirers says NO > http://www.theinquirer.net/?article=23708 > > Why should mac users care...we... if it isn't this old > chestnut again: (being dragged out by Wired this time) > > http://www.wired.com/news/mac/0,2125,67750,00.html?tw=wn_tophead_2 > > apparently because of this: > http://www.wired.com/news/mac/0,2125,67749,00.html?tw=wn_tophead_1 > > > is it true ... well maybe Apple have traveled that platform path in > the lab to test it as feasible IMO they would be crazy not to have been maintaining an in-house PC/x86 port of MacOS/X . Even if they never intended to use it, it'd be important future-proofing, and keeping it in good condition would ensure a much lower-cost transition if they *did* make one. Porting an OS once is hard (even if the OS started out on x86 like MacOS/X); maintaining the port is much easier and cheaper. > Personally I think Apple, for sure, would be wise pursuing low end > home users media "digital hub" on intel bassed platforms. I'll wait for the announcement - I'm not convinced. I'm more inclined to believe theories about new embedded devices and the XScale than a move from PPC to x86. They /could/ do it - MacOS/X should be fairly portable (at least to an x86 board with OpenFirmware) if Apple have even half a brain. Most well written MacOS/X native apps should be a simple recompile for their developers, plus a few days/weeks tweaking for endian issues, writing replacements for PPC asm segments, etc. Less if the app was well written in the first place. Developers could ship fat binaries to hide the arch differences from users, too. They could handle older apps with emulation like QEmu does for Linux - because the OS would have identical and compatible APIs they would only need to "translate" the binary at runtime - not emulate a whole OS. Classic could be emulated, or simply ditched entirely. The downside: users would need to buy new apps to get full performance, as most developers would be unlikely to release upgrades for free. Older apps would always just run slowly. The biggest problem, though, would be drivers. MacOS/X PPC drivers wouldn't work, and neither would Windows PC drivers. That'd be a lot of obsolete hardware and pissed-off users. Additionally, if they do make such a move - and again, I'm not convinced - I'd be stunned if they used off-the-shelf x86 chipsets and motherboards. I'd strongly expect to see Apple hardware with an x86 CPU instead of a PPC CPU. They'd probably bring in OpenFirmware, or maybe adopt EFI, but I'd be very surprised if they used the legacy PC BIOS. That means no dual-booting Windows, and still needing special Apple hardware to run MacOS/X. PearPC / One totally awesome thing that could be done if Apple /did/ switch is that emulation for Windows could be made *massively* faster. It'd also be much more practical to port WINE to MacOS/X so Windows apps could be "translated" like is sometimes done (unreliably) on Linux currently. The biggest thing that makes me skeptical is that they're rumoured to be talking to /Intel/. Now, if they were taking to AMD I'd be much more inclined to believe this - the Opteron is hot stuff. Intel's offerings are rather lacklustre at the moment - the overpriced, failing Itanium that never went anywhere, the power-hungry, not-that-fast Pentium 4 and the 64 bit Xeon that's really no match for the Opteron. Talking to Intel is much more likely to be about embedded products IMO ... though I guess they could be interested in the very cheap P4 Celeron. They could also just be pressuring IBM - but given their recent Cell deals, I'd suspect this to be a really bad time to try that. Anyway ... it's interesting to see what'll happen. There's some big conference on soon, right? -- Craig Ringer
Re: Question regarding what type of database to use
On Fri, 2005-06-03 at 09:15 +0800, Rod wrote: > I figure I can accomplish this quiet easily with a Filemaker/ > Filemaker Server combo, but price is a consideration ($0 would make > the boss very happy!). There are two solutions I am considering: It'd probably get you the best results, fastest, but at a considerable cost. It also means you need to send money their way for upgrades if, say, your version won't run on some newer version of MacOS/X. > a) Using Webobjects, so all I need on the client machines are a web > browser I thought WebObjects was very expensive. Am I mistaken? There's also the learning time (if any) and the no doubt considerable development time to consider, especially given that writing good web-based user interfaces for databases is quite difficult. > b) Using REALBasic to create a front end for an SQL database. Being > a multi platform language, REALBasic might be a good solution, as I > can also tailor the interface to what we need That does sound like a sensible option if you need a "real" client program. Building a good database user interface is non-trivial, though it's easier with a rich client than a web interface. > Anybody here had experience with either solution? Are they difficult > to implement? Can anybody point me to a good resource for SQL > (especially on the Mac), as I'm a bit 'green' when it comes to SQL. There's really nothing Mac specific about SQL. Whichever program you're using, the language is pretty much the same, so you can install PostgreSQL or some other database and not really have to worry at all about whether it's on a mac, a UNIX box, or something else. There are a LOT of SQL tutorials etc on the 'net. I found a fantastic one a while ago, but of course can't track it down now and I never noted down where it was. Sorry. I found the manual for the PostgreSQL database engine to be EXTREMELY useful as a resource and guide to SQL, though it doesn't start at the very basic level. There are also lots of books out there. I'd probably install PostgreSQL or some other database and start plodding through a few tutorials, then read the database manual in detail and go from there. If you do decide to go down the programming route, one alternative to WebObjects is PHP . PHP is a free language for writing web applications. It'll run on any platform and talk to most SQL databases - SQLite, PostgreSQL, MySQL, MS SQL, Oracle, etc. Like WebObjects, it's quite a bit of work to learn how to do good web applications in, though, and you'll never be able to do as good a user interface as in a "real" client program. Personally, I do all the in-house web apps and databases front-ends at work in Python, another programming language, but that's because it's what I know and am comfortable with. I'd suggest PHP for someone just getting into it, since it's much easier to get started with and there's much more info out there on the web about it. There are even books specifically about database programming with PHP. In your position, I'd be seriously considering FileMaker. How much time will it take to produce an adequate solution by writing it yourself? Is it actually worth doing so when you can buy a program that'll let you do it quickly and easily out of the box? If you write it yourself, what position will your employer be in if you leave - how will they maintain it, and how long will it take you to write the documentation to let them do so? I'd write it myself, but I already know the tools and the languages needed. In your position I'd seriously have to consider that FileMaker might be the best option and the lowest cost overall. Failing FileMaker, I'd probably want to choose a tool like PHP with PostgreSQL that doesn't cost anything, and didn't tie me down to any platform support and upgrade agendas other than my own - even if it does take me a lot longer to produce a good result with. -- Craig Ringer
Re: Firewall worm on Windows? Malware statistics
On Wed, 2005-06-01 at 11:56 +0800, Martin Hill wrote: > I've been trying to find the reference to a Firewall product on Windows that > despite only having something like 60,000 users, had a worm or virus written > specifically for it - a prime example of how the Theory of Obscurity (that > there aren't enough Macs out there to be worth a Hacker's time) is flawed. That sounds like BlackIce, which was attacked by the Witty worm. http://vil.nai.com/vil/content/v_101118.htm http://xforce.iss.net/xforce/alerts/id/167 http://blackice.iss.net/product_pc_protection.php I understand that there's evidence the attack was assisted by an insider. It's a very unusual attack too (highly destructive), and there may have been some other oddities involved. I personally don't find it compelling in terms of the obscurity argument. Yes, the user base was small - but the potential damage was high, they probably had inside information to make the attack easier, and the potential embarrasment to ISS was *massive*. Someone with a grudge... To me, it seems to be all about payoff. Effort vs impact. Both can be measured in many ways, so it's not very clear cut. Impact can have many factors - user base, ease of attack, damage caused, potential attention and press coverage, ability to use the attack to make money (spambots, information theft, etc), etc. Effort can be affected by many things - baseline security of the OS, ability to assist the attack with a vulnerability exploit, ease and reliability of exploit, availability of virus kit, availability of how-to documentation on the exploit, etc. Overall, I find this the most sensible way to explain current attack patterns and the best match I've yet found to what is actually going on. > Any more statistics on malware for Linux or other comments would also be > appreciated. There isn't much in the way of worms and malware for Linux/UNIX, but (just like MacOS/X) I think that's mostly because it's an uninteresting target. Low user base and relatively high baseline security level (can't easily modify system files, silently install drivers etc) probably mean that it's not really worth it. It's not like there's a shortage of security holes, but most of them would be pretty hard to exploit. Writing an effective "desktop-style" worm to attack Linux would probably be even harder than MacOS/X because (a) chances are there aren't as many desktop users, and (b) the variety of common mail programs etc is mind boggling - there's nothing even remotely close to a "basic standard" like Mail.app . That doesn't mean attacks won't happen, only that they're less likely, will take longer/be less frequent, and will probably affect fewer users. Mac users should remember this too - macs aren't inherently "virus free" or immune to viruses. As far as I can see, they're just more difficult and less interesting to attack. I get sick and tired of users of both Linux and MacOS trumpeting "malware immunity" when it's patently not true. A script called "runme.sh" with "sudo rm -rf /" can be considered malware and would work on a sufficiently stupid/inexperienced/gullible user, if it comes to that - and lots of Windows trojans and worms do rely on that kind of stupidity. As for "server" style worms exploiting running services rather than client applications, the only major worm to affect Linux systems I know of in recent history was Slapper - and that attacked Apache. My understanding is that attacks against UNIX and Linux systems are far more often direct crack attempts (exploiting mis-configurations or un- patched security weaknesses) to steal information and/or take control of a box on a high-bandwidth connection. How much that has to do with Linux being primarily considered a "server OS" isn't something I care to guess about. I'm afraid I don't have any links or stats for any of this off off-hand, though I'll see if I can hunt up some later. Google should help though, and I expect will largely confirm what I'm saying (ranting by zealots about "immunity" aside). Personally, I've never seen a virus or worm for anything but Windows; I've seen a couple of rootkit reports from Linux users, but only in mailing list posts - never on my systems. -- Craig Ringer
Re: Graphics/video workstation.
On Thu, 2005-05-26 at 17:34 +0800, Christian Kotz wrote: > I've been told that Mac's traditionally do better in the roles of > creativity anyway in comparison to PC's. A lot of PC users understand > this but is it a myth? It used to be true, back in the MacOS 7 vs win311 days when there was just no comparison. I don't think it's really all that significant anymore. > I would like to know as I planned to move into > this field as a career but if I won't be seeing many Macs it could be > concerning. I've used Premiere on a PC and gee, I was impressed... :P, > Constant crashing etc real productive. I think it's much of a muchness, really. Depends on what you prefer to work with. You can get all the important apps for both, and the colour management on both platforms is broken in various ways so you'll end up using the apps' built in colour management no matter what platform you use. For me, the biggest single factor is that Windows apps mostly use these ghastly MDIs (multiple-document-interface, ie window-within-a-window) that I simply cannot stand. Not that I'm fond of Apple's weird window management either (switching apps not windows), as a long term UNIX user, but it's at least a little less nasty than an MDI. I've found stability on both to be excellent (well, MacOS 9 is always poor stability-wise, I mean MacOS/X and above) if the OS and apps are properly configured, and stability on both to be awful if they're not. It's easier to get it wrong with a PC though, in particular because many virus scanners and personal firewalls are pure evil and worse than the problems they try to solve, and because bad software is more able to harm your system's stability on a PC. You also need to be very careful about spyware etc. On the other hand, in the mid-range you get a lot more grunt for the buck with a PC IMO. Depending on the work you do, that may matter a lot. Right now, I really don't know what I'd pick. I'm an experienced sysadmin, so I can make Windows behave its self, and I'm more familar with x86 - but I /like/ the UNIX-like innards of MacOS/X even after all the butchery Apple have commmited on them. In the end, I think I'd suggest you make the decision based on what platform you're most comfortable working on. -- Craig Ringer
