From: tml@tik.vtt.fi (Tor Lillqvist)
Newsgroups: comp.os.linux
Subject: Re: X386 and lame SVGA cards
Date: 7 May 92 14:06:30 GMT
Organization: Technical Research Centre of Finland, Laboratory for Information
Processing (VTT/TIK)
In-reply-to: ron_p@aruba.nysaes.cornell.edu's message of 7 May 92 03:12:21 GMT
In article < RON_P.92May6221221@aruba.nysaes.cornell.edu>
ron_p@aruba.nysaes.cornell.edu (Ron Pool) writes:
I very much like the fact that source is available for everything
I'm using in Linux right now, except for possibly X386 and gcc2.1
(as ported to Linux).
Excuse me, but am I missing something? Why is there lots of these
announcements "source for this and that for Linux is available"? If
Linux is POSIX compliant, most of the better free software (like GCC)
should be pretty straightforward to compile from the official sources.
Why is there a need for special Linux versions of the source?
Or is it simply the case that most Linux users are not used to compile
and install free software, and thus need a lot of handholding even for
trivial configuration things like do signal handlers need to be
reinstalled each time or not, or strchr vs. index.
--
Tor Lillqvist,
working, but not speaking, for the Technical Research Centre of Finland
Newsgroups: comp.os.linux
From: Paul Allen < paula@atc.boeing.com>
Subject: Sources (was: Re: X386 and lame SVGA cards)
Reply-To: paula@atc.boeing.com
Organization: The Internet
Date: Thu, 7 May 1992 20:32:15 GMT
tml@tik.vtt.fi (Tor Lillqvist) wrote:
|In article < RON_P.92May6221221@aruba.nysaes.cornell.edu>
ron_p@aruba.nysaes.cornell.edu (Ron Pool) writes:
|
| I very much like the fact that source is available for everything
| I'm using in Linux right now, except for possibly X386 and gcc2.1
| (as ported to Linux).
|
|Excuse me, but am I missing something? Why is there lots of these
|announcements "source for this and that for Linux is available"? If
|Linux is POSIX compliant, most of the better free software (like GCC)
|should be pretty straightforward to compile from the official sources.
|Why is there a need for special Linux versions of the source?
Linux appears to be considerably better than Minix when it comes to
ease of porting stuff, but there still seems to be some level of "port"
required for some software. I'm asuming that this will get better
as Linux matures and becomes more complete.
|Or is it simply the case that most Linux users are not used to compile
|and install free software, and thus need a lot of handholding even for
|trivial configuration things like do signal handlers need to be
|reinstalled each time or not, or strchr vs. index.
Many people who aren't used to compiling things are installing and
attempting to run Linux because it's available as a complete binary-
only distribution. Some of these folks will need handholding when
they start trying to customise things.
The last time I looked on tsx-11, the binaries to sources space ratio
was about 2:1. In many cases, Linux diffs for a package are included
with the binary rather than being separated out under the sources
hierarchy. (So you have to ftp a half meg of binary in order to get
the 20Kb of diffs!) When I commented on the apparent disregard for
sources (the "Binaries considered harmful" thread of a month ago), I
got a uniformly negative reaction from several of the major players
in the Linux arena. I found this surprising, since Unix software has
historically been distributed almost exclusively in source form.
If I were running things, Linux would be distributed as a minimal
binary bootstrap kit and sources (or diffs) for everything else.
It would be easy to assemble a complete source hierarchy without
having to deal with huge unwanted binaries. New users would need
to ftp the bootstrap kit in order to get started, but sources and
diffs would suffice after that. (Fortunately for me, I'm not running
things. I'd probably have half the net mad at me! :-))
Paul Allen
paul.allen@atc.boeing.com
Newsgroups: comp.os.linux
From: tytso@ATHENA.MIT.EDU (Theodore Ts'o)
Subject: Re: Sources (was: Re: X386 and lame SVGA cards)
Reply-To: tytso@athena.mit.edu
Organization: The Internet
Date: Thu, 7 May 1992 23:36:20 GMT
From: Paul Allen < paula@atc.boeing.com>
Reply-To: paula@atc.boeing.com
The last time I looked on tsx-11, the binaries to sources space ratio
was about 2:1. In many cases, Linux diffs for a package are included
with the binary rather than being separated out under the sources
hierarchy. (So you have to ftp a half meg of binary in order to get
the 20Kb of diffs!)
If this is true, then that is a bug. I generally try to break stuff
apart into a source tar file and a binary tar file, although there may
have been a few exceptions. (Hint for potential uploaders --- please
upload the binaries in a separate file from the sources/diffs!)
When I commented on the apparent disregard for
sources (the "Binaries considered harmful" thread of a month ago), I
got a uniformly negative reaction from several of the major players
in the Linux arena. I found this surprising, since Unix software has
historically been distributed almost exclusively in source form.
What I thought I heard as the general consensus (although I may be
biased since that's what *I* believe :-), wasn't that sources was bad
and that you should upload binaries. Rather, the consensus was that
people should upload sources or diffs whenever possible, but making
binaries available shouldn't be considered a bad thing.
I don't think it can be "harmful" to give people the option of either
grabbing sources or grabbing binaries. Even people who are comfortable
compiling distributions may not want to waste the time and disk space to
recompile the world. And if you attempt to force people who don't know
how to compile distributions, they will more likely just give up on
Linux rather than actually sitting down and learning the foibles of
make, learning how to edit appropriate config files, learning what
#ifdef POSIX, #ifdef SVR4. Eventually, this is something which all good
Unix hackers should know how to do. But that doesn't mean that you
start people out with that.
But if you don't like binaries, just pretend that the "bin" or
"binaries" subdirectory on your favorite FTP site doesn't exist. And if
you find yourself needing to grab 1/2 meg of binaries just to get a
diff, you should let your FTP administror know.
- Ted
Newsgroups: comp.os.linux
Path: sparky!uunet!europa.asd.contel.com!darwin.sura.net!blaze.cs.jhu.edu!bogstad
From: bogs...@blaze.cs.jhu.edu (Bill Bogstad)
Subject: Re: Sources (IMPORTANT to managers of LINUX ftp sites)
Message-ID: <1992May9.143843.13241@blaze.cs.jhu.edu>
Organization: Johns Hopkins Computer Science Department, Baltimore, MD
References: <1992May7.233620.29099@athena.mit.edu>
Date: Sat, 9 May 1992 14:38:43 GMT
Lines: 66
In article <1992May7.233620.29...@athena.mit.edu> ty...@athena.mit.edu writes:
>...
>What I thought I heard as the general consensus (although I may be
>biased since that's what *I* believe :-), wasn't that sources was bad
>and that you should upload binaries. Rather, the consensus was that
>people should upload sources or diffs whenever possible, but making
>binaries available shouldn't be considered a bad thing.
>...
I've never had any problems with binaries being distributed, but I
have complained in the past about sources for some packages either not being
clearly labeled as such and/or not being distributed at all. Well, a recent
posting in gnu.gcc.announce by Richard Stallman may be relevant to this
discussion. [BTW, I did not bring this to his attention. It isn't my
fault. :-)] To summarize: He considers that FTP sites that provide binaries
of programs covered under the GNU public license (GPL) without providing
FULL source code (not just diff files) "side by side" to be in violation of
the GPL.
I don't know if this is an official statement of policy for the Free
Software Foundation (FSF); but if anyone can make such a statement I would
think that he can. He specifically is commenting on GCC; but it would
appear that his comments would cover anything under the GPL for which he or
the FSF hold the copyright. I'm including the full text of his message
which appeared in gnu.gcc.announce below.
Bill Bogstad
bogs...@cs.jhu.edu
Newsgroups: gnu.gcc.announce
From: r...@gnu.ai.mit.edu (Richard Stallman)
Subject: Executables and anon ftp
Message-ID: <9205081831.AA03956@mole.gnu.ai.mit.edu>
Sender: gnuli...@ai.mit.edu
Organization: GNUs Not Usenet
Distribution: gnu
Date: Fri, 8 May 1992 10:31:22 GMT
Approved: info-...@prep.ai.mit.edu
Lines: 27
I've heard that there are ftp sites distributing executables of
versions of GCC, without full sources on line side by side with them.
This isn't allowed under the GPL, and for good reason: ensuring users
have easy access to the source code is one of the main aims of GNU.
This is not a matter of deliberate wrongdoing; the source code for
these versions is free and is circulating. (The people who wrote the
modified versions aren't in control of what ftp sites do.) But it is
still a significant matter because it can cause trouble for users.
Sometimes there are patch files available giving changes from an FSF
release, such as 2.1. But this is not sufficient. For example,
suppose the user copies the binaries and patches now, then tries to
use the patches in a few months when a problem arises. At that time
it will be hard to find a copy of GCC 2.1 sources to patch. Instead,
2.3 or 2.4 will be current--but the patches may not work in those
versions, due to the changes we will have made in GCC.
This is why the GPL says that the complete sources have to be made
available. In the case of distribution by anonymous FTP, users can
choose to copy just the executable, but they must have the option of
copying the source as well. So the source has to be on line alongside
the binary.
If you see an ftp site that has binaries available without complete
sources, please bring this to their attention.
Newsgroups: comp.os.linux
From: britt@leland.Stanford.EDU (Britt Park)
Subject: Re: Sources (IMPORTANT to managers of LINUX ftp sites)
Organization: DSG, Stanford University, CA 94305, USA
Date: Sat, 9 May 92 16:54:04 GMT
I think that a few comments about the GNU license are necessary. I do
not believe that the portion of the GNU license which states that modifications
to GNU software must fall under the GNU license has any legal basis. The
rights to the modifications to a program invest wholly with the author of those
modifications. As to the requirement that complete sources to a binary on an
archive site must be located on the same archive site; I don't believe this
has any basis either, at least not as the GNU license is currently phrased:
"You may copy and distibute the Program ... in object code or
executable form under the terms of Paragraphs 1 and 2 above provided that you
also do one of the following:
... Accompany it with the information you received as to where the
corresponding source code may be obtained."
Certainly I believe it's better for archive sites to make available
complete sources rather than just diffs. This is not, however, always
practicable. It costs a great deal in resources to maintain an archive site,
and GNU software, as we all know, weighs in on the hefty side. I'd prefer to
see archive sites with binaries only rather than archive sites which fold
because they don't have the space to hold all the sources.
Britt Park
britt@cb-iris.stanford.edu
Newsgroups: comp.os.linux
From: aclark@netcom.com (Al Clark)
Subject: Re: Sources (IMPORTANT to managers of LINUX ftp sites)
Date: Sat, 09 May 92 18:54:25 GMT
Organization: Netcom - Online Communication Services (408 241-9760 guest)
In article <1992May9.165404.20041@leland.Stanford.EDU>
britt@leland.Stanford.EDU (Britt Park) writes:
>
> I think that a few comments about the GNU license are necessary. I do
>not believe that the portion of the GNU license which states that modifications
>to GNU software must fall under the GNU license has any legal basis. The
>rights to the modifications to a program invest wholly with the author of those
>modifications. As to the requirement that complete sources to a binary on an
>archive site must be located on the same archive site; I don't believe this
>has any basis either, at least not as the GNU license is currently phrased:
>
> "You may copy and distibute the Program ... in object code or
>executable form under the terms of Paragraphs 1 and 2 above provided that you
>also do one of the following:
> ... Accompany it with the information you received as to where the
>corresponding source code may be obtained."
>
>
> Certainly I believe it's better for archive sites to make available
>complete sources rather than just diffs. This is not, however, always
>practicable. It costs a great deal in resources to maintain an archive site,
>and GNU software, as we all know, weighs in on the hefty side. I'd prefer to
>see archive sites with binaries only rather than archive sites which fold
>because they don't have the space to hold all the sources.
>
> Britt Park
> britt@cb-iris.stanford.edu
Not being a lawyer, I have no idea whether or not the above has any validity.
However, I would suggest that people be very careful about following it.
FSF and rms have a lot invested, and I suspect will be militant about
enforcing their copyright (copyleft). I have been told that one of the rights
protected by copyright is the right to modify, thus modifying a copyrighted
file without permission is in itself a copyright violation.
Be careful when you get into this legal stuff.
--
Al - aclark@netcom.com - My opinions are my own.
*** Practice random kindness and senseless acts of beauty! ***
From: tytso@ATHENA.MIT.EDU (Theodore Ts'o)
Subject: Re: Sources (IMPORTANT to managers of LINUX ftp sites)
Reply-To: tytso@athena.mit.edu
Date: Sun, 10 May 1992 16:14:57 GMT
From: britt@leland.Stanford.EDU (Britt Park)
Date: 9 May 92 16:54:04 GMT
As to the requirement that complete sources to a binary on an
archive site must be located on the same archive site; I don't
believe this as any basis either, at least not as the GNU license is
currently phrased:
"You may copy and distibute the Program ... in object code or
executable form under the terms of Paragraphs 1 and 2 above provided
that you also do one of the following:
... Accompany it with the information you received as to where the
corresponding source code may be obtained."
That's GPL version 1. GPL version 2 is not so clear; the same section
in GPL v. 2 states:
Accompany it with the information you received as to the offer
to distribute corresponding source code. (This alternative is
allowed only for noncommercial distribution and only if you
received the program in object code or executable form with such
an offer, in accord with Subsection b above.)
Grr.... I have sent mail to rms@gnu.ai.mit.edu, asking him to reconsider
the matter. In the meantime, I suggest that people who want to put
stuff under the GPL carefully read both the version 1 and version 2, and
consider which one you want to place your program under, if any.
Certainly I believe it's better for archive sites to make available
complete sources rather than just diffs. This is not, however, always
practicable. It costs a great deal in resources to maintain an
archive site, and GNU software, as we all know, weighs in on the hefty
side. I'd prefer to see archive sites with binaries only rather than
archive sites which fold because they don't have the space to hold
all the sources.
This is my concern also. I'm not sure what can be done, however.
(Except for stating that you wish your program to be placed under GPL
version 2.)
- Ted
Newsgroups: comp.os.linux
From: lee@uhunix.uhcc.Hawaii.Edu (Greg Lee)
Subject: Re: Sources (IMPORTANT to managers of LINUX ftp sites)
Nntp-Posting-Host: uhunix.uhcc.hawaii.edu
Organization: University of Hawaii
Date: Sun, 10 May 1992 20:29:16 GMT
In article <7!jkhpq.aclark@netcom.com> aclark@netcom.com (Al Clark) writes:
}In article <1992May9.165404.20041@leland.Stanford.EDU>
britt@leland.Stanford.EDU (Britt Park) writes:
}>...
}>modifications. As to the requirement that complete sources to a binary on an
}>archive site must be located on the same archive site; I don't believe this
}>has any basis either, ...
}...
}Not being a lawyer, I have no idea whether or not the above has any validity.
}However, I would suggest that people be very careful about following it.
}...
Since we are so heavily indebted to fsf, it would make sense to think first
about what is courteous and much later about what we could get away with.
--
Greg Lee < lee@uhunix.uhcc.hawaii.edu>
From: torvalds@klaava.Helsinki.FI (Linus Benedict Torvalds)
Newsgroups: comp.os.linux
Subject: Re: Sources (IMPORTANT to managers of LINUX ftp sites)
Date: 11 May 92 08:33:44 GMT
Organization: University of Helsinki
In article <1992May10.202916.26781@news.Hawaii.Edu> lee@uhunix.uhcc.Hawaii.Edu
(Greg Lee) writes:
>
>Since we are so heavily indebted to fsf, it would make sense to think first
>about what is courteous and much later about what we could get away with.
Indeed. I hope people try to follow the /intent/ of the GPL first, and
then start worrying about the legal implications. I don't feel too
strongly about the legalese: it's there just because it's required to
uphold that intent in a court of law, and wouldn't even be needed
otherwise. I'd suggest not reading the GPL searching for loopholes -
but on the other hand not being unnecessarily strict about it either.
Use a bit of common sense in it all (until somebody starts threatening
with legal action: but in that case you have probably not been following
the GPL even in intent).
There are probably quite a few binaries that should be removed from the
linux archives: especially the older ones that needed a bit more porting
than they need now with the better compiler/library, and that might not
have source (I think my original gcc-1.40 and bash-1.05 should probably
be removed: I've had to delete my sources to get X running.. Nobody
uses them any more anyway). It might even result in some general
cleanup of old binaries...
Linus
From: britt@leland.Stanford.EDU (Britt Park)
Subject: Re: Sources (IMPORTANT to managers of LINUX ftp sites)
Date: 12 May 92 17:41:08 GMT
I'm not quite sure that people have been reading to the end of my
original post. I'm not recommending wholesale theft of GNU software for
nefarious purposes. I'm recommending that people not allow GNU to
deprive them of what I believe to be their rights. The GNU license is
claiming the authority to dictate to programmers how thy wish to
distribute their work.
A modified program is a derived work and is covered by the
copyright of the work from which it is derived. The modifications are
probably not; at least no more so than commentary about a copyrighted
literary work is. The general rule appears to be that if parts of a
creative work are separable, then the separate parts are separate works.
Modifications to software are certainly physically separable from the
original. That's why we have diff. Of course the courts may decide
that this separability may not be relevant for software. They (the
courts) seem capable of deciding almost anything these days. Software
companies will certainly argue that modifications to programs are not
independant of the original program. In a very real sense they are not,
but then most software is not independant of the (copyrighted)
environment in which they run. Even ATT hasn't tried to claim copyright
over programs written for UNIX.
Legalities aside, FSF is, instead of promoting, hindering the
propagation of good free software by trying to require people to be not
just dutiful hackers who fix and port software, but distributors. What
I want is to make any modifications I make really freely available. I
don't have the room to keep multiple megs of source code lying around.
So, I just won't modify GNU software, at least, not until an attorney
tells me whether my legal theorizing is correct. Further, I don't think
its fair to bulletin boards or archive sites that they have to maintain
huge source files because thay want to make useful modifications
available.
Britt
britt@cb-iris.stanford.edu
P.S. As best as I can make out from the two versions of the GNU GPL, an
archive site is permitted to carry only binaries to a GNU work, as long
as they make it clearly known where the sources to the modified work can
be obtained. By "may be obtained" presumably is meant in the ways
outlined in paragraphs 1-4.
P.P.S. One hopeful note. The law generally takes a dim view of
contracts or other legally enforcible vehicles which try to extend their
terms past the original subject of the contract or those which try to
act in perpetuity. Entailment of property in wills has never been legal
in the U.S.
P.P.P.S. With my luck I'll probably be completely wrong about the legal
aspects. What I've said is all based on my understanding of the
Intellectual Property textbook I read, and my questioning of an attorney
friend. My understanding may be quite imperfect. I have a tendency to
confuse what's right with what's legal. They don't always correspond.
Keywords:
From: pgr@ecs.ox.ac.uk (Partially Grown Rhododendron)
Subject: Re: Sources (IMPORTANT to managers of LINUX ftp sites)
Date: 12 May 92 21:48:06 GMT
Reply-To: pgr@ecs.ox.ac.uk (Partially Grown Rhododendron)
This is a long reply to Britt Park's post...
In < 1992May12.174108.12892@leland.Stanford.EDU>, Britt Park
< britt@leland.Stanford.EDU> writes:
> I'm not quite sure that people have been reading to the end of my
> original post.
I think quite a lot of people have; that's probably why they are arguing
with you :-}
> I'm not recommending wholesale theft of GNU software for
> nefarious purposes. I'm recommending that people not allow GNU to
> deprive them of what I believe to be their rights. The GNU license is
> claiming the authority to dictate to programmers how thy wish to
> distribute their work.
I think the key phrase here is `what I believe' ...
The GPL can only affect you if you either try to distribute, use, or
modify code that has been put under it. Since the original author of a
piece of GPL'd code put it under this license, you only have as many
rights to the code as he (and it) grants. The GPL can `dictate'
*NOTHING* unless the author of the original work wanted it to; somebody
is not going round forcing programmers to put their *original* code
under the GPL. It only places constraints on people who then go on to
produce derived works; this is perfectly reasonable, since this is what
the original author wanted.
The GPL (version 2) states in its preamble:
| For example, if you distribute copies of such a program, whether
| gratis or for a fee, you must give the recipients all the rights that
| you have. You must make sure that they, too, receive or can get the
| source code. And you must show them these terms so they know their
| rights.
i.e., if you got a program from a site and compiled it, you must give
anybody who you give the binaries to the right to do the same thing.
You must tell them where the source code is. And you *must* supply them
with the COPYING notice.
Now, for the terms and conditions (precised where possible):
| 1. You may copy and distribute verbatim copies of the Program's
| source code as you receive it, in any medium, provided that you
| conspicuously and appropriately publish on each copy an appropriate
| copyright notice and disclaimer of warranty; keep intact all the
| notices that refer to this License and to the absence of any warranty;
| and give any other recipients of the Program a copy of this License
| along with the Program.
[bit about charging deleted]
Point: there is no prominent COPYING notice, or notice saying where
the source can be found, or who wrote it, in the gcc 2.1, libc, or
associated GNU code distribution for Linux that existed on tsx11 from
April 6th -- this is a requirement of the GPL. Hence, those tar files
are infringing the GPL, and thence the copyright (left). If I were the
author of GCC, I would feel rather peeved by that. Wouldn't you?
Perhaps that's what RMS noticed too...
| 2. You may modify your copy or copies of the Program or any portion
| of it, thus forming a work based on the Program, and copy and
| distribute such modifications or work under the terms of Section 1
| above, provided that you also meet all of these conditions:
[(a) prominent notices of date and time of changes made]
| b) You must cause any work that you distribute or publish, that in
| whole or in part contains or is derived from the Program or any
| part thereof, to be licensed as a whole at no charge to all third
| parties under the terms of this License.
[(c) interactive program must print no-warranty banner if a derived work]
And now the important bit:
*** Thus, it is not the intent of this section to claim rights or contest
*** your rights to work written entirely by you; rather, the intent is to
*** exercise the right to control the distribution of derivative or
*** collective works based on the Program.
> Modifications to software are certainly physically separable from the
> original. That's why we have diff. Of course the courts may decide
> that this separability may not be relevant for software.
[a bit more about modifications being independent of the original code]
The diff's may be owned by you; but the diff's without the original are
nothing -- when you can compile your own modifications without any of
the original code then you have written a new piece of software, which
you then own. Until then, you have to live under the previous
copyright. In order to make diff's useful they must be applied, and as
soon as they are applied they form a derived work. Which then falls
under the GPL. You don't like that? Fine, write everything yourself.
Then you won't have to worry about the GPL at all. Or work out a way of
compiling diff's without the original code...
> Even ATT hasn't tried to claim copyright over programs written for UNIX.
True, and the programs under the GPL do not try to claim copyright over
completely independently written code -- they only claim copyright over
code which uses code that was written by somebody else and licensed with
it.
> Legalities aside, FSF is, instead of promoting, hindering the
> propagation of good free software by trying to require people to be not
> just dutiful hackers who fix and port software, but distributors.
Point: It is not the FSF. This distinction is thrashed out repeatedly.
The GPL is trying to make sure that if Fred Blogg's BBS distributes
some of the GNU code in a compiled form, that anybody who picks it up
can also get the source. If Fred Blogg can not afford to put the
source on line, then Fred Blogg's must then think hard about whether he
should still be distributing the code -- if he does, he infringes the
license. Is this fair? Shoud he be forced to do it? It doesn't
really matter. The fact is that by distributing the code he has agreed
to abide by the license, and abide by it he *must*. Does this prevent
software distribution? In the short term, perhaps. However, if the
demand is sufficient, Fred Blogg's will surely find the space to put
both the binaries and the source -- and if he can't, somebody else
will.
> What I want is to make any modifications I make really freely available. I
> don't have the room to keep multiple megs of source code lying around.
You don't need to. Just put it on an FTP site where the original code
is sitting.
> Further, I don't think
> its fair to bulletin boards or archive sites that they have to maintain
> huge source files because thay want to make useful modifications
> available.
*I* don't think it's fair that some people are saying `Well, we like
this code, we use it, we like having the source code available, we like
being able to make modifications, but we don't like having to follow the
rules that are layed down about its distribution.' Nobody is forcing
you to use the code; if you don't use it then you don't need to worry
about the copyrights. If you do, then at least have the good manners to
do what is asked of you by it.
pihl
Disclaimer: I am not affiliated with the GNU project, or the FSF.
I do, however, agree with both their aims and their methods.
I am also not a lawyer -- I am, I hope, reading the GPL as
it was meant to be read, and not misinterpreting it.
If I am, apologies to all concerned :-}
--
|| Phil Richards pgr@prg.oxford.ac.uk, ...!uunet!mcsun!uknet!ox-prg!pgr ||
| \ \ \ Useless disclaimer: I can hardly speak intelligently for myself / / / |
| / / / so there is no chance that I will be speaking for anybody else. \ \ \ |
|| ``Intelligence is alien to me; I'm a computer scientist'' -- almost Spock ||
From: tytso@ATHENA.MIT.EDU (Theodore Ts'o)
Subject: Re: Sources (IMPORTANT to managers of LINUX ftp sites)
Reply-To: tytso@athena.mit.edu
Date: Wed, 13 May 1992 02:36:07 GMT
From: pgr@ecs.ox.ac.uk (Partially Grown Rhododendron)
Date: 12 May 92 21:48:06 GMT
Reply-To: pgr@ecs.ox.ac.uk (Partially Grown Rhododendron)
Point: there is no prominent COPYING notice, or notice saying where
the source can be found, or who wrote it, in the gcc 2.1, libc, or
associated GNU code distribution for Linux that existed on tsx11 from
April 6th -- this is a requirement of the GPL. Hence, those tar files
are infringing the GPL, and thence the copyright (left). If I were the
author of GCC, I would feel rather peeved by that. Wouldn't you?
Perhaps that's what RMS noticed too...
No, RMS was commenting in general; I don't think he specifically noticed
us.
The gcc 2.1 directory is mirrored off of banjo.concert.net, and I
believe those binaries are maintained by H.J. Lu
(hlu@yoda.eecs.wsu.edu).
However, if the
demand is sufficient, Fred Blogg's will surely find the space to put
both the binaries and the source -- and if he can't, somebody else
will.
Maybe. This certainly screws over smaller time operations, though,
where this may not be the case. I suppose if you're ideologically pure,
you would just say "tough" --- but remember, Linux started out that way.
In fact, some people might argue that it is still relatively small by
most standards --- the number of FTP sites in the U.S. (or in the
*world*, for that matter) that carry Linux is still pitifully small by
Internet standards. If one of the U.S. FTP sites were to fold due to
these requirements, it would probably mean a serious blow to Linux's
distribution.
In priciple, I agree with what the FSF is trying to do --- and for
programs where there have been serious changes made to do a "port", I
might even agree that it would be a good thing to upload the sources
along with binary. However, last I checked, many of the GNU packages
compile with little or no changes, except for maybe one or two #define's
or -D in a config file or in a Makefile. For those packages, the
argument that the binary might be from an older version than what's
currently available on prep.ai.mit.edu is much less compelling, since
compiling the new version will probably be just as trivial as compiling
the original version.
- Ted