within ten minutes the font has changed itself to arioso system wide?
When the font configuration data gets reset, or it can't find the default font (such as, if you are using bitmap helvetica as your font (which is the default), but using Xft rendering which does not support bitmap fonts), it goes to whatever is alphabetically first among the fonts it is aware of. Sometimes it ends up being Arial, which is ok (although its usually the wrong size), on my computer it typically chooses an absolutely horrid looking font called "Agate".
Before they went under, they made a 256-color VGA remake of KQ1 for sure, probably other early games too.
They did Space Quest I also. The first Leisure Suit Larry game may have been the other one, I'm not quite sure.
Re:This arguement needs to be put to rest
on
Apple Wants Your Input
·
· Score: 3, Insightful
This would be like going to a dealership and saying "Why should I spend $40,000 on this BMW when I can go across the street and get a Geo Metro for $9,000? It'll take me to work just as well as the BMW won't it?"
This is not a sound analogy. First of all, the cheap PC is as fast or faster. But more importantly, other than the CPU and the case, modern Macintoshes basically are PCs. They use standard PC memory (and not DDR, either), standard PC video, IDE hard drives, PCI/AGP expansion slots, USB and Firewire...inside the shiny case there is nothing unique about them. There might have been something to this in 1984, when you could get a Mac with real sound and a 32-bit CPU where no corresponding PC existed, or even 1989 or so with a Mac II and a high-resolution display. But the proprietary hardware that used to distinguish systems like Macs and (moreso) Amigas from PC clones can't compete with commodity PC hardware in price/performance any longer. Even the PowerPC CPU, which by the most optimistic estimates is only on par with Intel and AMD, really only functions as a glorified copy protection device, to make running the operating system on (even more) commodity hardware difficult. If they could do it without opening the door to simple OS X emulators for commodity PCs, I believe Apple would move to AMD or Intel CPUs for the cost savings.
Re:Why yEnv is good for the software companies
on
Usenet Encoding: yEnc
·
· Score: 2
Most progressive newsreaders group multi-part binary posts together pretty well. That really isn't a problem at all. And the reason for the multi-partness doesn't have to do with a particular encoding scheme, but with the line number limit many servers impose on posts.
$450 is a little steep compared with Intel hardware, but this is worlds better than the $3000 developer boards that have been options before. I don't think I can afford it now, but if the final publicly available version is anywhere near as cheap I will get one eventually.
WINE stands for Wine Is Not an Emulator. It implements native code to the function calls present in the Windows DLL's. An emulator is something that duplicates the environment that an application runs in. WINE doesn't bother.
I'm well aware of what their acronym means and their arguments therein. But what does emulate actually mean: 1. To imitate the function of (another system), as by modifications to hardware or software that allow the imitating system to accept the same data, execute the same programs, and achieve the same results as the imitated system. Now, what does WINE do? Is it windows? Clearly not. So windows is "another system". What is its goal? Clearly it puports to "achieve the same results as another system". I don't see any qualifiers in the definition requiring emulation to involve the original [windows] binaries or a virtual machine environment. It doesn't [i]need[/i] to emulate hardware because it is already running on x86-compatible PCs.
OS/2's Windows support was hokey. By the time Windows programs were really taking off, they were for Win32s (Win95-style, not Win3x). Those programs worked poorly at best under OS/2 if they worked at all.
Thats not the point. If you take a Windows program from the time OS/2 2.0 was released, it works totally transparently. Obviously a big part of that was the access to the Windows 3.0 source code, which nobody working on WINE or anything else has for current versions of Windows. I cannot simply take a windows binary of any kind and run it under x86 unix without any configuration. You have to install wine and grapple with it, and if you are lucky it will work.
As for your other points, no current versions of Word, IE, or Outlook work under OS/2
No kidding, they postdate any serious work on the Win-OS2 code by many years. But if you take any program with comparable ubiquity from the early nineties, it will work with the level of transparency I've described (and I wouldn't necessarily count Microsoft's office software as ubiquitous at this time...they really took over the world in that department about the time that OS/2 was being destroyed.).
No stable Gravis Ultrasound Support. Yeah someone had written a driver but it was still flakey as hell. Wave Table synthesis smashed the hell out of FM synthesis and I wasn't going to play my games in FM anymore.
Thats true of linux too. The only driver that supported all of its features does not compile against modern kernels or with modern compilers.
With modern versions of Linux (and windows also, actually) there's no Synthesizer support at all.
I hope this isn't going to happen with wine/linux. Its quite obvious that windows programs will never quite work perfect in wine, and I hope developers dont use wine as an excuse to not bother developing linux applications.
Lindows aside, I don't think we have to worry about this. You could literally just pop in the installation disks for most (non-MS) 16-bit windows software and install it and run it under OS/2. Wine is considerably more, err, byzantine. And plus it doesn't work very reliably for the software (well, Word, IE and Outlook) that most people want to use, and it seems like the developers are more interested in using the code for proprietary emulation for running specific programs (games, plugins) or porting (corel stuff, etc.) than producing a general, Free, universal windows emulator.
Moreover to that, IBM's devotion to OS/2 could have been seen as a competitor to its Unix line. So if IBM was about to fully support OS/2, that could have been considered as killing the Unix line.
Absurd. IBM actually talked for some time about porting OS/2 to the PowerPC (as discussed in that article), but it never happened. Mainly because they explicitly didn't want to sell cheap PowerPC machines that would compete with their own RS/6000s, but rather they wanted someone else (Apple, who never actually did) to adopt their PReP standard and sell the hardware for them. This is also why there was briefly a PowerPC port of Windows NT. As it was though, you could never actually run OS/2 and AIX on the same hardware, so they weren't exactly competitive (also IIRC, AIX was supposed to be ported to their microkernel also).
Why can't Mandrake wait another month for a release so it can include both KDE 3 and GNOME 2?
I don't think you'll see any distribution (except maybe some of those crazy expert-oriented, source-based distributions I've been hearing about) using either of those by default so soon after release. Meanwhile my friend claims kde 3 beta is alread available through "cooker", if you really want it fast you will probably be able to get it.
Re:True Type font support missing from official bu
on
Mozilla 0.9.9 Released
·
· Score: 2
Yeah thats what the (non-working) builds I was using report too. I did find a nightly which seems to have it compiled in, looks pretty nice actually. Seems to be some problems with bold and italic text when using freetype.
next dsistros for the desktop will be Xandros and Lycoris
I don't know anyone who has ever used either of those. Right now they have about as much credibility with the average linux user as Caldera or Turbolinux. Mandrake I think is excellent for new users (I favor Debian and Slackware myself, but I must concede that I figured things out on Red Hat first). Its just as easy to install as something like Corel/Xandros, but the software set is much more standard and up-to-date, and its easier to upgrade. Maybe in the long run commercial distributions in general are doomed, but for now Mandrake fills a useful niche and even though I don't use their distro myself, I might donate to them were I not completely broke.
Re:True Type font support missing from official bu
on
Mozilla 0.9.9 Released
·
· Score: 2
A alot of good it does all of us when the
0.9.9 release does not include the true type font support! Any link to builds that do?
Good to know I'm not the only one having problems with this. None of the non-RPM release files seemed to have it enabled. Enough people are reporting it to work that I assume the RPMs must support it. Hopefully when the slackware packages come out on linuxpackages.net in the next couple of weeks they'll have it compiled in (they're built off of the official source rpms after all).
ACPI has a lot of benefits, and the problem isn't really ACPI per se, but the poor support for ACPI in free operating systems is the real problem here. ACPI has been around for a while (my 4-year old socket 7 motherboard supported it optionally), and the PCI IRQ sharing that this person is griping about is actually part of the PCI specification and should be supported by the operating system exclusively of whether ACPI works or not. It does enough things better than before that its likely to be standard pretty soon. And if the linux (and bsd) acpi developers don't get on the ball, there could be no new notebooks at all with working power management in free operating systems within a year. This is no different than Microsoft demanding that system makers remove floppy drives and ISA slots. Which they've been doing or will do soon. Rather than whining about it would be much better for someone to write decent ACPI drivers.
Re:In case you don't know what they're talking abo
on
C
·
· Score: 2
Maybe you don't have to make your code run on an old VAX as well as a system with a ANSI-C compiler
Hasn't gcc been ported to the VAX? Seemed like it would have to have been, or the NetBSD Port would be impossible.
I saw some Sun boxen being demoed running NT. I was fairly sure that this was an effort on the part of Sun Microsystems.
You're sure it wasn't a Sparc box with one of the add-on PC compatibility cards (which are really seperate computers on a card, with an Intel CPU, dedicated RAM, etc.)?
I've often been intrigued by some of the older Unix workstations, particularly Alphas (for their wide compatibility with PC hardware, of which I have an abudance, and the mystique they carried when they were new). Articles like this insist that people are just throwing these things away, and you can get truckloads of them for nominal cost.
But everything I've seen, on eBay and elsewhere on the net, has been, while maybe inexpensive and even cheap, totally out of proportion to the cost for older PC and even Mac hardware. As the benchmarks in that article show, a 21066 Multia with no cache is barely faster than a 486 at half the clock speed. And yet a loaded multia can still sell for upwards of $200. And the AT-format 21066 board based on the same architecture as the Multia can cost $50 alone (with CPU). I can get a box of 486 or Pentium boards for that much. And of course there is much more abundant binary-packaged software that will run on those.
f they got a good settlement out of this, it might give them a new lease on life. The problem would be buying their intellectual property back from Palm. Does anyone know if Palm is planning to use it for anything?
Allegedly Palm is using the Be technology as the basis for their next OS. The current Palm OS can't really do the type of things that, thanks to Microsoft et. al, everyone thinks a handheld device needs to do.
Also don't think that even if they get some money, they will return as a viable software company. More likely any money they get will go to their creditors.
Is there anything on the books (FCC, et al.) now that would stop me from *legally* implementing any of these solutions?
I don't think you run into legal issues until you start selling them commercially. Then they probably need to be FCC certified, which costs losts of money, etc. Also, there may be some restrictions on maximum signal strength on the ISM bands that 802.11b uses...but that, I'm not sure about, they might just regulate power output.
First of all, there is already a "modified" RedHat out there, Peanut linux, which can be installed on more minimal systems. Second, Slackware and Debian, which use simple text based installers, can already be installed on machines with as little as 8 megabytes of RAM, and they aren't cut-down mini-distros, but real distributions which include lots of packages and can scale to almost any task. RedHat, with its resource-guzzling graphical installer and auto-configuration systems (which are absolutely useless and border on counter-productive on old machines with lots of non-PnP ISA hardware), is, with the possible exception of Mandrake, the worst possible basis I can think of for a minimalist linux distribution.
When I saw this, what came to mind was my memory of having installed Slackware 3.2 (kernel 2.0.30 IIRC) on a 386SX with 4 megabytes of RAM about 4 years ago. And I ran X on it (sort of)! To think that their target is "32mb or less", when the system requirements of quite a bit of the base software have not changed a lot, is ridiculous. There is a need for something that can install on machines with really low memory...I don't think the trick i used to get slackware 3 on my 386 (not mounting the initial root FS on a ramdisk, creating a swap partition and adding it immediately, using two floppy drives) would work with current versions of slackware. But this isn't it, not even close.
wait... is this freedos thing a virtual machine that runs on top of *nix?
I seem to remember there being a program called "dosemu" that was bundled with a lot of linux distributions in the past, which could run a virtualized dos session from *nix. In fact it used FreeDOS by default IIRC. I don't know what became of it though.
But FreeDOS itself is a standalone operating system, a drop-in replacement for older versions of DOS.
Microsoft did not port their NT to PPC.(you are right NT is portable and asm is not, but PPC's asm is open enough for them, at least as far as I know)
It was ported, IIRC it was even released. You can find a "ppc" directory on NT 3.51 and 4.0 CDs. IBM has been balking on open PPC hardware for almost 10 years.
NT can run on Alpha. I'm not sure whether NT5(aka W2K) can run on Alpha, but previous versions can. It's Microsoft who left Alpha, not vice versa.:)
Hard to say. NT on Alpha development was always funded by DEC, then Compaq, to a certain extent. Compaq pulled funding and support for it, and Microsoft stopped development (they actually kept using it to develop the 64-bit version until Itanium hardware became available). There are beta versions of Win2k for alpha, but the final version didn't support it.
Slashdot Mirror
When the font configuration data gets reset, or it can't find the default font (such as, if you are using bitmap helvetica as your font (which is the default), but using Xft rendering which does not support bitmap fonts), it goes to whatever is alphabetically first among the fonts it is aware of. Sometimes it ends up being Arial, which is ok (although its usually the wrong size), on my computer it typically chooses an absolutely horrid looking font called "Agate".
They did Space Quest I also. The first Leisure Suit Larry game may have been the other one, I'm not quite sure.
This is not a sound analogy. First of all, the cheap PC is as fast or faster. But more importantly, other than the CPU and the case, modern Macintoshes basically are PCs. They use standard PC memory (and not DDR, either), standard PC video, IDE hard drives, PCI/AGP expansion slots, USB and Firewire...inside the shiny case there is nothing unique about them. There might have been something to this in 1984, when you could get a Mac with real sound and a 32-bit CPU where no corresponding PC existed, or even 1989 or so with a Mac II and a high-resolution display. But the proprietary hardware that used to distinguish systems like Macs and (moreso) Amigas from PC clones can't compete with commodity PC hardware in price/performance any longer. Even the PowerPC CPU, which by the most optimistic estimates is only on par with Intel and AMD, really only functions as a glorified copy protection device, to make running the operating system on (even more) commodity hardware difficult. If they could do it without opening the door to simple OS X emulators for commodity PCs, I believe Apple would move to AMD or Intel CPUs for the cost savings.
Most progressive newsreaders group multi-part binary posts together pretty well. That really isn't a problem at all. And the reason for the multi-partness doesn't have to do with a particular encoding scheme, but with the line number limit many servers impose on posts.
$450 is a little steep compared with Intel hardware, but this is worlds better than the $3000 developer boards that have been options before. I don't think I can afford it now, but if the final publicly available version is anywhere near as cheap I will get one eventually.
I'm well aware of what their acronym means and their arguments therein. But what does emulate actually mean: 1. To imitate the function of (another system), as by modifications to hardware or software that allow the imitating system to accept the same data, execute the same programs, and achieve the same results as the imitated system. Now, what does WINE do? Is it windows? Clearly not. So windows is "another system". What is its goal? Clearly it puports to "achieve the same results as another system". I don't see any qualifiers in the definition requiring emulation to involve the original [windows] binaries or a virtual machine environment. It doesn't [i]need[/i] to emulate hardware because it is already running on x86-compatible PCs.
Thats not the point. If you take a Windows program from the time OS/2 2.0 was released, it works totally transparently. Obviously a big part of that was the access to the Windows 3.0 source code, which nobody working on WINE or anything else has for current versions of Windows. I cannot simply take a windows binary of any kind and run it under x86 unix without any configuration. You have to install wine and grapple with it, and if you are lucky it will work.
No kidding, they postdate any serious work on the Win-OS2 code by many years. But if you take any program with comparable ubiquity from the early nineties, it will work with the level of transparency I've described (and I wouldn't necessarily count Microsoft's office software as ubiquitous at this time...they really took over the world in that department about the time that OS/2 was being destroyed.).
Thats true of linux too. The only driver that supported all of its features does not compile against modern kernels or with modern compilers.
With modern versions of Linux (and windows also, actually) there's no Synthesizer support at all.
Lindows aside, I don't think we have to worry about this. You could literally just pop in the installation disks for most (non-MS) 16-bit windows software and install it and run it under OS/2. Wine is considerably more, err, byzantine. And plus it doesn't work very reliably for the software (well, Word, IE and Outlook) that most people want to use, and it seems like the developers are more interested in using the code for proprietary emulation for running specific programs (games, plugins) or porting (corel stuff, etc.) than producing a general, Free, universal windows emulator.
Absurd. IBM actually talked for some time about porting OS/2 to the PowerPC (as discussed in that article), but it never happened. Mainly because they explicitly didn't want to sell cheap PowerPC machines that would compete with their own RS/6000s, but rather they wanted someone else (Apple, who never actually did) to adopt their PReP standard and sell the hardware for them. This is also why there was briefly a PowerPC port of Windows NT. As it was though, you could never actually run OS/2 and AIX on the same hardware, so they weren't exactly competitive (also IIRC, AIX was supposed to be ported to their microkernel also).
I don't think you'll see any distribution (except maybe some of those crazy expert-oriented, source-based distributions I've been hearing about) using either of those by default so soon after release. Meanwhile my friend claims kde 3 beta is alread available through "cooker", if you really want it fast you will probably be able to get it.
Yeah thats what the (non-working) builds I was using report too. I did find a nightly which seems to have it compiled in, looks pretty nice actually. Seems to be some problems with bold and italic text when using freetype.
I don't know anyone who has ever used either of those. Right now they have about as much credibility with the average linux user as Caldera or Turbolinux. Mandrake I think is excellent for new users (I favor Debian and Slackware myself, but I must concede that I figured things out on Red Hat first). Its just as easy to install as something like Corel/Xandros, but the software set is much more standard and up-to-date, and its easier to upgrade. Maybe in the long run commercial distributions in general are doomed, but for now Mandrake fills a useful niche and even though I don't use their distro myself, I might donate to them were I not completely broke.
Good to know I'm not the only one having problems with this. None of the non-RPM release files seemed to have it enabled. Enough people are reporting it to work that I assume the RPMs must support it. Hopefully when the slackware packages come out on linuxpackages.net in the next couple of weeks they'll have it compiled in (they're built off of the official source rpms after all).
They didn't become russians for many thousands of years after crossing the land bridge. Russia's conquest of siberia actually postdates columbus.
ACPI has a lot of benefits, and the problem isn't really ACPI per se, but the poor support for ACPI in free operating systems is the real problem here. ACPI has been around for a while (my 4-year old socket 7 motherboard supported it optionally), and the PCI IRQ sharing that this person is griping about is actually part of the PCI specification and should be supported by the operating system exclusively of whether ACPI works or not. It does enough things better than before that its likely to be standard pretty soon. And if the linux (and bsd) acpi developers don't get on the ball, there could be no new notebooks at all with working power management in free operating systems within a year. This is no different than Microsoft demanding that system makers remove floppy drives and ISA slots. Which they've been doing or will do soon. Rather than whining about it would be much better for someone to write decent ACPI drivers.
Hasn't gcc been ported to the VAX? Seemed like it would have to have been, or the NetBSD Port would be impossible.
You're sure it wasn't a Sparc box with one of the add-on PC compatibility cards (which are really seperate computers on a card, with an Intel CPU, dedicated RAM, etc.)?
I've often been intrigued by some of the older Unix workstations, particularly Alphas (for their wide compatibility with PC hardware, of which I have an abudance, and the mystique they carried when they were new). Articles like this insist that people are just throwing these things away, and you can get truckloads of them for nominal cost.
But everything I've seen, on eBay and elsewhere on the net, has been, while maybe inexpensive and even cheap, totally out of proportion to the cost for older PC and even Mac hardware. As the benchmarks in that article show, a 21066 Multia with no cache is barely faster than a 486 at half the clock speed. And yet a loaded multia can still sell for upwards of $200. And the AT-format 21066 board based on the same architecture as the Multia can cost $50 alone (with CPU). I can get a box of 486 or Pentium boards for that much. And of course there is much more abundant binary-packaged software that will run on those.
Allegedly Palm is using the Be technology as the basis for their next OS. The current Palm OS can't really do the type of things that, thanks to Microsoft et. al, everyone thinks a handheld device needs to do.
Also don't think that even if they get some money, they will return as a viable software company. More likely any money they get will go to their creditors.
I don't think you run into legal issues until you start selling them commercially. Then they probably need to be FCC certified, which costs losts of money, etc. Also, there may be some restrictions on maximum signal strength on the ISM bands that 802.11b uses...but that, I'm not sure about, they might just regulate power output.
First of all, there is already a "modified" RedHat out there, Peanut linux, which can be installed on more minimal systems. Second, Slackware and Debian, which use simple text based installers, can already be installed on machines with as little as 8 megabytes of RAM, and they aren't cut-down mini-distros, but real distributions which include lots of packages and can scale to almost any task. RedHat, with its resource-guzzling graphical installer and auto-configuration systems (which are absolutely useless and border on counter-productive on old machines with lots of non-PnP ISA hardware), is, with the possible exception of Mandrake, the worst possible basis I can think of for a minimalist linux distribution.
When I saw this, what came to mind was my memory of having installed Slackware 3.2 (kernel 2.0.30 IIRC) on a 386SX with 4 megabytes of RAM about 4 years ago. And I ran X on it (sort of)! To think that their target is "32mb or less", when the system requirements of quite a bit of the base software have not changed a lot, is ridiculous. There is a need for something that can install on machines with really low memory...I don't think the trick i used to get slackware 3 on my 386 (not mounting the initial root FS on a ramdisk, creating a swap partition and adding it immediately, using two floppy drives) would work with current versions of slackware. But this isn't it, not even close.
I seem to remember there being a program called "dosemu" that was bundled with a lot of linux distributions in the past, which could run a virtualized dos session from *nix. In fact it used FreeDOS by default IIRC. I don't know what became of it though.
But FreeDOS itself is a standalone operating system, a drop-in replacement for older versions of DOS.
It was ported, IIRC it was even released. You can find a "ppc" directory on NT 3.51 and 4.0 CDs. IBM has been balking on open PPC hardware for almost 10 years.
Hard to say. NT on Alpha development was always funded by DEC, then Compaq, to a certain extent. Compaq pulled funding and support for it, and Microsoft stopped development (they actually kept using it to develop the 64-bit version until Itanium hardware became available). There are beta versions of Win2k for alpha, but the final version didn't support it.
I didn't say anything about getting it the DAY OF RELEASE. Jesus, if you're that impatient, just apt-get install mozilla-cvs.
You can usually get the latest version, as well as nightly cvs builds, from sid. And no, you don't have to upgrade to it to use the packages.