You know, they could also make gcc output "Red Hat is your friend. Trust Red Hat. Buy Red Hat products" on your speakers when you compile something.
But they won't. And even if they did, SuSE could just remove it (it's their right with GPL software)
What Red Hat gains from this (apart from revenue, which cygnus has been making quite a lot of) is several top-notch compiler/development environment developers.
Some people say Red Hat is monopolizing the Linux software market. Is that really even possible since you are giving out all your software?
The Linux support market is a different matter. They certainly have most of the top developers working for them, but is that really a bad thing? It's not like they're not paying them for their work. With stock options and all that they probably are doing quite well too. As are the shareholders.
It is necessary to have companies which can answer questions from "I get this login prompt and don't know what to do?" to "gcc creates broken code when cross-compiling our code to this and that architecture, here's a code sample" if Linux wants to be a credible operating system.
Except that getting Java to work properly on anything except green threads (which is a userland thread-emulation package), win32 threads or Solaris threads is a real bitch.
It took the blackdown people ages to get it right (in their latest 1.1.x release it's pretty good and supposedly they've gotten it to work with 1.2 too after a _lot_ of work). Just look at the amount of non-solaris/win32 ports of java2 that aren't in alpha/beta. There's not too many of them, is there.
I wouldn't expect it to be any easier for the Be people. No matter how elegant your threads are (some people argue linuxthreads are just that), that won't help you at all if your threads dont work the same way as they do in Solaris.
Oh, and this is UK law apparently (and looking at his e-mail address the guy in trouble is in the UK)
I really hope this doesn't end in some poor invidual being sued for everything he owns by that multi-billion DVD cabal (who thinks MS is too small to join them)
Nope, the idea of this project if I understood correctly is not to create a new language people around can use to talk to eachother, it's meant more for machines. English doesn't really qualify as it's a real language and thus doesn't have the required properties as it is ambiguous in many ways. I'm not familiar with esperanto, but I would assume it doesn't contain all the necessary information either to be useful as an intermediate language as it was designed to be a human language. Take for example the word "uncle". Is that on your mothers or fathers side? In english it's impossible to say but in some languages the difference is very important. If your translator program is smart it might guess correctly from the context (or atleast state it doesn't know so maybe a human operator can then choose one) Ok, so maybe things would be much easier if everyone spoke english, but that is not the case and won't be in the foreseeable future. In the meanwhile automatic translation is needed badly and this sounds like a good approach to the problem.
I heard an interesting story about 6 months ago in a seminar from an security researcher about an unnamed international company wanting to connect their UK network to the one somewhere in the middle-east. Being security conscious people they encrypted the connection (no idea if it was some 40-bit mickey mouse crypto from some american company or something decent) Everything worked just nicely. However, after a month they got a call from someone claiming to represent the french goverment asking them to stop encrypting their VPN. The next thing they did was to ask their telco to reroute the connection so it didn't go over anywhere near France;)
Yes I have compiled kernels with egcs and run them all the time. My point is that pgcc hasn't been tested nearly as heavily as egcs (which other major distributions have been using for quite a long time and which cygnus supports commercially)
I don't think the users of a "linux suitable for newbies" should be the ones to be the ones to find out if the compiler generates bad code or not. If it was "hackers linux" we were talking about I wouldn't mind at all, since the users would (hopefully) know what they're getting into.
This from the PGCC faq:
Older Linux kernels had many problems with illegal assembly constructs which prevented them from being compiled with egcs or pgcc (see the next question for details). Work has been done to correct these problems, and 2.2 kernels are much improved. However, there are still a number of problems which can cause kernels built using PGCC to fail to work correctly and neither the PGCC maintainers nor the kernel maintainers officially support this. While many people are successfuly using PGCC built kernels you may experience problems, and if you do the PGCC maintainers are only really interested in bug reports which also identify the source of the problem.
The kernel problem I'm referring to was a driver (not one of the simpler ones either) failing to work. Sure it could have been an user error, but the logical conclusion (same hardware, same kernel, same driver version) really would be that something got miscompiled with pgcc.
Don't see this as a "pgcc sucks" post. pgcc has it's purpose as a testbed for new x86 optimizations in egcs, but IMHO it really doesn't belong in a distribution that claims to be suitable for newbies.
Pentium optimization seems to be one of the major points in mandrake these days. However, I _really_ wouldn't trust my production systems to code generated by an experimental compiler.
Even egcs,which is quite conservative compared to pgcc, has had it's problems with generating correct code (especially for things like the kernel).
The amount of speed gain you get from pgcc isn't _THAT_ big in the common case. Sure you get impressive speed gains with something like gzip when you use -O99 -fexperimental-option-of-the-month -m686, but for most things it's a few percent max. glibc already has inline assembly for string handling/math.
For userland you probably can get away with it without too many critical problems, but using pgcc for the kernel is just irresponsible. I know of atleast one kernel developer who ignores reports from mandrake users because one "problem" went away after the user switched to another distribution that doesn't compile their kernel with pgcc.
Someone I know listened to Linus' talk in Helsinki a few days ago, and apparently someone asked him about NSA's implant chip that can be used to read peoples minds.
Linus replied: "Well, I really can't talk about the things Transmeta does"
The thing to remember about Nokia is that it's really a pile of smaller units (cell phones, monitors, phone exchanges, misc r&d etc.) and the different units do things very differently.
Some use VMS as a development platform (or did atleast a few years ago), some think NT is the solution to everything including world hunger and others (generally the ones that don't do end-user products) use whatever does the job best. Linux has increasingly been just that thing for quite a lot of stuff.
If I understood correctly this was done by the multimedia terminal people whereas the information the gnokii people want is from the cell phone people and they seem to want to keep their stuff pretty secret.
From what I've heard there are only a few games for the dreamcast that actually use Windows CE, and those are pretty unknown crappy ones too. All the others use the hardware more or less directly (probably some proprietary library involved, but I really don't know the details that well)
Just remove -ansi -pedantic from config/cf/xfree86.cf (atleast if you have glibc 2.1). The problem is in that it also #defines _GNU_SOURCE, which causes the glibc header files to have lots fun stuff that's not ANSI compliant.
Also, before upgrading you should know that it seems to still be in a very experimental state, I'm seeing funnyness like nothing happening after clicking a link in netscape.
14% of 69 million stocks at $72.625 each, so it's like $700M now. Funny how these large amounts of money that are so ridiculously large I can't even understand how large they really are change so much in just a few moments...
Well, you do have choices even now for accelerated 3D under linux
1) 3dfx
Binary-only drivers using glide. It's worked under linux for ages.
2) Matrox G200/G400
Vendor supplied near, but not complete documentation.
3) TNT/TNT2
Vendor supplied/supported 3D acceleration with full source.
Not a hard choice for me =) (well, ok, G400 is still a possibility, it seems to be slightly cheaper and some of the features are really nice. Not that I have space for two monitors, so the dual head support is useless)
Red Hat makes good products Good products sell well Selling well means getting lots of money Lots of money buys lots of hackers Hackers create free software Free software makes me happy Ergo Red Hat makes me happy i used to be a bootdisk-0.95 user, then a sls user, then a slackware user, briefly a debian user and back to slackware, but after RH 2.1 I really haven't wanted to tinker with my machine enough to compile every single new package I install from source, I really have better things to do with my computer than around installing new versions of packages from source.
In 2.0.37pre up to approximately 3.5 gigs is supported (and I believe work is being done to support the kludgy stuff that lets xeons do up to 64G in a sane as possible way)
Oh, and I believe Linux has been succesfully booted on a 64-cpu ultrasparc. That's certainly more than NT will do for a very long time I know, using them efficiently is a very different matter, but what I do know is people using multiple 6xppro linux systems happily in heavy production use. It might not be optimal with > 4 cpu's (yet), but well, NT isn't either:)
Sash is a simple, standalone, statically linked shell which includes simplified versions of built-in commands like ls, dd and gzip. Sash is statically linked so that it can work without shared libraries, so it is particularly useful for recovering from certain types of system failures. Sash can also be used to safely upgrade to new versions of shared libraries.
Slashdot Mirror
You know, they could also make gcc output "Red Hat is your friend. Trust Red Hat. Buy Red Hat products" on your speakers when you compile something.
But they won't. And even if they did, SuSE could just remove it (it's their right with GPL software)
What Red Hat gains from this (apart from revenue, which cygnus has been making quite a lot of) is several top-notch compiler/development environment developers.
Some people say Red Hat is monopolizing the Linux software market. Is that really even possible since you are giving out all your software?
The Linux support market is a different matter. They certainly have most of the top developers working for them, but is that really a bad thing? It's not like they're not paying them for their work. With stock options and all that they probably are doing quite well too. As are the shareholders.
It is necessary to have companies which can answer questions from "I get this login prompt and don't know what to do?" to "gcc creates broken code when cross-compiling our code to this and that architecture, here's a code sample" if Linux wants to be a credible operating system.
Except that getting Java to work properly on anything except green threads (which is a userland thread-emulation package), win32 threads or Solaris threads is a real bitch.
It took the blackdown people ages to get it right (in their latest 1.1.x release it's pretty good and supposedly they've gotten it to work with 1.2 too after a _lot_ of work). Just look at the amount of non-solaris/win32 ports of java2 that aren't in alpha/beta. There's not too many of them, is there.
I wouldn't expect it to be any easier for the Be people. No matter how elegant your threads are (some people argue linuxthreads are just that), that won't help you at all if your threads dont work the same way as they do in Solaris.
Oh, and this is UK law apparently
(and looking at his e-mail address the guy in trouble is in the UK)
I really hope this doesn't end in some poor invidual being sued for everything he owns
by that multi-billion DVD cabal (who thinks MS is
too small to join them)
Nope, the idea of this project if I understood correctly is not to create a new language people around can use to talk to eachother, it's meant more for machines. English doesn't really qualify as it's a real language and thus doesn't have the required properties as it is ambiguous in many ways. I'm not familiar with esperanto, but I would assume it doesn't contain all the necessary information either to be useful as an intermediate language as it was designed to be a human language. Take for example the word "uncle". Is that on your mothers or fathers side? In english it's impossible to say but in some languages the difference is very important. If your translator program is smart it might guess correctly from the context (or atleast state it doesn't know so maybe a human operator can then choose one) Ok, so maybe things would be much easier if everyone spoke english, but that is not the case and won't be in the foreseeable future. In the meanwhile automatic translation is needed badly and this sounds like a good approach to the problem.
I heard an interesting story about 6 months ago in a seminar from an security researcher about an unnamed international company wanting to connect their UK network to the one somewhere in the middle-east. Being security conscious people they encrypted the connection (no idea if it was some 40-bit mickey mouse crypto from some american company or something decent) Everything worked just nicely. However, after a month they got a call from someone claiming to represent the french goverment asking them to stop encrypting their VPN. The next thing they did was to ask their telco to reroute the connection so it didn't go over anywhere near France ;)
Yes I have compiled kernels with egcs and run them all the time. My point is that pgcc hasn't been tested nearly as heavily as egcs (which other major distributions have been using for quite a long time and which cygnus supports commercially)
I don't think the users of a "linux suitable for newbies" should be the ones to be the ones to find out if the compiler generates bad code or not. If it was "hackers linux" we were talking about I wouldn't mind at all, since the users would (hopefully) know what they're getting into.
This from the PGCC faq:
The kernel problem I'm referring to was a driver (not one of the simpler ones either) failing to work. Sure it could have been an user error, but the logical conclusion (same hardware, same kernel, same driver version) really would be that something got miscompiled with pgcc.Don't see this as a "pgcc sucks" post. pgcc has it's purpose as a testbed for new x86 optimizations in egcs, but IMHO it really doesn't belong in a distribution that claims to be suitable for newbies.
Pentium optimization seems to be one of
the major points in mandrake these days.
However, I _really_ wouldn't trust my
production systems to code generated by
an experimental compiler.
Even egcs,which is quite conservative compared to pgcc, has had it's problems with generating
correct code (especially for things like the kernel).
The amount of speed gain you get from pgcc
isn't _THAT_ big in the common case. Sure
you get impressive speed gains with something
like gzip when you use -O99 -fexperimental-option-of-the-month -m686, but
for most things it's a few percent max.
glibc already has inline assembly for string handling/math.
For userland you probably can get away with it
without too many critical problems, but using pgcc for the kernel is just irresponsible. I know of atleast one kernel developer who ignores reports from mandrake users because one "problem" went away after the user switched to another distribution that doesn't compile their kernel with pgcc.
Someone I know listened to Linus' talk
in Helsinki a few days ago, and apparently
someone asked him about NSA's implant chip
that can be used to read peoples minds.
Linus replied: "Well, I really can't talk about the things Transmeta does"
The thing to remember about Nokia is that it's really a pile of smaller units
(cell phones, monitors, phone exchanges, misc r&d etc.) and the different units do things very differently.
Some use VMS as a development platform (or did atleast a few years ago), some think NT is the solution to everything including world hunger
and others (generally the ones that don't do
end-user products) use whatever does the job best. Linux has increasingly been just that thing for quite a lot of stuff.
If I understood correctly this was done by the
multimedia terminal people whereas the information the gnokii people want is from the cell phone
people and they seem to want to keep their stuff pretty secret.
If the DC was behind NAT/MASQ the IP would show
up to be the router that does the NAT:ing.
The open ports are consistant with this (telnet, BGP4, http), all are services that are running
on pretty much every cisco router.
From what I've heard there are only a few games
for the dreamcast that actually use Windows CE, and those are pretty unknown crappy ones too.
All the others use the hardware more or less
directly (probably some proprietary library involved, but I really don't know the details that well)
Just remove -ansi -pedantic from config/cf/xfree86.cf
(atleast if you have glibc 2.1).
The problem is in that it also #defines
_GNU_SOURCE, which causes the glibc header
files to have lots fun stuff that's not ANSI
compliant.
Also, before upgrading you should know that
it seems to still be in a very experimental state,
I'm seeing funnyness like nothing happening after clicking a link in netscape.
It runs on Intel CPU's.... and... umn...
AMD... and on a good day maybe Cyrix too!
That's 3 platforms already!
14% of 69 million stocks at $72.625 each, so it's like $700M now. Funny how these large amounts of money that are so ridiculously large I can't even understand how large they really are change so much in just a few moments...
Well, you do have choices even now for accelerated
3D under linux
1) 3dfx
Binary-only drivers using glide. It's worked under linux for ages.
2) Matrox G200/G400
Vendor supplied near, but not complete documentation.
3) TNT/TNT2
Vendor supplied/supported 3D acceleration with full source.
Not a hard choice for me =) (well, ok, G400 is still a possibility, it seems to be slightly cheaper and some of the features are really nice. Not that I have space for two monitors, so the dual head support is useless)
Red Hat makes good products
Good products sell well
Selling well means getting lots of money
Lots of money buys lots of hackers
Hackers create free software
Free software makes me happy
Ergo Red Hat makes me happy
i used to be a bootdisk-0.95 user, then a sls user, then a slackware user, briefly a debian user and back to slackware,
but after RH 2.1 I really haven't wanted to tinker with my machine enough to compile every single new package I install from source, I really have better things to do with my computer than
around installing new versions of packages from source.
In 2.0.37pre up to approximately 3.5 gigs
:)
is supported (and I believe work is being done to support the kludgy stuff that lets xeons
do up to 64G in a sane as possible way)
Oh, and I believe Linux has been succesfully
booted on a 64-cpu ultrasparc. That's certainly more than NT will do for a very long time
I know, using them efficiently is a very
different matter, but what I do know is people using multiple 6xppro linux systems happily in heavy production use. It might not be optimal
with > 4 cpu's (yet), but well, NT isn't
either
and it comes with Red Hat
Sash is a simple, standalone, statically linked shell which includes
simplified versions of built-in commands like ls, dd and gzip. Sash
is statically linked so that it can work without shared libraries, so
it is particularly useful for recovering from certain types of system
failures. Sash can also be used to safely upgrade to new versions of
shared libraries.