Interesting Linus Quote
Tim Smith
source code control. Here's what Linus had to say on the kernel list:
I would suggest that if you are silently seething about the fact that a
commercial product can do something better than a free one, how about
_doing_ something about it?
Quite frankly, I don't _want_ people using Linux for ideological reasons. I
think ideology sucks. This world would be a much better place if people had
less ideology, and a whole lot more "I do this because it's FUN and because
others might find it useful, not because I got religion."
Would I prefer to use a tool that didn't have any restrictions on it for
kernel maintenance? Yes. But since no such tool exists, and since I'm
personally not very interested in writing one, _and_ since I don't have any
hangups about using the right tool for the job, I use BitKeeper.
Some of the more outspoken advocates here should keep this in mind when they
start to get carried away.
--Tim Smith
WarpKat
All hail Linus. For he is God.
=:P
Joe Potter
Good post, and good comment on your part.
Linus T. has his faults, but he is spot on target with this one.
--
Regards, Joe
mjt
Hash: SHA1
WarpKat wrote:
>> Some of the more outspoken advocates here should keep this in mind when
>> they start to get carried away.
>
> All hail Linus. For he is God.
>
"Martin Luther, meet Linus Torvalds"
http://www.salon.com/21st/feature/1998/11/12feature.html
- --
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Michael J. Tobler: motorcyclist, surfer, # Black holes result
skydiver, and author: "Inside Linux", # when God divides the
"C++ HowTo", "C++ Unleashed" # universe by zero
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iEYEARECAAYFAjzqueQACgkQs+/mRnVxfrondgCgqXiffl2WwGKEGHtfWVPtISh3
sJIAnRliJE9y462M4Iic62szGmGVmYfA
=GMAF
-----END PGP SIGNATURE-----
Kenneth Downs
Your other post asks the question "Is Linux Free?" According to RMS, no.
Then you tell us to keep in mind Linus's attitude, which pretends that
politics and ideology do not exist. Seems to me that without the
"ideology" of the FSF, Linus would have no OS for his kernel.
...and we would have no NG in which to debate these things.
Something else to keep in mind.
--
Ken
Linux, the more you learn, the more you love
maddman
I do like the stuff that Linus says. He does this, and gives us this
wonderful kernel, because he likes to. He's an extremly practical
man, and has a good sense of humor as well.
Matthew Gardiner
He is the way, truth and the light ;)
Matthew Gardiner
Travis Sparks
> Tim Smith wrote:
>
>> Some people are upset that Linus uses BitKeeper, which is non-free, for
>> source code control. Here's what Linus had to say on the kernel list:
>>
>> I would suggest that if you are silently seething about the fact that
>> a commercial product can do something better than a free one, how
>> about _doing_ something about it?
>>
>> Quite frankly, I don't _want_ people using Linux for ideological
>> reasons. I think ideology sucks. This world would be a much better
>> place if people had less ideology, and a whole lot more "I do this
>> because it's FUN and because others might find it useful, not because
>> I got religion."
>>
>> Would I prefer to use a tool that didn't have any restrictions on it
>> for kernel maintenance? Yes. But since no such tool exists, and since
>> I'm personally not very interested in writing one, _and_ since I
>> don't have any hangups about using the right tool for the job, I use
>> BitKeeper.
>>
>> Some of the more outspoken advocates here should keep this in mind when
>> they start to get carried away.
>>
>> --Tim Smith
I agree with Linus completely, way too many linux advocates get carried away
with the idea of Linux vs. Microsoft, do I hate M$? Yes, I believe that M$
is way too concerned about money, and that the company is too big to
provide any kind of serious innovation, but people who habitualy complain
about M$ just to complain about them, these people annoy me, I prefer to
focus on Linux, where it's headed, what could be improved, how I can help,
etc. Not about what's M$ up to? What up with their license? Etc. This stuff
is just a footnote to me, true Linux enthusiasts care about the OS, not
it's competition.
Donn Miller
>
> Some people are upset that Linus uses BitKeeper, which is non-free, for
> source code control. Here's what Linus had to say on the kernel list:
Years ago, when ftp.cdrom.com was in CA and had thousands of
simultaneous users, it used a special ftp daemon that was not the stock
FreeBSD ftp daemon. It was a binary-only version of a ftp server that
one of the FreeBSD developers had developed himself, and was licensed to
the FreeBSD project.
-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 80,000 Newsgroups - 16 Different Servers! =-----
Net Resident
<reply_i...@mouse-potato.com> wrote:
Having fun is all well and good but I have seen that viewpoint carried to
a 'idealogical' extremes also.
I have zero problems with Linus using a commercial app but why would he
seemingly bite the hand that has fed Linux so much ?.
It is in large part that a well thought out, publically benificial AND fun
ideology has carried Linux to where it is today in my view. It seems to me
that he is getting a bit full of himself, I only hope I am wrong on that
last part.
Net Resident
"mjt" <mtobler@removethis_sbcglobal.net> wrote in message
news:d3zG8.12041$BN3.244...@newssvr30.news.prodigy.com...
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> WarpKat wrote:
>
> >> Some of the more outspoken advocates here should keep this in mind
when
> >> they start to get carried away.
> >
> > All hail Linus. For he is God.
> >
>
> "Martin Luther, meet Linus Torvalds"
>
> http://www.salon.com/21st/feature/1998/11/12feature.html
>
That was a cute read and it is hard to be sure if it was a tongue in
cheek piece or an apropos analogy but either way Thomas Scoville made
one colossal mistake by replacing Richard Stallman with Linus Torvalds
as the protagonist.
While Linus deserves credit for the kernal I believe that few people
would have ever heard of it if not for Richard Stallman.
Jim Richardson
Hash: SHA1
What do you mean "bite the hand that has fed Linux so much" ?
Linus said if there was an less restricted tool to do this, he'd use it.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (GNU/Linux)
iD8DBQE86yprd90bcYOAWPYRAlGjAJ9dZ7Vz8fvrUaN3OxckRRyxR/sl3wCfSqgF
RlryVCITGxKrrGDWK9bwz+8=
=cB0U
-----END PGP SIGNATURE-----
--
Jim Richardson
Anarchist, pagan and proud of it
http://www.eskimo.com/~warlock
Linux, from watches to supercomputers, for grandmas and geeks.
rapskat
Resident" - dump details are as follows...
Well, it is his baby...or is it? If Linus declared tomorrow that he was
dropping all involvement with Linux, it would still continue to be
developed and supported by millions of people worldwide.
I have no hangups about commercial applications. In truth, I would like
to see more 3rd party closed source commercial applications make their way
to the platform. There's nothing wrong with someone wanting to make a
buck, and it would facilitate the growth of the userbase if people's
favorite products ran on Linux.
I use Linux and OSS because they work well for me. Fast, stable, secure,
hassle free, etc. But would I use Windows by choice if it offered the
same benefits? As it stands right now, no. For two reasons..
The first being that I could not conscientiously support an entity like
Microsoft which deems to take away personal choice.
Secondly, why? If I already have all the benefits of Linux and OSS, why
would I go out and pay for them? Why would I submit myself and my systems
to that sort of control? It just wouldn't make sense.
That was the practical side of it, now as for the ideological side...
I would like for people to have an opportunity to make an educated choice
as to which software they run on their computers. Linux is a viable
choie, but many people are unaware or misinformed of it. It is my
intention to present Linux and OSS to as many persons as possible for the
purposes of educating them and enabling them to be able to make that
choice.
If they choose Windows, then so beit, but at least they made that choice
on their own without being coerced to make it from a lack of knowledge.
That's my stance on it.
--
rapskat - 1:55am up 8:35, 2 users, load average: 0.44, 0.19, 0.11
drop the hot to mail me
Hallelujah, I've been Bourne-Again!
Cat
Hash: SHA1
> While Linus deserves credit for the kernal I believe that few people
> would have ever heard of it if not for Richard Stallman.
And I think it bugs him a bit. See the herd. Both very clever people and have
done a truckload for free software.
- --
Cat
Linux 2.4.9-31
7:27pm up 28 days, 23:45, 6 users, load average: 0.07, 0.06, 0.08
http://www.ratrobot.com/art/art.htm Piero Manzoni sold his own excrement in
tins and Britney Spears debut album went 12xPlatinum. Both of them are canned
shit but which one is art?
-----BEGIN PGP SIGNATURE-----
Version: PGP 6.5.8
iQA/AwUBPOtlyyh0Y2LcENUAEQIOIwCgisIW7xorDcDpAmEwXWMc5v+vvvAAoNNj
scBzRTQQu2Kh/w5rCvyaayR6
=K/My
-----END PGP SIGNATURE-----
Erik Funkenbusch
> Your other post asks the question "Is Linux Free?" According to RMS, no.
>
> Then you tell us to keep in mind Linus's attitude, which pretends that
> politics and ideology do not exist. Seems to me that without the
> "ideology" of the FSF, Linus would have no OS for his kernel.
>
> ...and we would have no NG in which to debate these things.
>
> Something else to keep in mind.
Are you honestly trying to suggest that Linus (or any other Free OS)
couldn't come up with a few utilities like "tail" or "more" himself (or his
group of friends)?
Sure, Linux might not have been so comprehensive (at least as quickly as it
did) without the FSF, but IMO the only thing that Linux actually *needed*
was gcc (which is from the FSF, but even RMS doesn't claim that anything
compiled with gcc should be called GNU).
kickaha
Well put, Rappy.
Marc Geerlings
Regards Marc Geerlings
maddman
I'm one of those that sees room for both commercial and free software.
Basically, I feel that the essentials for running computers and the
internet should be Free. Operating systems, web/file/print servers,
text editors, web browsers, and compilers are the types of things that
should be not only no cost, but have the source released both for
customization and furthur development. These things should be GPL, to
prevent them from ever being made non free.
Other applications - games, complex business apps, detailed network
management apps, etc. should either be commercial or released BSD. If
the creator wants to GPL them, great! But if not that's fine too.
These apps should also be forced (by the market) to support open
standards to maintain interoperability with each other and Free apps.
Jan
Huh? Are you saying the only thing the FSF made was little stuff like
'tail'? Do you think the only thing that makes up an operating system is
the kernel? The kernel is a very important part of an operating system,
but it isn't the only part of an operating system.
Kenneth Downs
>> Tim Smith wrote:
>> Your other post asks the question "Is Linux Free?" According to RMS, no.
>>
>> Then you tell us to keep in mind Linus's attitude, which pretends that
>> politics and ideology do not exist. Seems to me that without the
>> "ideology" of the FSF, Linus would have no OS for his kernel.
>>
>> ...and we would have no NG in which to debate these things.
>>
>> Something else to keep in mind.
>
> Are you honestly trying to suggest that Linus (or any other Free OS)
> couldn't come up with a few utilities like "tail" or "more" himself (or
> his group of friends)?
>
no. i'm saying he could not have come up with the entire GNU toolset,
another thing entirely.
> Sure, Linux might not have been so comprehensive (at least as quickly as
> it did) without the FSF, but IMO the only thing that Linux actually
> *needed* was gcc (which is from the FSF, but even RMS doesn't claim that
> anything compiled with gcc should be called GNU).
thank you for your opinion.
Erik Funkenbusch
Funny, but people here seem to think the opposite, especially when it comes
to MS.
Much of what the GNU project makes is available from other sources,
including shells, command line tools, etc.. and has been since before Linux
was even announced. IMO the only real "must have" thing that GNU makes that
there is no non-GNU alternative to is gcc. Sure, there are lots of third
party compilers, but none that are easily ported to a new platform.
Rob S. Wolfram
>> Tim Smith wrote:
>> Your other post asks the question "Is Linux Free?" According to RMS, no.
>>
>> Then you tell us to keep in mind Linus's attitude, which pretends that
>> politics and ideology do not exist. Seems to me that without the
>> "ideology" of the FSF, Linus would have no OS for his kernel.
>>
>> ...and we would have no NG in which to debate these things.
>>
>> Something else to keep in mind.
>
> Are you honestly trying to suggest that Linus (or any other Free OS)
> couldn't come up with a few utilities like "tail" or "more" himself (or his
> group of friends)?
I don't think he is, but fact is that the tools were there and they made
up the GNU OS that did exist back then, short of only a kernel. Back
then GNU was recognized for what it was and stood for. Take for example
this little quote in one of Andy Tanenbaum's posts in the infamous 1992
"linux is obsolete" thread in comp.os.minix:
| Making software free, but only for folks with enough money to buy first
| class hardware is an interesting concept. Of course 5 years from now
| that will be different, but 5 years from now everyone will be running
| free GNU on their 200 MIPS, 64M SPARCstation-5.
What was this "GNU" he was talking about?
The toolset that made up GNU was quite a bit more than just "tail" or
"more". Short of the kernel it was a fully functional OS.
Cheers,
Rob
--
Rob S. Wolfram <aze...@hamal.xs4all.nl> OpenPGP key 0xD61A655D
For something that is not sustainable, we seem to keep getting more
and more of it.
-- Kenneth Downs in COLA about free software
Rob S. Wolfram
> Some people are upset that Linus uses BitKeeper, which is non-free, for
> source code control. Here's what Linus had to say on the kernel list:
> Quite frankly, I don't _want_ people using Linux for ideological reasons. I
> think ideology sucks. This world would be a much better place if people had
> less ideology, and a whole lot more "I do this because it's FUN and because
> others might find it useful, not because I got religion."
/me thinks that if that is his opinion, he chose a wrong license for
spreading his kernel. He should have chosen the BSD license instead
(which actually did exist back in '91).
ISTR that Linus was _very_ outspoken in the first releases of the kernel
and that he did not approve that others would make money on his code. If
that is not ideology, I don't know what is.
Ideology is one of the main reasons why I use a GNU OS. The Unix
paradigm is a second one.
Cheers,
Rob
--
Rob S. Wolfram <aze...@hamal.xs4all.nl> OpenPGP key 0xD61A655D
Either SCO will wake up and smell the roses, or it will soon
feed the roses.
-- Ray Dassen on slashdot
Rob S. Wolfram
> I have no hangups about commercial applications. In truth, I would like
> to see more 3rd party closed source commercial applications make their way
> to the platform. There's nothing wrong with someone wanting to make a
> buck, and it would facilitate the growth of the userbase if people's
> favorite products ran on Linux.
OK, noted.
> I use Linux and OSS because they work well for me. Fast, stable, secure,
> hassle free, etc. But would I use Windows by choice if it offered the
> same benefits? As it stands right now, no. For two reasons..
>
> The first being that I could not conscientiously support an entity like
> Microsoft which deems to take away personal choice.
Now this is contradictory to the point you made above. You have no
hangups about applications that take away your choice and then you
complain about Microsoft doing the very same thing?
The GNU OS exists for one reason and one reason only, to have an OS
where the users do not have their choice taken away. Stallman's beef
with the name is just that so many people use a GNU OS without even
knowing that. They use this "cool, stable, gratis Unix-like" OS, but
they have absolutely no idea that it is the ideology of giving users
their freedom back that caused the OS to exist in the first place.
Cheers,
Rob
--
Rob S. Wolfram <aze...@hamal.xs4all.nl> OpenPGP key 0xD61A655D
The best way to accelerate a computer running Windows is at
9.8 meters per square second
Richard Revis
> The Unix
> paradigm
I thought UNIX was meant to be (at least where possible) PHB free?
--
'Microsoft is not the answer. Microsoft is the question. NO is the answer.'
12:51pm up 13 days, 15:45, 2 users, load average: 0.01, 0.02, 0.00
E-mail address munged to prevent spam.
Linus Torvalds
Rob S. Wolfram <aze...@hamal.xs4all.nl> wrote:
>[snip]
>> Quite frankly, I don't _want_ people using Linux for ideological reasons. I
>> think ideology sucks. This world would be a much better place if people had
>> less ideology, and a whole lot more "I do this because it's FUN and because
>> others might find it useful, not because I got religion."
>
>/me thinks that if that is his opinion, he chose a wrong license for
>spreading his kernel. He should have chosen the BSD license instead
>(which actually did exist back in '91).
No. I think the BSD license is a fine license, but I don't want any of
my code under it.
The point about the GPL (for me) is the continual improvements it allows
by everybody. The BSD license tends to inherently fracture the project
at some point: if it becomes interesting enough commercially, somebody
_will_ take the approach of not feeding back his changes, and the
project as a whole suffers.
Yes, the BSD apologists talk about how the original code is still
available, but that's a complete red herring. It doesn't matter if the
original code is available, if somebody else makes something that is
"better enough" to steal away a lot of resources from the project. I do
not want _my_ code to end up like that.
This is not just theory, you can see it in the "dilution" of the BSD
camps themselves, but you can also see it clearly in projects like Wine.
In Wine it got so bad that the project leaders themselves decided to opt
for the LGPL exactly because of these issues. And that kind of license
change is _painful_. They didn't do it lightly.
Finally, I don't see the point of the BSD license. If I really don't
care about what happens to my code, I'd just make it truly public
domain.
But licenses are like underwear: you get _really_ upset if somebody
tries to change them for you. To each his own.
Linus
Paul Cooke
what the heck do you say after that...
Thank you Linus... for putting it so clearly.
--
Paul Cooke
Registered Linux user 273897 Machine registration number 156819
Linux Counter: Home Page = http://counter.li.org/
7:40pm up 2 days, 10:34, 2 users, load average: 0.22, 0.15, 0.16
Erik Funkenbusch
Linus Torvalds <torv...@penguin.transmeta.com> wrote:
> No. I think the BSD license is a fine license, but I don't want any of
> my code under it.
>
> The point about the GPL (for me) is the continual improvements it
> allows by everybody. The BSD license tends to inherently fracture
> the project at some point: if it becomes interesting enough
> commercially, somebody _will_ take the approach of not feeding back
> his changes, and the project as a whole suffers.
>
> Yes, the BSD apologists talk about how the original code is still
> available, but that's a complete red herring. It doesn't matter if the
> original code is available, if somebody else makes something that is
> "better enough" to steal away a lot of resources from the project. I
> do not want _my_ code to end up like that.
>
> This is not just theory, you can see it in the "dilution" of the BSD
> camps themselves, but you can also see it clearly in projects like
> Wine. In Wine it got so bad that the project leaders themselves
> decided to opt for the LGPL exactly because of these issues. And
> that kind of license change is _painful_. They didn't do it lightly.
So what you're saying is that, at least for you, the GPL is about control.
As in, controlling your code so that others can't go somewhere else?
That doesn't make sense to me, since the GPL doesn't prevent forking at all.
It might give you more opportunity to get access to someone elses changes,
but it certainly doesn't guarantee it.
> Finally, I don't see the point of the BSD license. If I really don't
> care about what happens to my code, I'd just make it truly public
> domain.
Making it public domain would also allow others to patent it (Yeah, I know..
prior art and all that, but if you don't "own" the code, you have very
little chance of standing up to it in a patent suit) and then turn around
and force you to license your own code.
> But licenses are like underwear: you get _really_ upset if somebody
> tries to change them for you. To each his own.
I guess that depends on who the "someone" is ;)
Colin Day
>
> Note: This appears to be a real message from Linus based on headers
>
> Linus Torvalds <torv...@penguin.transmeta.com> wrote:
> > No. I think the BSD license is a fine license, but I don't want any of
> > my code under it.
> >
> > The point about the GPL (for me) is the continual improvements it
> > allows by everybody. The BSD license tends to inherently fracture
> > the project at some point: if it becomes interesting enough
> > commercially, somebody _will_ take the approach of not feeding back
> > his changes, and the project as a whole suffers.
> >
> > Yes, the BSD apologists talk about how the original code is still
> > available, but that's a complete red herring. It doesn't matter if the
> > original code is available, if somebody else makes something that is
> > "better enough" to steal away a lot of resources from the project. I
> > do not want _my_ code to end up like that.
> >
> > This is not just theory, you can see it in the "dilution" of the BSD
> > camps themselves, but you can also see it clearly in projects like
> > Wine. In Wine it got so bad that the project leaders themselves
> > decided to opt for the LGPL exactly because of these issues. And
> > that kind of license change is _painful_. They didn't do it lightly.
>
> So what you're saying is that, at least for you, the GPL is about control.
> As in, controlling your code so that others can't go somewhere else?
Quite the contrary, they can go somewhere else. They just can't prevent
you from tagging along.
>
> That doesn't make sense to me, since the GPL doesn't prevent forking at all.
> It might give you more opportunity to get access to someone elses changes,
> but it certainly doesn't guarantee it.
>
> > Finally, I don't see the point of the BSD license. If I really don't
> > care about what happens to my code, I'd just make it truly public
> > domain.
>
> Making it public domain would also allow others to patent it (Yeah, I know..
> prior art and all that, but if you don't "own" the code, you have very
> little chance of standing up to it in a patent suit) and then turn around
> and force you to license your own code.
>
> > But licenses are like underwear: you get _really_ upset if somebody
> > tries to change them for you. To each his own.
>
> I guess that depends on who the "someone" is ;)
True. But I suspect that Elizabeth Hurley wouldn't want to change my
underwear :-(.
Colin Day
Linonut
> Finally, I don't see the point of the BSD license. If I really don't
> care about what happens to my code, I'd just make it truly public
> domain.
>
> But licenses are like underwear: you get _really_ upset if somebody
> tries to change them for you. To each his own.
>
> Linus
Howdy, stranger! You must be new 'round here!
Chris
--
We have DIFFERENT amounts of HAIR --
Kenneth Downs
Great, with your heavy programming experience, we're all waiting for the
product announcements.
Rob S. Wolfram
Linus, I couldn't agree more with these statements, and I understand
that your choice for the GPL was pragmatic, rather that ideological.
However, I disagree with your LKML post that the world would be a better
place if people had less ideology. I think that we would not have a GPL
if RMS was not so stubbornly ideological.
I also disagree with the BSD apologists' reasoning that the original
code is still available. A nice example is the fact that zlib is used in
all sorts of closed-source applications like Microsoft's IE. Now it is
not poven that the double free() bug is exploitable in anything but
glibc but still, even though zlib is free code, IE users are not able to
patch the bug themselves.
> Finally, I don't see the point of the BSD license. If I really don't
> care about what happens to my code, I'd just make it truly public
> domain.
>
> But licenses are like underwear: you get _really_ upset if somebody
> tries to change them for you. To each his own.
I know you care, but the point was directed to the fact that you dislike
that some users care about the ideology behind the GPL. The BSD license
seems to be one where ideology doesn't count, only practical
applicability. Thus, my remark about the BSDL.
Sidenote: I really appreciate that you still visit COLA now and again
and "meet the people on the street", even though the signal to noise
ratio has dropped to an all time low...
Cheers,
Rob
--
Rob S. Wolfram <aze...@hamal.xs4all.nl> OpenPGP key 0xD61A655D
If the Internet was meant to be like TV, you would have been
given a remote control.
-- Rop Gongrijp
Tom Gough
Symbiosis. Well nearly.
Tom Gough
> Tim Smith wrote:
>
>> Some people are upset that Linus uses BitKeeper, which is non-free, for
>> source code control. Here's what Linus had to say on the kernel list:
>>
>> a commercial product can do something better than a free one, how
>> about _doing_ something about it?
>>
>> reasons. I think ideology sucks. This world would be a much better
>> place if people had less ideology, and a whole lot more "I do this
>> because it's FUN and because others might find it useful, not because
>> I got religion."
>>
>> for kernel maintenance? Yes. But since no such tool exists, and since
>> I'm personally not very interested in writing one, _and_ since I don't
>> have any hangups about using the right tool for the job, I use
>> BitKeeper.
>>
>> they start to get carried away.
>>
>
> Your other post asks the question "Is Linux Free?" According to RMS, no.
>
> Then you tell us to keep in mind Linus's attitude, which pretends that
> politics and ideology do not exist. Seems to me that without the
> "ideology" of the FSF, Linus would have no OS for his kernel.
>
> ...and we would have no NG in which to debate these things.
>
> Something else to keep in mind.
>
is free. Linus tends to be more of a pragmatic person. He does what works
and that's that. I've never seen any evidence he believes ideology or
politics do not exist, simply that they are less important that pragmatism.
Tom Gough
You just made my point better than me. Bollocks
--
http://www.guild.bham.ac.uk/chess-club
Tom Gough
> Erik Funkenbusch wrote:
>
>>> Tim Smith wrote:
>>> Your other post asks the question "Is Linux Free?" According to RMS, no.
>>>
>>> Then you tell us to keep in mind Linus's attitude, which pretends that
>>> politics and ideology do not exist. Seems to me that without the
>>> "ideology" of the FSF, Linus would have no OS for his kernel.
>>>
>>> ...and we would have no NG in which to debate these things.
>>>
>>> Something else to keep in mind.
>>
>> Are you honestly trying to suggest that Linus (or any other Free OS)
>> couldn't come up with a few utilities like "tail" or "more" himself (or
>> his group of friends)?
>>
>
> no. i'm saying he could not have come up with the entire GNU toolset,
> another thing entirely.
>
Why not? He would have to have worked differently, but given what Linux
has achieved, there is no reason why versions of the whole GNU toolset
could not have been produced independantly of GNU.
>> Sure, Linux might not have been so comprehensive (at least as quickly as
>> it did) without the FSF, but IMO the only thing that Linux actually
>> *needed* was gcc (which is from the FSF, but even RMS doesn't claim that
>> anything compiled with gcc should be called GNU).
>
> thank you for your opinion.
>
Kenneth Downs
Say Linus, I'm only a few months into this open-source thing, so the idea
of the author dropping by to chat still strikes me as completely unexpected.
So as long as you're here, please accept my thanks for your cool kernel.
Linux and tools that make it a complete system have reinvigorated my love
of the craft, which had been quietly dying inside of me for years.
I still wonder at the "sound byte" quotes you offer on ideology, they give
me the distinct impression that you think there is no place for ideology.
But your response clearly indicates otherwise. Perhaps sometime you could
expand on what you mean by "ideology sucks." Methinks you are holding on
out on us :)
Ed Cogburn
>
> Linus, I couldn't agree more with these statements, and I understand
> that your choice for the GPL was pragmatic, rather that ideological.
> However, I disagree with your LKML post that the world would be a better
> place if people had less ideology. I think that we would not have a GPL
> if RMS was not so stubbornly ideological.
I agree Linus is largely pragmatic rather than ideological, but those he
works with are very different. Has anyone given thought to the
"tainting the kernel" code that is in the modutils programs and the
kernel? Now, this, IMHO, is exactly the kind of ideologically driven
silliness that has no value or constructive reason for existing. Its
just there to annoy people who have no way to avoid non-GPL code because
they have hardware with closed source drivers. Linus never should have
allowed this into the kernel, but he did, so how pragmatic is he really?
--
It is dangerous to be right when the government is wrong. -- Voltaire
Chronos Tachyon
The reasoning behind tainting's acceptance, IIRC, was that people were
frequently reporting kernel oopses and panics that were caused by buggy
proprietary drivers, and that making the kernel "tainted" after loading a
no-source driver would reduce the number of spurious bug reports sent to
the kernel developers. At least, that's what I remember from following
Kernel Traffic at the time it was implemented.
--
Chronos Tachyon
http://chronos.dyndns.org/ -- WWED?
Guardian of Eristic Paraphernalia
Gatekeeper of the Region of Thud
3:02am up 1:15, 1 user, load average: 0.00, 0.02, 0.00
Colin Day
Until he storms into my apartment and uninstalls my "tainted" Nvidia
drivers,
I will assume that he did not introduce "tainting" for ideological
purposes.
Colin Day
Ed Cogburn
> On Sat 25 May 2002 12:49, Ed Cogburn wrote:
>
> The reasoning behind tainting's acceptance, IIRC, was that people were
> frequently reporting kernel oopses and panics that were caused by buggy
> proprietary drivers, and that making the kernel "tainted" after loading a
> no-source driver would reduce the number of spurious bug reports sent to
> the kernel developers. At least, that's what I remember from following
> Kernel Traffic at the time it was implemented.
I read some of the arguments, and I think I see some ideology playing a
large role here. Their claim is its supposed to reduce "bad" support
requests, yet what does it say when I boot my machine which includes
non-GPL NVIDIA drivers for my GeForce2 based video card:
"Loading modules: NVdriver warning: loading [module-file-path] will
taint the kernel: non-GPL license - NVIDIA"
I find a couple of things about this to be bizarre. First, how do they
prevent invalid support requests with this language? Driver support is
never mentioned. Second, why do they make it sound like the error is
being reported by the non-GPL driver? The "NVdriver warning:" for
example. Isn't this going to confuse the heck out of people? Third,
why don't they just call the offending driver "unsupported"? Instead
they use the word "taint" which has a certain negative connotation, like
something has been "poisoned" or "corrupted", and they mention its
"non-GPL license" but they never explain *why* the user is getting the
error/warning message. All they do is specifically single out "non-GPL
licenses" for abuse. How is a newbie, a virgin to the ideology wars
surrounding Open Source, going to react to this warning message? He's
probably going to think something is wrong with the kernel!
I agree with this guy: http://www.smcc.demon.nl/webcam/tainting.html
Bob Hauck
> "Loading modules: NVdriver warning: loading [module-file-path] will
> taint the kernel: non-GPL license - NVIDIA"
>
> I find a couple of things about this to be bizarre. First, how do they
> prevent invalid support requests with this language?
I don't think it is so much preventing the requests, but ignoring them
if they involve a tainted kernel.
> All they do is specifically single out "non-GPL licenses" for abuse.
They're singling out licenses that are "less free" than GPL. Sure,
there is likely some ideology here, but there are also valid practical
reasons to not encourage closed drivers:
o They go unsupported if the company goes under.
o They go unsupported if the company decides to drop the product.
o They are unlikely to support non-x86 architectures.
o Supporting pure binary drivers would require freezing internal
structures that the developers don't want to freeze.
Perhaps the word "taint" is a bit loaded, but I think there are valid
reasons to warn people about why they ought to avoid closed drivers if
they can.
--
-| Bob Hauck
-| To Whom You Are Speaking
-| http://www.haucks.org/
Ed Cogburn
> On Sun, 26 May 2002 01:41:46 -0400, Ed Cogburn <ecog...@xtn.net> wrote:
>
>
>>"Loading modules: NVdriver warning: loading [module-file-path] will
>>taint the kernel: non-GPL license - NVIDIA"
>>
>>I find a couple of things about this to be bizarre. First, how do they
>>prevent invalid support requests with this language?
>>
>
> I don't think it is so much preventing the requests, but ignoring them
> if they involve a tainted kernel.
But they claim they did this to REDUCE support requests from those with
hardware that has closed drivers.
>>All they do is specifically single out "non-GPL licenses" for abuse.
>>
>
> They're singling out licenses that are "less free" than GPL. Sure,
> there is likely some ideology here, but there are also valid practical
> reasons to not encourage closed drivers:
>
> o They go unsupported if the company goes under.
> o They go unsupported if the company decides to drop the product.
> o They are unlikely to support non-x86 architectures.
> o Supporting pure binary drivers would require freezing internal
> structures that the developers don't want to freeze.
You used the word "support" and "unsupported" in every one of those
points. I agree with your points, but why is it that their "tainted"
messages never use that word?
> Perhaps the word "taint" is a bit loaded, but I think there are valid
> reasons to warn people about why they ought to avoid closed drivers if
> they can.
There ARE valid reasons for wanting to point out to people that the
Linux people you ask for help from can't help you if you are using
closed source drivers, but these "tainted" messages attacking "non-GPL"
licensed software DOES NOT DO THAT. Its Free Software extremism run
amok in the kernel.
Bob Hauck
> Bob Hauck wrote:
> There ARE valid reasons for wanting to point out to people that the
> Linux people you ask for help from can't help you if you are using
> closed source drivers, but these "tainted" messages attacking "non-GPL"
> licensed software DOES NOT DO THAT.
Perhaps you could write a patch to change all instances of "taint" to
"unsupported" then. The word itself seems to be your big beef rather
than the purpose for putting it there.
> Its Free Software extremism run amok in the kernel.
Maybe that's not such a bad thing. Closed source software is simply a
pain in my ass. Keeping track of licenses, dongles, license servers,
node-locked software, etc, it is all overhead that adds no value to the
product at all. Then there's all the stuff I've paid for over the years
and the company went out of business or got bought out or just doesn't
care any more and so now I have to spend _more_ money and _more_ time to
migrate to something else.
Maybe the world needs more "free software extremeism".
Ed Cogburn
> On Mon, 27 May 2002 03:39:53 -0400, Ed Cogburn <ecog...@xtn.net> wrote:
>
>>Bob Hauck wrote:
>>
>
>>There ARE valid reasons for wanting to point out to people that the
>>Linux people you ask for help from can't help you if you are using
>>closed source drivers, but these "tainted" messages attacking "non-GPL"
>>licensed software DOES NOT DO THAT.
>>
>
> Perhaps you could write a patch to change all instances of "taint" to
> "unsupported" then. The word itself seems to be your big beef rather
> than the purpose for putting it there.
More than the word needs to be changed, the entire messages need to be
replaced. Something like this:
"This hardware driver is not Open Source. The Open Source community and
Linux developers can not support proprietary Linux hardware drivers,
therefore only the vender who provided you this driver can help you if
you have problems."
Besides, I've got a better idea, download the modutils package and
commented out the messages, then at least they won't clutter your syslog
anymore.
>>Its Free Software extremism run amok in the kernel.
>>
>
> Maybe that's not such a bad thing. Closed source software is simply a
> pain in my ass. Keeping track of licenses, dongles, license servers,
> node-locked software, etc, it is all overhead that adds no value to the
> product at all. Then there's all the stuff I've paid for over the years
> and the company went out of business or got bought out or just doesn't
> care any more and so now I have to spend _more_ money and _more_ time to
> migrate to something else.
>
> Maybe the world needs more "free software extremeism".
Yes, the world would be a lovely place if all software were free,
unfortunately the world is far from a lovely place, and will remain so
for a very long time. I don't like closed-source software any more than
you do, but when it comes to closed-source *hardware* *drivers* there is
a reasonable excuse: the hardware people don't want to give their
competitors an advantage by seeing the details of how their hardware is
driven. As a moderate in the ideology wars, I can understand that. I
don't like it, but I can understand it, and few if anyone out there
anytime soon is going to buy hardware based solely on whether its got
open source drivers or not. 99.9% of consumers out there will buy the
best hardware available at the time, regardless of drivers. Open Source
should at least acknowledge that reality and not make it worse for the
people who buy hardware based on features and price, not ideology. I
don't mind the basic ideal behind open source software, in essence I
agree with it, but when free software starts deliberately making itself
difficult/annoying to use with closed-source, just because its
closed-source, then you've gone to far. There is absolutely no
constructive reason for the tone or wording of those warning messages.
Bob Hauck
> Bob Hauck wrote:
>> Perhaps you could write a patch to change all instances of "taint" to
>> "unsupported" then. The word itself seems to be your big beef rather
>> than the purpose for putting it there.
> More than the word needs to be changed, the entire messages need to be
> replaced. Something like this:
>
> "This hardware driver is not Open Source. The Open Source community and
> Linux developers can not support proprietary Linux hardware drivers,
> therefore only the vender who provided you this driver can help you if
> you have problems."
Well, there you go then. Maybe you can talk Linus or Red Hat or
whomever into it.
> Besides, I've got a better idea, download the modutils package and
> commented out the messages, then at least they won't clutter your syslog
> anymore.
Yes, it sure is nice to be able to make things work the way _you_ want
them to work. Yet another reason to dislike closed software.
>> Maybe the world needs more "free software extremeism".
> Yes, the world would be a lovely place if all software were free,
And we'll never get there if we keep compromising when there are other
choices.
> but when it comes to closed-source *hardware* *drivers* there is
> a reasonable excuse: the hardware people don't want to give their
> competitors an advantage by seeing the details of how their hardware is
> driven.
But, Ed, that is *always* the excuse. Drivers or regular software,
makes no difference. The thinking is *always* that "if we let our
competitors see this then we won't have an advantage". The thing that
gets forgot in the marketing meetings is the _customer_ and what his
needs are.
In the case of closed source drivers, it seems clear to me that if the
method of driving the thing has such a big impact on performance, then
they aren't selling _just_ hardware. They are selling "proprietary IP"
or some such thing. Pretending that they are selling hardware is blowing
smoke.
Note that I don't include loadable firmware in this, as Stallman does,
because that is usually a "burn file" for some kind of FPGA or on-board
micro. Having that kind of binary in the driver does not prevent me
from porting it to a new platform, updating it for a new kernel, etc.
In this case I view the binary as a legitmate part of the hardware.
> don't like it, but I can understand it, and few if anyone out there
> anytime soon is going to buy hardware based solely on whether its got
> open source drivers or not.
If they want to ever escape from the lock-in that vendors are trying to
perpetuate then they'll have to change their ways. Maybe they won't,
but it doesn't hurt to point out the tradeoffs.
Ed Cogburn
> On Mon, 27 May 2002 11:49:12 -0400, Ed Cogburn <ecog...@xtn.net> wrote:
>
>>Bob Hauck wrote:
>>
>
>>>Perhaps you could write a patch to change all instances of "taint" to
>>>"unsupported" then. The word itself seems to be your big beef rather
>>>than the purpose for putting it there.
>>>
>
>>More than the word needs to be changed, the entire messages need to be
>>replaced. Something like this:
>>
>>"This hardware driver is not Open Source. The Open Source community and
>>Linux developers can not support proprietary Linux hardware drivers,
>>therefore only the vender who provided you this driver can help you if
>>you have problems."
>>
>
> Well, there you go then. Maybe you can talk Linus or Red Hat or
> whomever into it.
No way, they like the "tainted" ideology too much.
>>Besides, I've got a better idea, download the modutils package and
>>commented out the messages, then at least they won't clutter your syslog
>>anymore.
>>
>
> Yes, it sure is nice to be able to make things work the way _you_ want
> them to work. Yet another reason to dislike closed software.
I agree, always did. I'm making a distinction between closed software
in general and closed-source hardware drivers.
>>>Maybe the world needs more "free software extremeism".
>>>
>
>>Yes, the world would be a lovely place if all software were free,
>>
>
> And we'll never get there if we keep compromising when there are other
> choices.
The world is full of compromises. They are necessary on a daily basis
so life can go on. We couldn't get anywhere without them. Only
extremists and religious fanatics believe a good world is possible with
no compromising.
>>but when it comes to closed-source *hardware* *drivers* there is
>>a reasonable excuse: the hardware people don't want to give their
>>competitors an advantage by seeing the details of how their hardware is
>>driven.
>>
>
> But, Ed, that is *always* the excuse. Drivers or regular software,
> makes no difference. The thinking is *always* that "if we let our
> competitors see this then we won't have an advantage". The thing that
> gets forgot in the marketing meetings is the _customer_ and what his
> needs are.
I disagree, there is a difference between the code of a word processor
and the code of a video card hardware driver. Driver code speaks
*volumes* about the hardware its driving. It can be the difference
between success or failure of the hardware vendor.
> In the case of closed source drivers, it seems clear to me that if the
> method of driving the thing has such a big impact on performance, then
> they aren't selling _just_ hardware. They are selling "proprietary IP"
> or some such thing. Pretending that they are selling hardware is blowing
> smoke.
Pretending? Strange, what is the card in my computer then? An illusion?
That hardware makers also provide software drivers is not a revelation.
Of course they are selling more than just hardware. You're just
falling back on the ideal that all software be Free regardless of its
purpose. But I'm telling you there is a distinct difference between
general purpose software and hardware drivers. We may not like it, but
this reality isn't going to change any time soon.
>>don't like it, but I can understand it, and few if anyone out there
>>anytime soon is going to buy hardware based solely on whether its got
>>open source drivers or not.
>>
>
> If they want to ever escape from the lock-in that vendors are trying to
> perpetuate then they'll have to change their ways. Maybe they won't,
> but it doesn't hurt to point out the tradeoffs.
"Lock in" has absolutely no meaning when speaking of hardware. If you
buy the hardware, then of course you're locked in. You're locked in for
as long as you use that hardware. If using that hardware also means
using a closed-source hardware driver then again you have no choice.
With general purpose software you can get yourself a different word
processor if you want, but by its nature, hardware is a natural, and
unavoidable, "lock in". Unlike general purpose software there is a
valid reasoning, a necessity, for not wanting the details of your
hardware driver to be public knowledge. So, if you want to keep arguing
in terms of general purpose software, I won't argue with you as I agree,
but I still maintain there is a difference when it comes to
closed-source hardware drivers.
Bob Hauck
> Bob Hauck wrote:
> The world is full of compromises. They are necessary on a daily basis
> so life can go on. We couldn't get anywhere without them. Only
> extremists and religious fanatics believe a good world is possible with
> no compromising.
I hope you aren't saying that it isn't legitimate to try to change the
way things are.
>> But, Ed, that is *always* the excuse. Drivers or regular software,
>> makes no difference. The thinking is *always* that "if we let our
>> competitors see this then we won't have an advantage".
> I disagree, there is a difference between the code of a word processor
> and the code of a video card hardware driver.
Nope. The word processor can have equally unique algorithms and
implementations as the driver. Vendors always want to protect what they
think is their secret sauce.
>> They are selling "proprietary IP" or some such thing. Pretending that
>> they are selling hardware is blowing smoke.
> Pretending? Strange, what is the card in my computer then? An illusion?
They are pretending to sell just hardware, which will normally work on a
variety of software platforms and which typically has a long life. They
are glossing over the secret IP in the software that will have a much
shorter lifespan.
Having secret IP is fine, but I reserve the right not to buy something
based on that and I refuse to pretend that it is somehow different if
there is hardware involved as compared to if there is not. I use closed
drivers sometimes, but I'm not blind to the implications.
> That hardware makers also provide software drivers is not a revelation.
> Of course they are selling more than just hardware.
What do you mean "of course"? Having large hunks of functionality in
the driver is a relatively recent thing. Once upon a time, hardware
vendors published schematics.
> But I'm telling you there is a distinct difference between
> general purpose software and hardware drivers. We may not like it, but
> this reality isn't going to change any time soon.
And I'm telling you that there is no such distinction. These vendors
are selling closed-source software that provides part of their
functionality. I fail to see how that's different from any other closed
source software.
I agree with what you say about the way things are presented now, that's
something that I'd like changed.
> "Lock in" has absolutely no meaning when speaking of hardware.
Sure it does, it just means something slightly different. Closed-source
drivers prevent you from using the hardware on the platform of your
choice, or even on a newer version of a once-supported platform if the
vendor doesn't feel like updating old drivers, or is out of business, or
whatever. You're locked-in to what the vendor wants to or can support,
even if the hardware will theoretically work.
This is a very similar problem to what you have with closed software,
don't you think?
> If you buy the hardware, then of course you're locked in. You're
> locked in for as long as you use that hardware.
Which may not be as long as _you'd_ like it to be. If the hardware
vendor decides not to support Linux 2.6 for that old card, then you
can't use it and have to buy another one.
This is exactly the same as any other old and unsupported software, and
if it is closed source you can't do anything to fix it.
> Unlike general purpose software there is a valid reasoning, a
> necessity, for not wanting the details of your hardware driver to be
> public knowledge.
As someone who sometimes writes hardware drivers, I'm afraid I still
don't see the difference. Something like Mathematica might contain
intellectual property that you don't want to reveal. Many ERP systems
are another example of that. How's that different? In either case it is
simply a matter of not wanting to reveal the secret info for whatever
reason. In either case basing part of my purchase decision on the
freeness of the code is at least as legitimate as the vendor's decison
to keep secrets.
Linus' decision to disrespect closed-source drivers with the "taint"
mechanism is likewise legitimate.
Ed Cogburn
Bob Hauck wrote:
> On Tue, 28 May 2002 10:59:14 -0400, Ed Cogburn <ecog...@xtn.net> wrote:
>
>>Bob Hauck wrote:
>>
>
>>The world is full of compromises. They are necessary on a daily basis
>>so life can go on. We couldn't get anywhere without them. Only
>>extremists and religious fanatics believe a good world is possible with
>>no compromising.
>>
>
> I hope you aren't saying that it isn't legitimate to try to change the
> way things are.
No, its all right to *try* but until you succeed you still have to live
with the rest of the world. You can be stubborn and obstinate and go
out of your way not to cooperate, but that is the kind of extreme
behavior that is giving Open Source a bad name. Ideology is what I
expect from RMS and the FSF, I thought the OS movement was a little more
practical.
>>>But, Ed, that is *always* the excuse. Drivers or regular software,
>>>makes no difference. The thinking is *always* that "if we let our
>>>competitors see this then we won't have an advantage".
>>>
>
>>I disagree, there is a difference between the code of a word processor
>>and the code of a video card hardware driver.
>>
>
> Nope. The word processor can have equally unique algorithms and
> implementations as the driver. Vendors always want to protect what they
> think is their secret sauce.
Its not algorithms that we're talking about. Its not secret sauce
either. Again, you seem to be focusing on the software side of things,
and my point is because there is a very important hardware side to this
issue. The OS community should recognize the current reality, they
don't have to accept it in principle, but as a practical matter they
should recognize how things are really going to be for quite some time.
With hardware drivers there is the issue of how much internal secrets
of its *hardware* does a company want to be openly seen by its
competitors. I guess its just this: I have no problem with proprietary
hardware, and that's what the vendors want to protect, its NOT about the
software that you're focused on. You want to apply the Open Source
concept to hardware too, which just isn't going to happen anytime soon.
The Open Source concept works for software because of the nature of
software, but I don't think it applies well to hardware widgets. I like
the idea of having an Open Source alternative for a given function like
a word processor or an operating system, but I don't see anyone
demanding that Ford release all its blueprints and internal
specifications for every component in its cars as well as forfeit all
patents it may have on that technology. Widgets != Software, these 2
things are controlled by different rules and laws. You don't copyright,
or copyleft, widgets. Software patents are an abomination, but patents
on widgets are nearly universally accepted. Widgets can be owned with
normally straight forward rules and rights for both seller and buyer,
software can only be "rented", and the rules and rights of "seller" and
"buyer" are far from being coherent, it is locked up in a morass of
legal mumbo jumbo about "fair use" and EULAes. They are apples and oranges.
>>>They are selling "proprietary IP" or some such thing. Pretending that
>>>they are selling hardware is blowing smoke.
>>>
>
>>Pretending? Strange, what is the card in my computer then? An illusion?
>>
>
> They are pretending to sell just hardware, which will normally work on a
> variety of software platforms and which typically has a long life. They
> are glossing over the secret IP in the software that will have a much
> shorter lifespan.
>
> Having secret IP is fine, but I reserve the right not to buy something
> based on that and I refuse to pretend that it is somehow different if
> there is hardware involved as compared to if there is not. I use closed
> drivers sometimes, but I'm not blind to the implications.
I see, you are applying the OS concept to hardware as well as software.
I guess we'll just have to disagree on this then.
> Linus' decision to disrespect closed-source drivers with the "taint"
> mechanism is likewise legitimate.
I didn't say it wasn't legitimate, by definition Linus makes it
legitimate, only that its pointless and doesn't solve the "problem" they
claim it does, it creates a new one.
Bob Hauck
> Its not algorithms that we're talking about. Its not secret sauce
> either. Again, you seem to be focusing on the software side of things,
I am, because it is the issues of portability and support going forward
that are my primary concerns.
> With hardware drivers there is the issue of how much internal secrets
> of its *hardware* does a company want to be openly seen by its
> competitors.
I suppose you're talking about graphics cards where the driver compiles
drawing instructions into primitives that the hardware can execute.
Knowing how the driver works might help somebody figure out how the
card works, or more importantly, what operations the hardware actually
implements and which are done in the driver. Ah, we're back to secret
software sauce.
The real issue in the graphics business is that many of the vendors
think that they are in the IP business rather than the graphics card
business. As such, they have become paranoid and have started treating
mechanisms as Intellectual Property.
I've already said that I'm ok with things like FPGA bitstreams and
firmware for on-board CPU's being only available as binaries. The
driver merely loads those into the board, that works the same on all
platforms so it doesn't affect the issues I care about. A similar
argument can be made about non-operating factory test modes and the
like. I just want to know how to operate the device.
> I guess its just this: I have no problem with proprietary
> hardware, and that's what the vendors want to protect, its NOT about the
> software that you're focused on.
I think that if they are selling hardware then they should sell
hardware. That means documenting it so people can use it as with any
other physical device.
> The Open Source concept works for software because of the nature of
> software, but I don't think it applies well to hardware widgets. I like
> the idea of having an Open Source alternative for a given function like
> a word processor or an operating system, but I don't see anyone
> demanding that Ford release all its blueprints and internal
> specifications for every component in its cars as well as forfeit all
> patents it may have on that technology.
Um, I guess you haven't noticed how much Ford actually _does_ release.
They seem to have pretty complete specifications and tolerances
available for repair shops to use. And even if they didn't, you can buy
a car and make measurements for most of the things you'd like to know.
It is interesting that about the only thing you _can't_ get specs for is
the electronics. I think the reason for that is obvious...it is the
only place they can get away with it. For just about everything else
you can buy a car and take it apart and _see_ how it works.
> Software patents are an abomination, but patents on widgets are
> nearly universally accepted.
So why don't they patent stuff instead of keeping it secret? Because
they want it both ways, that's why. They are selling a physical device
but want to keep how it works a secret. That's pretty hard to do with
most non-electronic devices. Yet the companies making these products do
manage to stay in business.
> I see, you are applying the OS concept to hardware as well as software.
No, I don't think I am. But you're right, we aren't going to be seeing
eye-to-eye on this anytime soon.
Deven T. Corzine
> They're singling out licenses that are "less free" than GPL. Sure,
> there is likely some ideology here, but there are also valid practical
> reasons to not encourage closed drivers:
>
> o They go unsupported if the company goes under.
> o They go unsupported if the company decides to drop the product.
> o They are unlikely to support non-x86 architectures.
> o Supporting pure binary drivers would require freezing internal
> structures that the developers don't want to freeze.
These are also valid, practical reasons to support a project such as
Project UDI (Uniform Driver Interface) <http://project-udi.org/>,
which would allow device drivers to be written in an OS-neutral
fashion. It would keep device drivers (free and closed alike) from
being so closely tied to a particular kernel version.
Unfortunately, Richard Stallman jumped to the conclusion that UDI is
some sort of Intel ploy to take advantage of free software developers.
Actually, Project UDI spent years working on the design before Intel
ever joined their group, and it was developed by proprietary UNIX
vendors and hardware vendors for mutual advantage, not to take
advantage of free software. Moreover, Microsoft has no incentive to
implement UDI (and dilute their monopoly lock-in), especially since
all hardware vendors target their platform already. Meanwhile, UDI is
probably the best hope for the HURD and other alternative kernels to
match the level of driver support currently enjoyed by Linux without
massive porting efforts...
Meanwhile, Linux kernel developers have expressed reservations about
UDI for a different reason -- they seem unconvinced that UDI driver
performance can match native driver performance, and they're very
reluctant to sacrifice performance despite the obvious portability
advantages of UDI. Interestingly, existing UDI prototypes (on
proprietary UNIX systems) have been able to achieve comparable
performance to the system's native drivers, DESPITE being layered on
top of those very same drivers, and not being optimized for
performance! This bodes well for the possible performance of a more
"native" UDI implementation which could be optimized for performance,
and possibly improve performance over the more traditional "native"
driver types. If this is true, chances are that the Linux kernel
developers would support UDI, but they're likely to need proof.
As for closed drivers, we already deal with them, and the biggest
problem with them is the need to keep them up to date when the kernel
changes around them. Since UDI requires drivers to conform to a very
constrained API, and the OS has enormous flexibility in implementing
the "UDI environment" to support that API, it would allow most kernel
design changes (even pervasive ones) to be done in the UDI environment
alone, saving time replicating changes throughout the many device
drivers, and obviating the need to update closed drivers. Also, while
UDI might not require source code to be provided, UDI compatibility is
only guaranteed at the source level, which provides some incentive to
release source code. (Of course, where the platform allows it, UDI
will also strive for binary compatibility where possible, which is why
kernel upgrades on the same hardware shouldn't present a problem.)
UDI is an impressive achievement, and it has the potential to offer
enormous benefits to everyone but Microsoft. And even if closed,
proprietary software could benefit from UDI, free software could
benefit just as much, if not more. Should we cut off our nose to
spite our face? The proprietary vendors will continue to move forward
with UDI with or without the free software community. Why should we
continue to suffer from the Balkanization of device drivers?
The free software community should embrace UDI, not reject it!
mlw
>
> Bob Hauck <b...@this-is.invalid> wrote in message news:<slrnaf230...@wren.haucks.org>...
> > They're singling out licenses that are "less free" than GPL. Sure,
> > there is likely some ideology here, but there are also valid practical
> > reasons to not encourage closed drivers:
> >
> > o They go unsupported if the company goes under.
> > o They go unsupported if the company decides to drop the product.
> > o They are unlikely to support non-x86 architectures.
> > o Supporting pure binary drivers would require freezing internal
> > structures that the developers don't want to freeze.
>
> These are also valid, practical reasons to support a project such as
> Project UDI (Uniform Driver Interface) <http://project-udi.org/>,
> which would allow device drivers to be written in an OS-neutral
> fashion. It would keep device drivers (free and closed alike) from
> being so closely tied to a particular kernel version.
[snip]
This is a great idea. One of the things I liked about NT was that it had a
defined API. A standardized driver API for multiple operating systems is a good
idea, PROVIDED it is done mostly at the C preprocessor level.
I will look at it, hmmm, maybe I should write the NT version. :-)
Matthias Warkus
...and Deven T. Corzine <de...@ties.org> wrote:
> These are also valid, practical reasons to support a project such as
> Project UDI (Uniform Driver Interface) <http://project-udi.org/>,
> which would allow device drivers to be written in an OS-neutral
> fashion. It would keep device drivers (free and closed alike) from
> being so closely tied to a particular kernel version.
ISTR these UDI drivers would run setuid in user space and thus pose
enormous security problems?
Or am I confusing something here?
mawa
--
"Die Deutschen werden den Juden Auschwitz nie verzeihen."
-- Zvi Rex, zitiert von Henryk M. Broder