I like the basic part of your idea, it provides for healthy competition, which is good for the consumer. The only problem I see with it is that it includes Windows. Windows is a classic example of a software project that has grown beyond the ability of the developers to control. They have what, 30 million lines of code in their latest incarnation? Considering the sheer amount of code in the thing, it is surprising it does anything at all. My solution would be to force Microsoft to source out an old version of their OS (really old, I would say no later than DOS 6.22) to a few companies like you suggested, or maybe just go all out and GPL the thing. Keeping it to a few companies is probably better, as compatibility could become a nightmare with 20 different versions of the OS out there. Then the companies would be interested in producing the best version of the OS, so theirs would sell better. This might be unrealistic, but I think some guidelines should be set out that the companies must follow when programming their OS:
1: It must be stable. This should be the top priority.
2: It must be reasonably secure. Security holes happen in the best projects (look at Linux, it is a security nightmare, though not as bad as Windows, at least you can lock down a Linux install without too much trouble), but careless bugs must not be present.
3: API's must be available and well documented. Ideally, header files and compilers should be available with the OS, as an option during installation. If the companies do not want to distribute this with the OS, then they should make a development kit available at a price individuals can easily afford (under $100).
4: Standard system calls should be present. By this I mean UNIX system calls. These are the only standard ones. This goes along with the API thing, but if these are available, then UNIX software can be built easily on the new OS.
5: The user interface must be: A) reasonably consistent across the different versions. B) difficult and full of cryptic commands. This isn't a typographical error or anything. From past experience, I have seen that it must be impossible for someone to create an interface that is both easy to use and stable. I know it can be done, but I also know that when left to these companies, it won't. As stability is more important than ease of use, ease of use must be sacrificed. C) tasteful. This is really a matter of personal preference, but it will pretty much take care of itself if item B is upheld.
6: This kind of goes along with the tasteful UI, and it really isn't that important, but the OS shouldn't have a stupid name. The name "Windows" is stupid. Make the name cryptic like the interface and it will be fine. Basically, it should not be immediately apparent what the name means when you look at it.
You might ask why I think things should be cryptic and difficult to use. It is because ever since computers have become easy to use and accessible to the masses, they have turned into cheap, flimsy, unreliable shit. Because of this, I think computers shouldn't be accessible for the masses to play with. I think that they should be available to anyone who wants one, but the consumer should actually have to learn how to use the damn thing (god forbid). Also, cheap (less than $2000, hell, less than $3500!) PC's should NOT be available. That way computers will again be regarded as a tool to accomplish some meaningful task, not a toy to play games on. Hell, let's get the hardware makers to go back to the 486 and start again, maybe they won't let the retards in Marketing royally fuck things up this time. Well, it can't be all Marketing's fault. Some idiot had to design the ATX spec for power supplies and motherboards and implement it, for example.
Anyway, this rant is well off-topic by now, so I'll just shut up now. And, while you are at it, might as well moderate me down to -8, Troll/Flamebait.
The virus will just sit there like a dumb shit unless you have Windows Scripting Host installed. I don't think this is an option with NT, so you can't be infected by it. Scripting Host is the real key to the thing. It will mess up your system even if you don't have Outlook installed, all it needs Outlook for is to propagate to everyone in your address book.
It's the other way around: WinNT contains code from OS/2. Originally OS/2 was going to be a collaboration between IBM and Micro$oft, but they went their separate ways after a short time. If OS/2 were opensourced, the only thing that could not be included would be the DOS layer, as that belongs to Micro$oft.
Actually, Intel or other companies still support P90 chips, and older. They don't for the purpose of putting them into PC's, obviously, as there are better alternatives, but these chips are good solutions for some embedded applications. I think you can still get 486 processors as well. (They cost about $2, if anyone knows if there is a 4 or 8 processor board available for this let me know!)
I have thought about doing the first post thing too, I just never get there in time.
You are right, I can't tell you what to do with your own machine, and I don't think MP3 files are an "inappropriate" use of a computer, just that it is wasteful when this is the ONLY use a computer gets. Computers have many uses, and to spend $2K on a 500 mHz box to use exclusively for something that you can do with a P75 is a waste. At least create something new with the computer if you are going to spend that much on it.
And yes, I do think seti@home is a waste of time. You are running an unoptimized program which spends most of its time with NOP's so it won't process the data too fast and run out, processing meaningless data, because even if some idiot alien civilization bothered to send out a radio signal (I say idiot because radio waves travel so slowly that the civilization would crumble to dust before the signal got past the nearest star) and we recieved it by some strange chance, there is almost no chance that it could even be decoded.
No, I don't go around deleting screensavers. I don't run any on my machines, but I quite frankly don't give a fuck what is run on others. I am entitled to my opinion on the usefulness of such things, however. And I hate getting flak from people because I deleted their illegal music files when I reimaged their machine. I can back up a few megabytes of data for people, but I don't have time to copy tens of gigabytes of music files out before I format a drive.
I haven't policed a computer lab in years! Brings back memories! I don't think Quake was released when I did, we did play Doom a lot in the evenings when not too many people used the lab. We have traditionally reloaded the drives of public machines daily, so students installing games and playing them isn't a problem as long as the machines aren't needed for real work. I'm doing systems and network administration these days (as well as servicing those PC's we sold to students), and while we're on the subject, I haven't blocked Napster! Its bandwidth is throttled, as our two T3's can't support over a thousand Napster servers running all the time. Gnutella is completely blocked (very easy to do, contrary to the authors' opinions) until an effective way of throttling it without throttling most other traffic along with it can be found. My opinions on things may suggest otherwise, but I feel that it isn't any of my business what people are doing, although certain things are throttled so that network resources will still be available for more academic purposes.
Disagreeing with the general opinion means that when I wrote my first post, there were about ten "MP3's good, RIAA bad" posts in the forum, and nothing else. By the time I finished posting, there was more variety, so I can see why you didn't get what I was talking about. And I'm not "imposing my arbitrary set of priorities" on anyone, I was stating my opinion on the subject. Apparently the days when you could discuss things in a polite manner on slashdot are long past. Oh, well, I did expect to be flamed, thanks for not disappointing! At least there was something there when I checked back.
You know, I am probably going to get flamed because I disagree with the general opinion here, but here goes...
First, I completely support Metallica in their lawsuit. The band feels that their music is their unique art form, and they do not approve of it being traded as a commodity on the 'net, as in their eyes those trading it do not respect that the music is art. As a fellow musician, I can see where they are coming from, and I think they are completely justified in their pursuit of this. I don't object to the downloading and playing of MP3 files, and I have downloaded a few myself, but it is something that too easily becomes an obsession, and that is what I object to, mostly because it is a waste of a good computer to fill the drive with music and do nothing but play music on it. I work for a college computing center, and we provide full service on machines students purchase from the college. On average two or three students bring in a machine every week that has had its Win98 install toasted (because Windoze SUCKS!). We have drive images on bootable CD-R's to quickly restore the machine to its default settings, but all data on the drive is lost. When we tell the owner this, they don't care about the term papers saved on it that they spent hours creating, but God forbid if even ONE of the damn songs they spent thirty seconds on is missing when they get the computer back. Maybe I am way off here, but I think that is a pretty fucked up set of priorities.
Next, and again maybe I am wrong, but isn't the unauthorized trafficking of copyrighted material ILLEGAL? You know, whether a law is unfair or not, we still have to abide by it. If we break laws en masse it could have unpleasant repercussions, such as mass ignorance of other, more important laws.
Also, I feel that musicians, just like programmers, have the right to choose their own license (Stallman will probably hate me now). If you want to distribute something for free, fine. If you want to be paid for it, that is fine too, just don't expect me to buy expensive crap. Granted, record companies do charge sickening amounts of money for CD's, but that doesn't give anyone an excuse to ignore their copyright.
Interesting you should mention Photoshop. You are right, it certainly isn't optimized for AMD chips, but benchmarks show an Athlon to run it faster than a PIII of the same clock speed. I think it probably has to do with the different pipelining strategies between the chips (the Athlon uses shorter pipelines for its integer and floating-point math units than the PIII).
The Celeron may have 128K L2 cache as opposed to the PIII with 256K, but on the Celeron the L2 cache runs at the core speed instead of 1/2 the core speed (PIII). Also, a Celeron runs cooler than a PIII, so you get a longer CPU life. And it's faster overall (especially overclocked by 200 mHz).
Since the topic is Athlon vs. PIII, the Athlon has 128K L1 (yes L1) cache, while the PIII has a measly 32K. This pretty much cancels any adverse effects the slower L2 cache may have on performance.
There was an incompatibility between video cards with Nvidia's GeForce 256 chipset and motherboards with the AMD-750 core-logic chipset. This was really more of a driver issue under Windows, where the system would experience occasional hard freezes (not a BSOD). The workaround was a driver update from Nvidia and AMD, and it involved temporarily restricting the AGP bus speed to 1x until a better solution could be implemented (sounded like they wanted to get a quick solution out the door to please their customers). This was just a Win9x issue anyway, and from what I have seen, it works fine anyway (no more problems than any other hardware under Windows).
Newer x86 processors (Intel, AMD, Cyrix, etc.) aren't very good for real-time work. This is because they are so fast that they sit idle most of the time waiting for instructions to come from much slower RAM. They work, but it is overkill when you can get a 68K processor from Motorola for a few dollars, or an HC12 or something for a dollar, that will do the same thing with much lower power requirements. (By the way, just because I suggest Motorola chips doesn't mean I am endorsing Apple. I am trying to give you some information based on my experience in real-time computing, and an old 68000 is a good chip for this kind of thing.)
For straight number-crunching, a fast x86 CPU is probably a decent choice. You will probably want to go SMP for maximum performance. (Are there SMP boards for a G4? That might be a good choice, as it has one of the best FPUs out there right now.) You might also want to look into an Alpha, they have very good floating point, and some really good SMP solutions exist for them.
Another issue to keep in mind is the software side of things. You may want to get a processor that you can write, or can learn to write, assembly code for. You can write much more efficient code in assembly, and (and this isn't always a bad idea) if you do things this way, you can make things even more efficient and not run an OS at all. (Even a 12mHz 386 screams when you don't run an OS on it while you run your code.)
The thing to keep in mind when approaching this problem is that there are a lot more solutions than Intel and AMD, and many of these are well worth investigating further. I am just trying to give you a few ideas here, you will probably want to do your own research and choose the solution that works best for you.
dialpad.com works fine under Linux/Netscape (it is a Java applet, if I remember correctly), and it is the only one I have seen that be used to call an actual telephone.
Instead of using a mechanical device to translate actual movement into simulated movement, use electrodes attached to the brain to intercept movement signals and translate them directly. And while we're at it, why don't we do all other I/O this way? I admit it probably can't be done at this point, and there are some potential problems (make it too real and it might screw with someone's head real good, and what if the simulation crashes (picture Micro$oft DirectMind or something as the API for this thing)), but the technology can probably be developed, and if done right, could probably make for a pretty realistic sim.
>Frankly, I think that RedHat is going to be in >seriouis trouble once a lot of people realize >that they're recent releases (4.0, 5.0, 6.0, 6.1 >for instance) are horrible and that the >competition (Linux-Mandrake for instance) is way >ahead in terms of actually getting things working >correctly.
I have installed both RedHat and Mandrake's 6.0 (i386) releases on a number of machines, including laptops. I have found RedHat's distro to be much better than Mandrake's in terms of things working correctly. For example, Mandrake 6.0 has a problem where filesystems may not be unmounted correctly during system shutdown. I think this is an issue with Mandrake's 2.2.9 kernel, as when I use the Mandrake kernel source rpms on a RedHat system and compile a kernel, the problem will occur on that system. A "regular?" 2.2.9 kernel does not have this issue. Linux-Mandrake's advantage, IMHO, is that it has some useful packages that aren't included in RedHat's distro (some of these made it into 6.1, rock on RedHat!) and it has some nice updated versions of others. My solution to this has been to install the less buggy RedHat, and do some upgrades from the Mandrake CD.
The thing I don't like about both distros is the initialization scripts. I would really prefer a *BSD style bootup, rather than that modified SySV thing.
Slashdot Mirror
I like the basic part of your idea, it provides for healthy competition, which is good for the consumer. The only problem I see with it is that it includes Windows. Windows is a classic example of a software project that has grown beyond the ability of the developers to control. They have what, 30 million lines of code in their latest incarnation? Considering the sheer amount of code in the thing, it is surprising it does anything at all. My solution would be to force Microsoft to source out an old version of their OS (really old, I would say no later than DOS 6.22) to a few companies like you suggested, or maybe just go all out and GPL the thing. Keeping it to a few companies is probably better, as compatibility could become a nightmare with 20 different versions of the OS out there. Then the companies would be interested in producing the best version of the OS, so theirs would sell better. This might be unrealistic, but I think some guidelines should be set out that the companies must follow when programming their OS:
1: It must be stable. This should be the top priority.
2: It must be reasonably secure. Security holes happen in the best projects (look at Linux, it is a security nightmare, though not as bad as Windows, at least you can lock down a Linux install without too much trouble), but careless bugs must not be present.
3: API's must be available and well documented. Ideally, header files and compilers should be available with the OS, as an option during installation. If the companies do not want to distribute this with the OS, then they should make a development kit available at a price individuals can easily afford (under $100).
4: Standard system calls should be present. By this I mean UNIX system calls. These are the only standard ones. This goes along with the API thing, but if these are available, then UNIX software can be built easily on the new OS.
5: The user interface must be:
A) reasonably consistent across the different versions.
B) difficult and full of cryptic commands. This isn't a typographical error or anything. From past experience, I have seen that it must be impossible for someone to create an interface that is both easy to use and stable. I know it can be done, but I also know that when left to these companies, it won't. As stability is more important than ease of use, ease of use must be sacrificed.
C) tasteful. This is really a matter of personal preference, but it will pretty much take care of itself if item B is upheld.
6: This kind of goes along with the tasteful UI, and it really isn't that important, but the OS shouldn't have a stupid name. The name "Windows" is stupid. Make the name cryptic like the interface and it will be fine. Basically, it should not be immediately apparent what the name means when you look at it.
You might ask why I think things should be cryptic and difficult to use. It is because ever since computers have become easy to use and accessible to the masses, they have turned into cheap, flimsy, unreliable shit. Because of this, I think computers shouldn't be accessible for the masses to play with. I think that they should be available to anyone who wants one, but the consumer should actually have to learn how to use the damn thing (god forbid). Also, cheap (less than $2000, hell, less than $3500!) PC's should NOT be available. That way computers will again be regarded as a tool to accomplish some meaningful task, not a toy to play games on. Hell, let's get the hardware makers to go back to the 486 and start again, maybe they won't let the retards in Marketing royally fuck things up this time. Well, it can't be all Marketing's fault. Some idiot had to design the ATX spec for power supplies and motherboards and implement it, for example.
Anyway, this rant is well off-topic by now, so I'll just shut up now. And, while you are at it, might as well moderate me down to -8, Troll/Flamebait.
The virus will just sit there like a dumb shit unless you have Windows Scripting Host installed. I don't think this is an option with NT, so you can't be infected by it. Scripting Host is the real key to the thing. It will mess up your system even if you don't have Outlook installed, all it needs Outlook for is to propagate to everyone in your address book.
It's the other way around: WinNT contains code from OS/2. Originally OS/2 was going to be a collaboration between IBM and Micro$oft, but they went their separate ways after a short time. If OS/2 were opensourced, the only thing that could not be included would be the DOS layer, as that belongs to Micro$oft.
Actually, Intel or other companies still support P90 chips, and older. They don't for the purpose of putting them into PC's, obviously, as there are better alternatives, but these chips are good solutions for some embedded applications. I think you can still get 486 processors as well. (They cost about $2, if anyone knows if there is a 4 or 8 processor board available for this let me know!)
I have thought about doing the first post thing too, I just never get there in time.
You are right, I can't tell you what to do with your own machine, and I don't think MP3 files are an "inappropriate" use of a computer, just that it is wasteful when this is the ONLY use a computer gets. Computers have many uses, and to spend $2K on a 500 mHz box to use exclusively for something that you can do with a P75 is a waste. At least create something new with the computer if you are going to spend that much on it.
And yes, I do think seti@home is a waste of time. You are running an unoptimized program which spends most of its time with NOP's so it won't process the data too fast and run out, processing meaningless data, because even if some idiot alien civilization bothered to send out a radio signal (I say idiot because radio waves travel so slowly that the civilization would crumble to dust before the signal got past the nearest star) and we recieved it by some strange chance, there is almost no chance that it could even be decoded.
No, I don't go around deleting screensavers. I don't run any on my machines, but I quite frankly don't give a fuck what is run on others. I am entitled to my opinion on the usefulness of such things, however. And I hate getting flak from people because I deleted their illegal music files when I reimaged their machine. I can back up a few megabytes of data for people, but I don't have time to copy tens of gigabytes of music files out before I format a drive.
I haven't policed a computer lab in years! Brings back memories! I don't think Quake was released when I did, we did play Doom a lot in the evenings when not too many people used the lab. We have traditionally reloaded the drives of public machines daily, so students installing games and playing them isn't a problem as long as the machines aren't needed for real work. I'm doing systems and network administration these days (as well as servicing those PC's we sold to students), and while we're on the subject, I haven't blocked Napster! Its bandwidth is throttled, as our two T3's can't support over a thousand Napster servers running all the time. Gnutella is completely blocked (very easy to do, contrary to the authors' opinions) until an effective way of throttling it without throttling most other traffic along with it can be found. My opinions on things may suggest otherwise, but I feel that it isn't any of my business what people are doing, although certain things are throttled so that network resources will still be available for more academic purposes.
Disagreeing with the general opinion means that when I wrote my first post, there were about ten "MP3's good, RIAA bad" posts in the forum, and nothing else. By the time I finished posting, there was more variety, so I can see why you didn't get what I was talking about. And I'm not "imposing my arbitrary set of priorities" on anyone, I was stating my opinion on the subject. Apparently the days when you could discuss things in a polite manner on slashdot are long past. Oh, well, I did expect to be flamed, thanks for not disappointing! At least there was something there when I checked back.
You know, I am probably going to get flamed because I disagree with the general opinion here, but here goes...
First, I completely support Metallica in their lawsuit. The band feels that their music is their unique art form, and they do not approve of it being traded as a commodity on the 'net, as in their eyes those trading it do not respect that the music is art. As a fellow musician, I can see where they are coming from, and I think they are completely justified in their pursuit of this. I don't object to the downloading and playing of MP3 files, and I have downloaded a few myself, but it is something that too easily becomes an obsession, and that is what I object to, mostly because it is a waste of a good computer to fill the drive with music and do nothing but play music on it. I work for a college computing center, and we provide full service on machines students purchase from the college. On average two or three students bring in a machine every week that has had its Win98 install toasted (because Windoze SUCKS!). We have drive images on bootable CD-R's to quickly restore the machine to its default settings, but all data on the drive is lost. When we tell the owner this, they don't care about the term papers saved on it that they spent hours creating, but God forbid if even ONE of the damn songs they spent thirty seconds on is missing when they get the computer back. Maybe I am way off here, but I think that is a pretty fucked up set of priorities.
Next, and again maybe I am wrong, but isn't the unauthorized trafficking of copyrighted material ILLEGAL? You know, whether a law is unfair or not, we still have to abide by it. If we break laws en masse it could have unpleasant repercussions, such as mass ignorance of other, more important laws.
Also, I feel that musicians, just like programmers, have the right to choose their own license (Stallman will probably hate me now). If you want to distribute something for free, fine. If you want to be paid for it, that is fine too, just don't expect me to buy expensive crap. Granted, record companies do charge sickening amounts of money for CD's, but that doesn't give anyone an excuse to ignore their copyright.
Interesting you should mention Photoshop. You are right, it certainly isn't optimized for AMD chips, but benchmarks show an Athlon to run it faster than a PIII of the same clock speed. I think it probably has to do with the different pipelining strategies between the chips (the Athlon uses shorter pipelines for its integer and floating-point math units than the PIII).
The Celeron may have 128K L2 cache as opposed to the PIII with 256K, but on the Celeron the L2 cache runs at the core speed instead of 1/2 the core speed (PIII). Also, a Celeron runs cooler than a PIII, so you get a longer CPU life. And it's faster overall (especially overclocked by 200 mHz).
Since the topic is Athlon vs. PIII, the Athlon has 128K L1 (yes L1) cache, while the PIII has a measly 32K. This pretty much cancels any adverse effects the slower L2 cache may have on performance.
There was an incompatibility between video cards with Nvidia's GeForce 256 chipset and motherboards with the AMD-750 core-logic chipset. This was really more of a driver issue under Windows, where the system would experience occasional hard freezes (not a BSOD). The workaround was a driver update from Nvidia and AMD, and it involved temporarily restricting the AGP bus speed to 1x until a better solution could be implemented (sounded like they wanted to get a quick solution out the door to please their customers). This was just a Win9x issue anyway, and from what I have seen, it works fine anyway (no more problems than any other hardware under Windows).
Newer x86 processors (Intel, AMD, Cyrix, etc.) aren't very good for real-time work. This is because they are so fast that they sit idle most of the time waiting for instructions to come from much slower RAM. They work, but it is overkill when you can get a 68K processor from Motorola for a few dollars, or an HC12 or something for a dollar, that will do the same thing with much lower power requirements. (By the way, just because I suggest Motorola chips doesn't mean I am endorsing Apple. I am trying to give you some information based on my experience in real-time computing, and an old 68000 is a good chip for this kind of thing.)
For straight number-crunching, a fast x86 CPU is probably a decent choice. You will probably want to go SMP for maximum performance. (Are there SMP boards for a G4? That might be a good choice, as it has one of the best FPUs out there right now.) You might also want to look into an Alpha, they have very good floating point, and some really good SMP solutions exist for them.
Another issue to keep in mind is the software side of things. You may want to get a processor that you can write, or can learn to write, assembly code for. You can write much more efficient code in assembly, and (and this isn't always a bad idea) if you do things this way, you can make things even more efficient and not run an OS at all. (Even a 12mHz 386 screams when you don't run an OS on it while you run your code.)
The thing to keep in mind when approaching this problem is that there are a lot more solutions than Intel and AMD, and many of these are well worth investigating further. I am just trying to give you a few ideas here, you will probably want to do your own research and choose the solution that works best for you.
It's not integrated at all. Go to 98lite.net and see for yourself.
Of course you could find three bugs in Linux faster. It's one of the wonderful things about Open Source.
Try traceroute, it should show you a real gateway somewhere in the list. It might be called TRACERT in NT, I don't know for sure.
dialpad.com works fine under Linux/Netscape (it is a Java applet, if I remember correctly), and it is the only one I have seen that be used to call an actual telephone.
Instead of using a mechanical device to translate actual movement into simulated movement, use electrodes attached to the brain to intercept movement signals and translate them directly. And while we're at it, why don't we do all other I/O this way? I admit it probably can't be done at this point, and there are some potential problems (make it too real and it might screw with someone's head real good, and what if the simulation crashes (picture Micro$oft DirectMind or something as the API for this thing)), but the technology can probably be developed, and if done right, could probably make for a pretty realistic sim.
>Frankly, I think that RedHat is going to be in
>seriouis trouble once a lot of people realize
>that they're recent releases (4.0, 5.0, 6.0, 6.1
>for instance) are horrible and that the
>competition (Linux-Mandrake for instance) is way
>ahead in terms of actually getting things working
>correctly.
I have installed both RedHat and Mandrake's 6.0 (i386) releases on a number of machines, including laptops. I have found RedHat's distro to be much better than Mandrake's in terms of things working correctly. For example, Mandrake 6.0 has a problem where filesystems may not be unmounted correctly during system shutdown. I think this is an issue with Mandrake's 2.2.9 kernel, as when I use the Mandrake kernel source rpms on a RedHat system and compile a kernel, the problem will occur on that system. A "regular?" 2.2.9 kernel does not have this issue.
Linux-Mandrake's advantage, IMHO, is that it has some useful packages that aren't included in RedHat's distro (some of these made it into 6.1, rock on RedHat!) and it has some nice updated versions of others. My solution to this has been to install the less buggy RedHat, and do some upgrades from the Mandrake CD.
The thing I don't like about both distros is the initialization scripts. I would really prefer a *BSD style bootup, rather than that modified SySV thing.
I don't know if this is exactly what you had in mind, but you can use:
.rpm file from a tarball.
rpm --tarbuild [tarball]
to create a
Hope this helps.
Chapters 1 & 2 can be read here.
this is set up for the root account by default, in the file ~/.bashrc. just kill the line aliasing 'rm -i' onto 'rm'.