RMS speaks on SCO issue at ZDNET
Steve
Jerry Nash
>http://techupdate.zdnet.com/techupdate/stories/main/0,14179,2914132,00.html
cp to groups.google.com ->
--
SCO's contract dispute with IBM has been accompanied by a smear
campaign against the whole GNU/Linux system. But SCO made an obvious
mistake when it erroneously quoted me as saying that "Linux is a copy
of Unix." Many readers immediately smelled a rat--not only because I
did not say that, and not only because the person who said it was
talking about published ideas (which are uncopyrightable) rather than
code, but because they know I would never compare Linux with Unix.
Unix is a complete operating system, but Linux is just part of one.
SCO is using the popular confusion between Linux and the GNU/Linux
system to magnify the fear that it can spread. GNU/Linux is the GNU
operating system running with Linux as the kernel. The kernel is the
part of the system that allocates the machine's resources to the other
programs you run. That part is Linux.
We developed GNU starting in 1984 as a campaign for freedom, whose aim
was to eliminate non-free software from our lives. GNU is free software,
meaning that users are free to run it, study it and change it (or pay
programmers to do this for them), redistribute it (gratis or for a fee),
and publish modified versions.
(See http://www.gnu.org/gnu/the-gnu-project.html.)
In 1991, GNU was mostly finished, lacking only a kernel. In 1992, Linus
Torvalds made his kernel, Linux, free software. Others combined GNU and
Linux to produce the first complete free operating system, GNU/Linux.
(See http://www.gnu.org/gnu/gnu-linux-faq.html.) GNU/Linux is also free
software, and SCO made use of this freedom by selling their version of it.
Today, GNU runs with various kernels including Linux, the GNU Hurd
(our kernel), and the NetBSD kernel. It is basically the same system,
whichever kernel you use.
Those who combined Linux with GNU didn't recognize that's what they were
doing, and they spoke of the combination as "Linux." The confusion spread;
many users and journalists call the whole system "Linux." Since they
also properly call the kernel "Linux," the result is even more confusion:
when a statement says "Linux," you can only guess what software it refers
to. SCO's irresponsible statements are shot through with ambiguous
references to "Linux." It is impossible to attribute any coherent meaning
to them overall, but they appear to accuse the entire GNU/Linux system of
being copied from Unix.
The name GNU stands for "GNU's Not Unix." The whole point of developing
the GNU system is that it is not Unix. Unix is and always was non-free
software, meaning that it denies its users the freedom to cooperate and
to control their computers. To use computers in freedom as a community,
we needed a free software operating system. We did not have the money to
buy and liberate an existing system, but we did have the skill to write
a new one. Writing GNU was a monumental job. We did it for our freedom,
and your freedom.
To copy Unix source code would not be ethically wrong, but it is illegal;
our work would fail to give users lawful freedom to cooperate if it were
not done lawfully. To make sure we would not copy Unix source code or
write anything similar, we told GNU contributors not even to look at Unix
source code while developing code for GNU. We also suggested design
approaches that differ from typical Unix design approaches, to ensure our
code would not resemble Unix code. We did our best to avoid ever copying
Unix code, despite our basic premise that to prohibit copying of software
is morally wrong.
Another SCO tool of obfuscation is the term "intellectual property." This
fashionable but foolish term carries an evident bias: that the right way
to treat works, ideas, and names is as a kind of property. Less evident
is the harm it does by inciting simplistic thinking: it lumps together
diverse laws--copyright law, patent law, trademark law and others--which
really have little in common. This leads people to suppose those laws are
one single issue, the "intellectual property issue," and think about
"it"--which means, to think at such a broad abstract level that the
specific social issues raised by these various laws are not even visible.
Any "opinion about intellectual property" is thus bound to be foolish.
(See http://www.gnu.org/philosophy/words-to-avoid.html.)
In the hands of a propagandist for increased copyright or patent powers,
the term is a way to prevent clear thinking. In the hands of someone
making threats, the term is a tool for obfuscation: "We claim we can
sue you over something, but we won't say what it is."
In an actual lawsuit, such ambiguity would make their case fail, or
even prevent it from getting off the ground. If, however, SCO's aim
is to shake the tree and see if any money falls down, or simply to
spread fear, they may regard vagueness and mystery as advantageous.
I cannot prognosticate about the SCO vs IBM lawsuit itself: I don't
know what was in their contract, I don't know what IBM did, and I am
not a lawyer. The Free Software Foundation's lawyer, Professor Moglen,
believes that SCO gave permission for the community's use of the code
that they distributed under the GNU GPL and other free software
licenses in their version of GNU/Linux.
However, I can address the broader issue of such situations. In a
community of over half a million developers, we can hardly expect
that there will never be plagiarism. But it is no disaster; we discard
that material and move on. If there is material in Linux that was
contributed without legal authorization, the Linux developers will
learn what it is and replace it. SCO cannot use its copyrights, or
its contracts with specific parties, to suppress the lawful
contributions of thousands of others. Linux itself is no longer
essential: the GNU system became popular in conjunction with Linux,
but today it also runs with two BSD kernels and the GNU kernel. Our
community cannot be defeated by this.
--
Copyright 2003 Richard Stallman. Verbatim copying and redistribution of
this entire article are permitted without royalty in any medium provided
this notice is preserved.
Jochem Huhmann
> On 26 Jun 2003 07:55:44 -0700, steves...@yahoo.com (Steve) wrote:
>
>>http://techupdate.zdnet.com/techupdate/stories/main/0,14179,2914132,00.html
>
> RMS is STILL pissed that he did all the work and Linus became a star
> and got all the credit for Linux while he is a relitive nobody.
No, if Linux had not happened, he'd be a nobody. Probably. Linux (as in
The Os, not The Kernel) wouldn't have happened without the GPL. Well,
there would have been BSD...
Jochem
--
"A designer knows he has arrived at perfection not when there is no
longer anything to add, but when there is no longer anything to take
away." - Antoine de Saint-Exupery
Christopher Browne
> On 26 Jun 2003 07:55:44 -0700, steves...@yahoo.com (Steve) wrote:
>
>>http://techupdate.zdnet.com/techupdate/stories/main/0,14179,2914132,00.html
>
> RMS is STILL pissed that he did all the work and Linus became a star
> and got all the credit for Linux while he is a relitive nobody.
On the one hand, he's partly right; many people don't bother giving
the FSF _any_ credit for the user space code they provided.
On the other hand, if Linux hadn't made the user space code popular,
then it would probably be FreeBSD that would be the "star" today, and
RMS still would be, in your terms, a "relative nobody."
--
let name="cbbrowne" and tld="acm.org" in name ^ "@" ^ tld;;
http://www3.sympatico.ca/cbbrowne/lsf.html
"You shouldn't make my toaster angry."
-- Household security explained in "Johnny Quest"
Brian Masinick
> On 26 Jun 2003 07:55:44 -0700, steves...@yahoo.com (Steve) wrote:
> >http://techupdate.zdnet.com/techupdate/stories/main/0,14179,2914132,00.html
> RMS is STILL pissed that he did all the work and Linus became a star
> and got all the credit for Linux while he is a relitive nobody.
I'm not sure that's the case at all. From what I can tell, there is
little egotistical behavior present with RMS. Richard M. Stallman,
from the very beginning, merely wanted to preserve the software
freedoms that he felt should be available.
Historical accounts indicate that RMS was bothered by the fact that a
certain printer driver did not have freely available source code, so
that he could make some simple modifications to the printer status
indicators. It seemed that in his office, which had some really fast
printers, it was common for the printer to get clogged and lots of
jobs to back up, without giving much, if any, notice to users at their
desks about what was going on. Richard sought to change that, only to
find out that the guy who created the code was unwilling to share it,
and his company wanted to protect what they felt was their
"intellectual property".
Do you remember seeing mention of "intellectual property" in Richard's
message? That's what this is all about. The story I relayed above is
but one of many that Richard experienced as he worked at the
M.I.T. Artificial Intelligence Labs in Cambridge, MA. At one time,
extremely bright software researchers were able to freely share code
and ideas. Richard saw all of that dwindle before his own eyes, and
was bent on doing something about it, and made it his lifelong
pursuit.
Now there may be an element of Stallman wanting to get some credit
where credit is due, but look at the facts. He and a small array of
talented professionals spent ten years (now nearly twenty years)
developing some really solid tools, beginning with an editor, (Emacs)
a compiler, (gcc), then a whole range of commands and utilities to
completely replace UNIX software, hence the name Gnu's NOT UNIX.
About all they had left to do was to write a kernel. That work was
beginning right about the same time that Linus Torvalds began his work
on his own Linux kernel. The rest of the system (or at least the main
components) are actually GNU software, not Linux, not something else.
Commercial vendors and individuals create collections of software
called distributions, that piece together customized Linux kernels, an
assortment of GNU system and application programs, and then create
their own installation and configuration tools.
Richard may be accused of being a stickler on this, but when 75% or
more of the actual work comes from projects that were initiated from
the GNU ideas, is it too much to ask that they include at least an
acknowledgement that those parts contain GNU software? In fact, if
you look at them carefully, the licenses do that. It's my opinion
that this is not an egotistical response from Stallman, he just
doesn't want to see the ideas of software freedom get dwindled into
some other thing. That's why he's against calling it all "Open Source
Software", even though upwards of 90% of the concepts are in agreement
with the ideas of "Free Software". It's just that the one area where
they disagree happens to be the thing that perhaps matters most of all
to Richard, and that's the need to vigorously maintain software
freedom as he sees it.
Now people can disagree with Richard on this point, and they do. He
still works with so-called Open Source people and collaborates on
projects without agreeing 100% on everything. I don't see anything
egotistical at all about someone who vigorously and vehemently adheres
to what they believe in. I doubt that I agree with either Richard or
anyone else in all areas, whether they pertain to software or to life
in general. What I can say is that I have great respect for
individuals who devote their life to something they deeply believe in,
and consistently make choices that agree with those beliefs. How many
of us can honestly say that we unwaveringly do that much? In that way
alone, if for no other reason, I think that Richard is someone who is
a great role model for others to examine.
Ego? There may be one in their somewhere. Somehow, I believe that's
the least of the things that concern the gentleman we are discussing
today. I don't think that Torvalds is a big ego-head, either. Both
men do what they do because it's their life's passion. They do it
with all their energy, and whether they have a big head or not,
they've achieved more of what they've set out to do than most of us
combined. So if any of them occasionally get bloated in the head
(which I question), then I certainly forgive them, should that be the
case. Would I have a big head if I got a lot of attention? I hope
not, but who knows? Power or a sense of power does weird things to
the mind. I'd be very slow to judge anyone on this matter.
--
Brian W. Masinick
mailto:masi...@yahoo.com
Barry Margolin
Brian Masinick <masi...@yahoo.com> wrote:
>Sam Richards <m...@privacy.net> writes:
>
>> On 26 Jun 2003 07:55:44 -0700, steves...@yahoo.com (Steve) wrote:
>
>> >http://techupdate.zdnet.com/techupdate/stories/main/0,14179,2914132,00.html
>
>> RMS is STILL pissed that he did all the work and Linus became a star
>> and got all the credit for Linux while he is a relitive nobody.
>Do you remember seeing mention of "intellectual property" in Richard's
>message? That's what this is all about.
I think Sam was focusing on the earlier paragraphs that go on and on about
the difference between Linux (the kernel) and GNU/Linux (the whole system
that combines the Linux kernel and the GNU libraries and commands).
I don't know how relevant that distinction is to the SCO vs. IBM court
case. If it is, then RMS is justified in bringing it up. If not, then
he's confusing the important issue in this situation by rehashing a
nit-picking detail.
--
Barry Margolin, barry.m...@level3.com
Level(3), Woburn, MA
*** DON'T SEND TECHNICAL QUESTIONS DIRECTLY TO ME, post them to newsgroups.
Please DON'T copy followups to me -- I'll assume it wasn't posted to the group.
Brian Masinick
> In article <87wuf7d...@yahoo.com>,
> Brian Masinick <masi...@yahoo.com> wrote:
>>Sam Richards <m...@privacy.net> writes:
>>
>>> On 26 Jun 2003 07:55:44 -0700, steves...@yahoo.com (Steve) wrote:
>>
>>> >http://techupdate.zdnet.com/techupdate/stories/main/0,14179,2914132,00.html
>>
>>> RMS is STILL pissed that he did all the work and Linus became a star
>>> and got all the credit for Linux while he is a relitive nobody.
> ...
>>Do you remember seeing mention of "intellectual property" in Richard's
>>message? That's what this is all about.
>
> I think Sam was focusing on the earlier paragraphs that go on and on about
> the difference between Linux (the kernel) and GNU/Linux (the whole system
> that combines the Linux kernel and the GNU libraries and commands).
>
> I don't know how relevant that distinction is to the SCO vs. IBM court
> case. If it is, then RMS is justified in bringing it up. If not, then
> he's confusing the important issue in this situation by rehashing a
> nit-picking detail.
Richard, being Richard, takes every opportunity he can to remind
people of why GNU software is important. You should note that he
explicitly mentioned that if someone actually DID place proprietary
code anywhere in either the GNU utilities, the Linux kernel, or both,
it could and should be removed and replaced. He knows exactly what
we're dealing with here. He also knows that there are at least three
significant kernel alternatives that can be used with the GNU
application and utility layer of the system: the popular Linux kernel,
the very secure BSD kernel, or the original HURD that he intended to
couple with the other GNU components. It's a good thing that there
are all those options. Richard simply wants to make sure that people
understand the clear distinction between GNU and Linux. They're used
together, but they are distinct things. Linux is the kernel, GNU is
the application/utility layer, and distributions put their distinctive
characteristics around both pieces.
FreeBSD sometimes uses GNU components, but they typically use their
own completely rewritten utilities, and they have their own kernel.
There is no clear distinction between the FreeBSD kernel and the rest
of the system as there is with GNU and Linux because the FreeBSD
project wrote the core stuff themselves. That said, you can certainly
get many GNU utilities with FreeBSD, and you can also get plenty of
other software components that rightly ought to be considered "Open
Source Software", as opposed to the clearly distinctive GNU "Free
Software". That's a distinction that Stallman wants everyone to
understand, because it's really a moral and social issue, in his
mind. I see that point and have great respect for it. I just hope
I'm accurately representing the point. If not, refer back to the
article; he lays it out clearly.
--
Brian Masinick
mailto:masi...@yahoo.com
Artis Rozentals
> case. If it is, then RMS is justified in bringing it up. If not, then
> he's confusing the important issue in this situation by rehashing a
> nit-picking detail.
It is not very relevant in the SCO vs. IBM case. It's much more relevant
in the "SCO saying that the Linux operating system is tainted" case,
because Linux is not an operating system at all! If there would be real
problems for users with the Linux kernel it could be substituted with i.e.
the NetBSD kernel.
Wayne Throop
: [...] hence the name Gnu's NOT UNIX.
I'd have liked it better if it'd been named GINSU, even though
that would play havoc with the animal metaphors and logos.
But it would have the advantage that, upon provocation, one could say
something like "Unix license? I don't got no Unix license. I don't got
to show you no stinkin' Unix license. GINSU is no stinking Unix!"
Wayne Throop thr...@sheol.org http://sheol.org/throopw
Per Abrahamsen
> I don't know how relevant that distinction is to the SCO vs. IBM court
> case. If it is, then RMS is justified in bringing it up.
It is kind of relevant to the FUD from SCO about the Linux kernel
containing some kind of IP (code or patents) belonging to SCO. If SCO
can actually back it up, then it is important to be able to
distinguish the Linux kernel from the OS that contains a Linux
kernel.
Losing the kernel would be bad, but it is only one component of the
OS, and a component that have several pretty good free alternatives.
It would be much worse to lose e.g. X11 or GCC.
David Steuber
> RMS is STILL pissed that he did all the work and Linus became a star
> and got all the credit for Linux while he is a relitive nobody.
Even if this was true, could you blame him?
I'm sure RMS also remembers the whole Gosling Emacs thing as well.
In anycase, RMS deserves his due credit.
--
One Editor to rule them all. One Editor to find them,
One Editor to bring them all and in the darkness bind them.
(do ((a 1 b) (b 1 (+ a b))) (nil a) (print a))
Brian Masinick
> Sam Richards <m...@privacy.net> writes:
>
>> RMS is STILL pissed that he did all the work and Linus became a star
>> and got all the credit for Linux while he is a relitive nobody.
>
> Even if this was true, could you blame him?
>
> I'm sure RMS also remembers the whole Gosling Emacs thing as well.
>
> In anycase, RMS deserves his due credit.
Oh, you can be SURE that RMS remembers the Gosling Emacs thing. In
fact, that may have been one of the final things that actually helped
RMS and the FSF develop the care and accuracy to which they develop
and license software today. There were a number of differences in
ideas when the XEmacs project first came into being, but there may
have been a possibility that at least some parts of the code in XEmacs
may have been folded into GNU Emacs, but one of the issues was a
concern over who actually wrote certain parts of the code, could there
be assurance that no part of it was commercial or encumbered in any
way. I know that there were other differences, but I think that may
have been one of the biggest stumbling blocks of all to bring the two
projects together. No need to argue over it, the Free Software
Foundation's #1 goal is to protect software freedom. I think when the
XEmacs differences came about, the sting from the Gosling issues were
still very fresh in people's mind, particularly to RMS, but probably
to others, too.
So I don't have any problem at all with the fact that the FSF protects
software freedom. I don't have any problem that others want to do
things somewhat different ways, either. I'm not particularly
enchanted with software that's entirely proprietary, though. I use it
at times, but I'd prefer to use software that's really free,
especially free from licenses that prevent it from being freely shared
with others. Therefore, when I see a chance to speak up for free
software, I try to do so. I'm starting to get better at it.
Jim Janney
> Sam Richards <m...@privacy.net> writes:
>
> > On 26 Jun 2003 07:55:44 -0700, steves...@yahoo.com (Steve) wrote:
> >
> >>http://techupdate.zdnet.com/techupdate/stories/main/0,14179,2914132,00.html
> >
> > RMS is STILL pissed that he did all the work and Linus became a star
> > and got all the credit for Linux while he is a relitive nobody.
>
> No, if Linux had not happened, he'd be a nobody. Probably. Linux (as in
> The Os, not The Kernel) wouldn't have happened without the GPL. Well,
> there would have been BSD...
What does BSD use for a compiler? GCC, or something else?
--
Jim Janney
D. C. Sessions
gcc would be insecure -- the BSD Core Team hand-translate
every line to make sure that nothing unintended sneaks in.
--
| Microsoft: "A reputation for releasing inferior software will make |
| it more difficult for a software vendor to induce customers to pay |
| for new products or new versions of existing products." |
end
Kin Cho
Kin Cho
NetBSD isn't totally standalone, the NetBSD kernel emulates
(some/all?) Linux system calls for supporting the running of
applications like netscape and realplayer in Linux emulation.
Life to a NetBSD user would be pretty boring if Linux goes away,
and its vast application support dries up.
-kin
Per Abrahamsen
> Or emacs!
We have two emacsen, I could live with the other one.
> Losing Netscape would be bad too.
I think mean Mozilla or gecko, Netscape is not "ours".
Losing gecko would be bad, in the ballpark of losing Linux-the-kernel,
bug again not-to-bad alternatives (KHTML) exist.
Ron House
> Richard may be accused of being a stickler on this, but when 75% or
> more of the actual work comes from projects that were initiated from
> the GNU ideas,
Was it really? Star/Open office, Netscape, Mozilla, W3C, KDE, all the
imaging, music, drivers, X, wine, fonts, TEX, LATEX, Apache, MySQL, a
few dozen more that don't occur to me just now? And these don't "come
from GNU ideas" just because some of them use the GPL.
> is it too much to ask that they include at least an
> acknowledgement that those parts contain GNU software?
When the person wanting the acknowledgement wrote a licence that denies
others the right to enforce a claim for similar acknowledgement, and
also forces people to release their own code under that licence if they
desire to contribute, then yes, that's too much to ask.
--
Ron House ho...@usq.edu.au
http://www.sci.usq.edu.au/staff/house
Linønut
>> is it too much to ask that they include at least an
>> acknowledgement that those parts contain GNU software?
>
> When the person wanting the acknowledgement wrote a licence that denies
> others the right to enforce a claim for similar acknowledgement, and
You've misread the GPL gravely on this issue.
--
Rejuventate your hardware with Linux!
Barry Margolin
>Brian Masinick wrote:
>
>> Richard may be accused of being a stickler on this, but when 75% or
>> more of the actual work comes from projects that were initiated from
>> the GNU ideas,
>
>Was it really? Star/Open office, Netscape, Mozilla, W3C, KDE, all the
>imaging, music, drivers, X, wine, fonts, TEX, LATEX, Apache, MySQL, a
>few dozen more that don't occur to me just now? And these don't "come
>from GNU ideas" just because some of them use the GPL.
Most of those things are fairly recent additions, and not really part of
the operating system at all. They're mostly applications that have been
ported to run on the GNU/Linux OS, and which distributors often package
together with it for convenience.
Kai Großjohann
> I'm sure RMS also remembers the whole Gosling Emacs thing as well.
Stupid newbie question: what was the Goslong Emacs thing?
--
~/.signature
Barry Margolin
I think this is the issue they're referring to:
Shortly after Unipress acquired Gosling Emacs, they claimed that the
redisplay code in GNU Emacs was copied from Gosling Emacs. RMS disagreed,
but rather than debate it and eventually end up in court, he (or someone he
delegated to) totally rewrote the redisplay module.
It was probably in need of a rewrite anyway.
Joe Brenner From:
>: Brian Masinick <masi...@yahoo.com>
>: [...] hence the name Gnu's NOT UNIX.
>I'd have liked it better if it'd been named GINSU, even though
>that would play havoc with the animal metaphors and logos.
>But it would have the advantage that, upon provocation, one could say
>something like "Unix license? I don't got no Unix license. I don't got
>to show you no stinkin' Unix license. GINSU is no stinking Unix!"
Keep that in mind should a code-fork, gnu forbid, ever
become necessary.
T. Max Devlin
Thank you, Jerry
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
Dominique Devriese
> In article <3EFFE208...@usq.edu.au>, Ron House
> <ho...@usq.edu.au> wrote:
>> Brian Masinick wrote:
>>
>>> Richard may be accused of being a stickler on this, but when 75%
>>> or more of the actual work comes from projects that were initiated
>>> from the GNU ideas,
>>
>> Was it really? Star/Open office, Netscape, Mozilla, W3C, KDE, all
>> the imaging, music, drivers, X, wine, fonts, TEX, LATEX, Apache,
>> MySQL, a few dozen more that don't occur to me just now? And these
>> don't "come from GNU ideas" just because some of them use the GPL.
> Most of those things are fairly recent additions, and not really
> part of the operating system at all. They're mostly applications
> that have been ported to run on the GNU/Linux OS, and which
> distributors often package together with it for convenience.
"The GNU utilities are not part of the operating system at all. They
are mostly applications that that have been ported to run on the Linux
kernel, and which distributors often package together with it for
convenience."
Both statements are nonsense imho.
cheers
domi
Kin Cho
> Was it really? Star/Open office, Netscape, Mozilla, W3C, KDE, all
> the imaging, music, drivers, X, wine, fonts, TEX, LATEX, Apache,
> MySQL, a few dozen more that don't occur to me just now? And
> these don't "come from GNU ideas" just because some of them use
> the GPL.
The success of these apps totally rely on a solid foundation of
the free high quality GNU tools such as gcc, gdb, etc... and of
course emacs!
-kin
Brian Masinick
RMS would argue that the GNU utilities ARE a part of the operating
system, but he'd also acknowledge that they've been used on many
different systems (which was something that he may not have originally
anticipated happening). In fact, quite a bit of work has been done
over the years maintaining and improving support for the various
programs to run on many different platforms. That still doesn't
change the fact that the utilities were intended to be part of the GNU
operating system (where Gnu's NOT UNIX), and that the last part to be
written, the kernel itself, can readily be plugged in.
Right now, there are at least three distinct types of kernels that can
run with the GNU "operating system", the GNU Hurd kernel (which RMS
once intended to call Alis, after someone he was seeing at the time),
the Linux kernel, (which is probably the most common use today) or the
BSD kernel, which has been done, too (and their are multiple
variations of the BSD kernel (found in FreeBSD, OpenBSD, NetBSD, and
possibly others).
Thien-Thi Nguyen
> That still doesn't
> change the fact that the utilities were intended to be part of
if intention were fact, i'd be WAY more accomplished,
not to mention more humane, witty, and supremely exact.
if intention were fact, i could explain how i missed
opportunities sublime, exquisite, and overall, back to back. :-/
if intention were fact, we would recurse on the null list,
and M-x bootstrap-emacs-42 (not a bad hack).
if intention were fact, there would be those who insist
that the world be zero-tolerance -- a pox on the slack!
if intention were fact, how would the great laws of physics
agree to let doggerel (such as this) be posted, answer that!
thi
Ron House
> Most of those things are fairly recent additions, and not really part of
> the operating system at all. They're mostly applications that have been
> ported to run on the GNU/Linux OS, and which distributors often package
> together with it for convenience.
>
So what makes X any less "part of the operating system" than the cat
command?
Ron House
So what? Are these tools free or is there a secret price in dollops of
admiration? If the bloke who thinks others shouldn't be paid for their
intellectual creations also expected nothing for his own, or if,
expecting something for himself, promoted a licence that gave something
to others, we might at least give him the credit for being sincere. As
it is, bleating on year after year that he didn't get any kudos for his
own work, he deserves nothing.
But of course it isn't a "so what?" situation. Linux would be nowhere
without the apps such as X, just as the GNU tools would be nowhere
without the Linux kernel. (After all the noise about hurd or whatever,
it was Linux the public wanted.) The whole lot, contributions of tens of
thousands of people, all hangs together, and to sourgrape that the name
should recognise your little bit and not all the rest is quite
insupportable.
Anyway, can't stay here gabbing! Back to work with my X/Mozilla/Linux!
Chong Yidong
> Barry Margolin wrote:
>
> > Most of those things are fairly recent additions, and not really part of
> > the operating system at all. They're mostly applications that have been
> > ported to run on the GNU/Linux OS, and which distributors often package
> > together with it for convenience.
> >
>
> So what makes X any less "part of the operating system" than the cat
> command?
Suppose we assume that the X server is part of the operating system,
just as the shell utilities are. The GNU folks decided early on not to
write their own X server, since XFree86 was free and available. They
made a similar decision for Don Knuth's document processing system,
TeX. It's therefore not unreasonable to call the operating system
GNU/XFree86/Knuth/Linux.
But the GNU folks wrote most of the missing parts of the system, which
consisted of a wide variety of software: from the aforementioned shell
utilities to the shell itself, to the compiler, to the C library.
Therefore, it's also not unreasonable to call the system GNU for
short, since most of the pieces came from GNU.
Wayne Throop
: the GNU folks wrote most of the missing parts of the system, which
: consisted of a wide variety of software: from the aforementioned shell
: utilities to the shell itself, to the compiler, to the C library.
: Therefore, it's also not unreasonable to call the system GNU for
: short, since most of the pieces came from GNU.
I use the GNU tools quite a bit. Compiler, libraries, text utils,
shell, and so on and so forth. But the things I use most often aren't
GNU at all. Mozilla, X, tcl/tk, perl, python, sometimes OpenOffice,
some of the KDE tools, mbone tools... the list goes on, and mostly
it's not GNU. If linux never had the textutils, I'd use perl.
If not bash, then ksh, zsh, or tcsh. And so on. Yes, yes, they
are compiled with gcc using various gnu libs, many of them use
autoconf and make. But then... that's the same relationship
that gnu has to the linux implementation of unixoid syscalls,
yet in that case, GNU deserves to have top billing?
In short, the GNU tools are most wonderful, and the FSF deserves
plenty of credit. But they don't dominate what I use in a unixoid
environment, especially not nowdays.
Kai Großjohann
--
~/.signature
Paolo Gianrossi
> So what? Are these tools free or is there a secret price in dollops of
> admiration?
Be aware: free is free as in free speech, not free beer... ;)
free software developers (and rms included) should get kudos, credit, even
money maybe, for their work: you do expect something for your work, you
just give liberties to other...
cheers
paolino
--
God is dead. Nietzche /"\
Nietzche is dead. God \ / ASCII Ribbon Campaign
- X Against HTML Mail
Paolo Gianrossi / \
paoli...@disi.unige.it
Member of the GNU@DISI project.
Roy Culley
Barry Margolin <barry.m...@level3.com> writes:
> In article <84k7b3w...@lucy.is.informatik.uni-duisburg.de>,
> Kai Großjohann <kai.gro...@gmx.net> wrote:
>>David Steuber <david....@verizon.net> writes:
>>
>>> I'm sure RMS also remembers the whole Gosling Emacs thing as well.
>>
>>Stupid newbie question: what was the Goslong Emacs thing?
>
> I think this is the issue they're referring to:
>
> Shortly after Unipress acquired Gosling Emacs, they claimed that the
> redisplay code in GNU Emacs was copied from Gosling Emacs. RMS disagreed,
> but rather than debate it and eventually end up in court, he (or someone he
> delegated to) totally rewrote the redisplay module.
>
> It was probably in need of a rewrite anyway.
Barry Margolin
>Barry Margolin wrote:
>
>> Most of those things are fairly recent additions, and not really part of
>> the operating system at all. They're mostly applications that have been
>> ported to run on the GNU/Linux OS, and which distributors often package
>> together with it for convenience.
>>
>
>So what makes X any less "part of the operating system" than the cat
>command?
Unix systems are completely usable without X, and in fact they were used
for many years without *any* window system at all. Suns with Sunview were
just as much Unix systems as other Unix systems running X. On the other
hand, if you take out the "cat" command, you've removed a significant piece
of a system's "Unix-ness" -- zillions of shell scripts that should be
portable to any flavor of Unix would break as a result.
I admit that there's no clear dividing line between OS and non-OS, it's a
fuzzy boundary. But some things are tightly associated with Unix, while
others are "just applications" that are frequently used on Unix, but not
essential (e.g. if you don't have Emacs you can still use vi, ex, or ed).
Greg Menke
> In article <3F0245CB...@usq.edu.au>, Ron House <ho...@usq.edu.au> wrote:
> >Barry Margolin wrote:
> >
>
> I admit that there's no clear dividing line between OS and non-OS, it's a
> fuzzy boundary. But some things are tightly associated with Unix, while
> others are "just applications" that are frequently used on Unix, but not
> essential (e.g. if you don't have Emacs you can still use vi, ex, or ed).
>
I think the dividling line is quite clear. cat and X are application
software- and as you say, it is true that cat is more necessary for a
"unixy" user environment with shells and whatnot. But, if a bit of
code is on the userspace side of a system call, its not OS- and if its
on the kernel side, it is OS. To do something useful, you have to
supply some kind of userspace software, but thats entirely up to the
user.
Gregm
Barry Margolin
Greg Menke <gregm...@toadmail.com> wrote:
>I think the dividling line is quite clear. cat and X are application
>software- and as you say, it is true that cat is more necessary for a
>"unixy" user environment with shells and whatnot. But, if a bit of
>code is on the userspace side of a system call, its not OS- and if its
>on the kernel side, it is OS. To do something useful, you have to
>supply some kind of userspace software, but thats entirely up to the
>user.
So you're saying that OS == Kernel? I think common usage distinguishes
these two -- the kernel is a major component of the OS, but it's not the
whole thing.
I don't think anyone would consider a system running a Linux or BSD kernel,
but with no way to login, and none of the common commands, to be a Unix
Operating System. It's just a bare kernel. It might be usable for an
embedded system (e.g. a TiVo box), but it's not a Unix OS.
Brian Masinick
> Barry Margolin <barry.m...@level3.com> writes:
>
> I think the dividling line is quite clear. cat and X are application
> software- and as you say, it is true that cat is more necessary for a
> "unixy" user environment with shells and whatnot. But, if a bit of
> code is on the userspace side of a system call, its not OS- and if its
> on the kernel side, it is OS. To do something useful, you have to
> supply some kind of userspace software, but thats entirely up to the
> user.
I'd define kernel plus the low level interface shell as the core
operating system. Whether you'd also include core utilities as part
of the operating system might be a debatable point, I'd say that base
utilities would also be considered part of the operating system.
Window managers, editors, navigational tools, and application
software, while useful parts of a complete system, do not, by my
definition, constitute an operating system.
One Web site, http://www.webopedia.com/TERM/o/operating_system.html,
makes these distinctions and classifications:
Operating systems can be classified as follows:
# multi-user : Allows two or more users to run programs at the same
time. Some operating systems permit hundreds or even thousands of
concurrent users.
# multiprocessing : Supports running a program on more than one CPU.
# multitasking : Allows more than one program to run concurrently.
# multithreading : Allows different parts of a single program to run
concurrently.
# real time: Responds to input instantly. General-purpose operating
systems, such as DOS and UNIX, are not real-time.
Operating systems provide a software platform on top of which other
programs, called application programs, can run. The application
programs must be written to run on top of a particular operating
system. Your choice of operating system, therefore,determines to a
great extent the applications you can run
Greg Menke
> In article <m3he64g...@europa.pienet>,
> Greg Menke <gregm...@toadmail.com> wrote:
> >I think the dividling line is quite clear. cat and X are application
> >software- and as you say, it is true that cat is more necessary for a
> >"unixy" user environment with shells and whatnot. But, if a bit of
> >code is on the userspace side of a system call, its not OS- and if its
> >on the kernel side, it is OS. To do something useful, you have to
> >supply some kind of userspace software, but thats entirely up to the
> >user.
>
> So you're saying that OS == Kernel? I think common usage distinguishes
> these two -- the kernel is a major component of the OS, but it's not the
> whole thing.
It is not the whole thing if OS is viewed as including some portion of
the user interface. I do think its true that OS is more vague than
Kernel, so perhaps it is fair to say the "OS" does include that extra
stuff. Which I suppose makes Sun's "Operating Environment" possibly
more appropriate.
> I don't think anyone would consider a system running a Linux or BSD kernel,
> but with no way to login, and none of the common commands, to be a Unix
> Operating System. It's just a bare kernel. It might be usable for an
> embedded system (e.g. a TiVo box), but it's not a Unix OS.
Yet Linux would be considered to be a different Unix(like) OS than
FreeBSD even if the exact same suite of software were installed on
both. Perhaps my issue is I come from an embedded system perspective
where an operating system pretty much is just its programming
interface.
Gregm
Christopher Browne
> Greg Menke <gregm...@toadmail.com> writes:
>> Barry Margolin <barry.m...@level3.com> writes:
>> I think the dividling line is quite clear. cat and X are
>> application software- and as you say, it is true that cat is more
>> necessary for a "unixy" user environment with shells and whatnot.
>> But, if a bit of code is on the userspace side of a system call,
>> its not OS- and if its on the kernel side, it is OS. To do
>> something useful, you have to supply some kind of userspace
>> software, but thats entirely up to the user.
> I'd define kernel plus the low level interface shell as the core
> operating system. Whether you'd also include core utilities as part
> of the operating system might be a debatable point, I'd say that
> base utilities would also be considered part of the operating
> system.
That's likely easy to agree on.
> Window managers, editors, navigational tools, and application
> software, while useful parts of a complete system, do not, by my
> definition, constitute an operating system.
There seem to be multiple legitimate judgements that are possible on
this.
I think a not-unreasonable perspective is to consider that the
"operating system" ends where you start implementing things on top of
it.
In an "embedded systems" perspective, there doesn't need to be
terribly much there for you to start implementing additional things on
top of it.
On the other hand, for someone implementing "web services" in Java,
the "operating system" certainly includes a Java implementation and
possibly other stuff including J2EE libraries and even a relational
database.
Continuing on that note, in the latter days of VMS, I seem to recall
it including RDB, an SQL DBMS (good enough that Oracle bought them out
both to get the customer list as well as the implementors who had some
interesting performance techniques), and they seemed to be starting to
look towards treating that as a "base component" for other apps to
depend on having there. Some Unix-like systems won't accept "ordinary
users" until they get the LDAP or PostgreSQL database that manages the
user list running...
In these various cases, it is _not_ unnatural to consider the DBMS
part of the "operating system."
If you're a user that wants to use OpenOffice.org, Mozilla, and some
GNOME/KDE apps, it's not going to be unnatural to consider lower
layers that you don't want to pay attention to such as X and graphical
libraries to be "part of the operating system."
> Operating systems provide a software platform on top of which other
> programs, called application programs, can run. The application
> programs must be written to run on top of a particular operating
> system. Your choice of operating system, therefore, determines to a
> great extent the applications you can run
If I want to run OpenOffice.org, then I need to have a kernel, init, C
libraries, X, and PostgreSQL running. It doesn't seem too wild to
consider that this makes all of these underlying components part of my
"operating system."
If your applications are different such that some of those components
are spurious to your requirements, then you need "less operating
system" than I do. Doesn't make either of us wrong...
--
"cbbrowne","@","cbbrowne.com"
http://www.ntlug.org/~cbbrowne/oses.html
BASIC is not a language. It's a plot to sucker poor unsuspecting
consumers into believing that they should buy a computer because
ANYONE can learn how to program.
Ron House
> Ron House <ho...@usq.edu.au> writes:
>
>
>>So what? Are these tools free or is there a secret price in dollops of
>>admiration?
>
> Be aware: free is free as in free speech, not free beer... ;)
Okay, people have chosen to call it "Linux". Stallman, the advocate of
free speech, doesn't have a valid argument in the world for telling them
they should do otherwise.
But of course the whole "free speech not free beer" thing is a hypocrisy
at the outset, because the predicted and known effect of the licence is
to prevent authors of any code released under it from having any way to
obtain payment for that code, so it well and truly is free in the sense
of beer. Irrelevant persiflage about "they can charge for support"
notwithstanding, the fact is they can't get paid for their work output,
the software product, because the licence ensures otherwise. Therefore
the whole free speech furphy qualifies precisely for the true definition
of hypocrisy: pretending to do good.
Greg Menke
This is incorrect. Its true you cannot charge for access to the
sourcecode, but you can certainly make money using it for consulting-
or as you say, charging for support. You can't make monopoly profits,
but its entirely possible to make a reasonable living with it.
I'll give you an example, the RTEMS realtime operating system. Go to
http://www.rtems.com. The company (OAR) that administers (and wrote
it), releases the source under a BSD-ish license- you can download the
entire source tree, compiler suite, documentation, etc and start using
it right away. There are no royalties or licensing fees and even
better, if you find a bug, you can trace it and fix it without any
fooling around with NDA's and support plans. Its hard to overstate
the advantages of being able to cm everything from the compiler source
up through the OS and application software.
RTEMS has been under active development for about 10 years and
competes directly with vxWorks on a variety of platforms, many of
which are long abandoned or at least nominally unsupported by Wind
River.
Theres an active mailing list you can use to ask technical support
questions, both of the authors of much of the RTEMS kernel and of the
many contributors. You can also hire OAR to write you a driver or
support some local hardware, or just for help in using RTEMS for some
purpose. The tradeoff in investing time learning the OS internals &
compiler suite vs working on your application is entirely at your
discretion.
We've done both; we bought a bunch of help from OAR to implement
support for our weirdo MIPS boards, developing a remote GDB stub and
porting various GPL software packages to RTEMS- and we've contributed
a good bit back to RTEMS as well. I get more & better support from
the RTEMS mailing list than I've ever gotten from a for-pay support
arrangement.
I make money using RTEMS on projects, and the better it is, the more
effective I am at using it, so its in my best interest to cultivate
other users and contribute my work back into it. My clients like it
because its cheap (no install licensing issues), it performs well, its
flexible (you can trivially retarget to any number of entirely
different architectures with no makefile changes or OS-specific
refactoring), and they can inspect the entire source tree anytime they
like. I imagine other users may have a similar perspective and it
engenders a dynamic and highly cooperative community.
I would no doubt disagree with RMS on a variety of issues, but I
regularly contribute money to the FSF because the idea works. Its not
perfect or universally applicable, but its a far better deal for the
user than a software monopoly.
Gregm
Ron House
> Ron House <ho...@usq.edu.au> writes:
>>But of course the whole "free speech not free beer" thing is a
>>hypocrisy at the outset, because the predicted and known effect of the
>>licence is to prevent authors of any code released under it from
>>having any way to obtain payment for that code, so it well and truly
>>is free in the sense of beer. Irrelevant persiflage about "they can
>>charge for support" notwithstanding, the fact is they can't get paid
>>for their work output, the software product, because the licence
>>ensures otherwise. Therefore the whole free speech furphy qualifies
>>precisely for the true definition of hypocrisy: pretending to do good.
>
>
> This is incorrect. Its true you cannot charge for access to the
> sourcecode,
Then it is CORRECT. I said you can't get paid for the software product,
and you agree that is CORRECT.
> but you can certainly make money using it for consulting-
> or as you say, charging for support. You can't make monopoly profits,
> but its entirely possible to make a reasonable living with it.
Which is exactly what I said. Make no mistake, I have no problem
whatsoever with open-source licences - I even authored one. And I
applaud those who want to give away their work. But I cannot agree with
RMS's religious view that code writers are not entitled to charge for
their output. Putting in work to make software is no different from
putting in work to make a tangible thing. The fact that the former can
be infinitely reproduced does indeed make a very big difference in what
one would consider moral charging models, but it doesn't invalidate the
right to be paid. RMS disagrees with that right and wrote a licence to
enforce it on anyone who wants to cooperate in creating a software
system. That is immoral. As for the work you describe on RTEMS, I have
nothing but praise for that. The problem isn't the cooperation or the
open source, it is the enslavement to RMS's notions for anyone who
wishes to make a contribution, THAT is the problem I have. In other
words, I disagree with the GPL specifically, not with any other
open-source or free licence that I know of.
Kin Cho
> Which is exactly what I said. Make no mistake, I have no
> problem whatsoever with open-source licences - I even authored
> one. And I applaud those who want to give away their work. But
> I cannot agree with RMS's religious view that code writers are
> not entitled to charge for their output. Putting in work to
> make software is no different from putting in work to make a
> tangible thing. The fact that the former can be infinitely
> reproduced does indeed make a very big difference in what one
> would consider moral charging models, but it doesn't invalidate
> the right to be paid. RMS disagrees with that right and wrote a
> licence to enforce it on anyone who wants to cooperate in
> creating a software system. That is immoral. As for the work
> you describe on RTEMS, I have nothing but praise for that. The
> problem isn't the cooperation or the open source, it is the
> enslavement to RMS's notions for anyone who wishes to make a
> contribution, THAT is the problem I have. In other words, I
> disagree with the GPL specifically, not with any other
> open-source or free licence that I know of.
Free software isn't the same thing as the software you write for
your company. Free software isn't work, it's software you write
in your free time. As the book says, it's "Just for Fun".
-kin
calmar
>
>> but you can certainly make money using it for consulting-
>> or as you say, charging for support. You can't make monopoly profits,
>> but its entirely possible to make a reasonable living with it.
>
When you're auto-mechanic, you would probably hate when you have
to bring your car to a garage even when you know you could do it
cheaper and better at home or so.
cars are allowed to setup/repair by private person --> source
available..
--> what is very desirable when you know something about cars.
Still you can make enough money with people hiring you for
repairing their cars.
Anyway, of course there should always be a possibility to get
paied for your work, does this automatically mean that the source
should not be availabe and the right to work on it (what you
would prefer when you know something about it)
--
calmar (o_
//\ <--- GNU/Linux is GREAT
V_/_ www.calmar.ws
Martin Thornquist
> cars are allowed to setup/repair by private person --> source
> available..
One rather big difference is that you cannot download cars for free
off the Internet. It's more like getting code along with a program you
buy and being allowed to modify it for your own use, but not freely
redistribute it.
Martin
--
"An ideal world is left as an exercise to the reader."
-Paul Graham, On Lisp
calmar
> [ m...@calmar.ws ]
>
>> cars are allowed to setup/repair by private person --> source
>> available..
>
> One rather big difference is that you cannot download cars for free
> off the Internet.
That's true, but many Linux-users buy distributions like Suse and
whatever there is out anyway - which would be similar like buying
a car. (even so in fact they buy especiallly the support from
those companies (even when it's only `built-in' support by
admin-tools written by them and of course they can get a complete
system without much hassle with such distros)
> It's more like getting code along with a program you
> buy and being allowed to modify it for your own use, but not freely
> redistribute it.
>
Would also be some kind of reasonable Licence I guess.
Of course it's weird that you work very long on a program, and
finally you earn not that much with that kind of licences.
But finally, the FSF.. concepts seems to work greatly anyway :-)
greetings
Henrik Enberg
> Free software isn't the same thing as the software you write for
> your company. Free software isn't work, it's software you write
> in your free time. As the book says, it's "Just for Fun".
My, this is a new strange definition I've never heard before. What book
are you talking about?
Greg Menke
> Greg Menke wrote:
> > Ron House <ho...@usq.edu.au> writes:
>
> >>But of course the whole "free speech not free beer" thing is a
> >>hypocrisy at the outset, because the predicted and known effect of the
> >>licence is to prevent authors of any code released under it from
> >>having any way to obtain payment for that code, so it well and truly
> >>is free in the sense of beer. Irrelevant persiflage about "they can
> >>charge for support" notwithstanding, the fact is they can't get paid
> >>for their work output, the software product, because the licence
> >>ensures otherwise. Therefore the whole free speech furphy qualifies
> >>precisely for the true definition of hypocrisy: pretending to do good.
> > This is incorrect. Its true you cannot charge for access to the
> > sourcecode,
>
> Then it is CORRECT. I said you can't get paid for the software
> product, and you agree that is CORRECT.
So the GPL is hypocrisy simply because you cannot charge for access to
sourcecode? Even when its the author's choice when to use it?
Gregm
Rau, Roland
Best,
Roland
Detlev Zundel
> But of course the whole "free speech not free beer" thing is a
> hypocrisy at the outset, because the predicted and known effect of the
> licence is to prevent authors of any code released under it from
> having any way to obtain payment for that code, so it well and truly
> is free in the sense of beer. Irrelevant persiflage about "they can
> charge for support" notwithstanding, the fact is they can't get paid
> for their work output, the software product, because the licence
> ensures otherwise. Therefore the whole free speech furphy qualifies
> precisely for the true definition of hypocrisy: pretending to do good.
Well, just because you cannot imagine a business model involving Free
Software doesn't mean a few (or even lots) of us making money with it
;)
In my case for example, I write extensions to GPLed programs as a free
lancer and I get paid for the effort to write the code which is of
course GPLed in the end. So I /do get paid/ for writing the software,
I just cannot copy and sell it zillions of times, but then again think
about a carpenter trying to copy a table he just built ;)
Cheers
Detlev
--
.. the tools we are trying to use and the language we or notation we are using
to express or record our thoughts, are the major factors determining what we
can think or express at all!
-- Edsger W. Dijkstra
Matthew Kennedy
> is free in the sense of beer. Irrelevant persiflage about "they can
> charge for support" notwithstanding, the fact is they can't get paid
> for their work output, the software product, because the licence
How convenient to call it an "irrelevant persiflage", when it is in
fact the core business model of several large commercial GNU/Linux
distributions.
Matt
--
Give them RADAR-GUIDED SKEE-BALL LANES and VELVEETA BURRITOS!!
Paul Jarc
> But of course the whole "free speech not free beer" thing is a
> hypocrisy at the outset, because the predicted and known effect of the
> licence is to prevent authors of any code released under it from
> having any way to obtain payment for that code, so it well and truly
> is free in the sense of beer.
You have misunderstood "free as in free speech, not free beer". The
point of it is *not* that "free as in beer" is denied, but that "free
is in speech" is what's important. I.e., it is important that we have
freedom with our software, but it is not important (one way or the
other) whether the software is available for zero price. You may
disagree with this position; you may say that your misinterpretation
was someone else's fault for not presenting the position clearly; but
it is not hypocritical.
paul
Paul Jarc
> But I cannot agree with RMS's religious view that code writers are
> not entitled to charge for their output. Putting in work to make
> software is no different from putting in work to make a tangible
> thing. The fact that the former can be infinitely reproduced does
> indeed make a very big difference in what one would consider moral
> charging models, but it doesn't invalidate the right to be paid.
You can certainly charge for your work - once. If you want to charge
for the *same* work many times, by charging for each
trivially-produced additional copy, that is harder (but still not
impossible; RedHat et. al. make money on copies due to brand
recognition). What is your moral justification for the "right" to
charge for the same work many times?
> RMS disagrees with that right and wrote a licence to enforce it on
> anyone who wants to cooperate in creating a software system.
"... cooperate in creating *his* software system", you mean. You're
under no obligation to use any particular license for your own works.
paul
Mart van de Wege
> Barry Margolin writes:
>
> > In article <3EFFE208...@usq.edu.au>, Ron House
> > <ho...@usq.edu.au> wrote:
> >> Brian Masinick wrote:
> >>
> >>> Richard may be accused of being a stickler on this, but when 75%
> >>> or more of the actual work comes from projects that were initiated
> >>> from the GNU ideas,
> >>
> >> Was it really? Star/Open office, Netscape, Mozilla, W3C, KDE, all
> >> the imaging, music, drivers, X, wine, fonts, TEX, LATEX, Apache,
> >> MySQL, a few dozen more that don't occur to me just now? And these
> >> don't "come from GNU ideas" just because some of them use the GPL.
>
> > Most of those things are fairly recent additions, and not really
> > part of the operating system at all. They're mostly applications
> > that have been ported to run on the GNU/Linux OS, and which
> > distributors often package together with it for convenience.
>
> "The GNU utilities are not part of the operating system at all. They
> are mostly applications that that have been ported to run on the Linux
> kernel, and which distributors often package together with it for
> convenience."
>
> Both statements are nonsense imho.
OTOH,
Try running a non-embedded Linux system without glibc.
You do know what the g in glibc stands for, don't you?
I'd say that the C library would be a fairly important part of the
operating system, and on Linux systems the C library is GNU libc
almost without fail.
I personally think that the most compelling argument for GNU/Linux,
although I am somewhat sloppy in the use of the correct name.
Mart
--
The body bags and little rags of children torn in two
And the jellied brains of those who remain to put the finger right on you.
As the madmen play on words and make us all dance to their song,
To the tune of starving millions to make a better kind of gun.
Thien-Thi Nguyen
> You may disagree with this position; you may say that your
> misinterpretation was someone else's fault for not presenting
> the position clearly; but it is not hypocritical.
but i gotta be me, gotta be free,
gotta fight for my right
to ignorantly NOT see!
forget about reason, forget about pleasin',
forget everything else
just hear my great wheezin'!
WHO banged the keyboard, WHO hacked while you snored?
my momma didn't raise no sucker;
i won't stand to be IGnored!
besides, my code is better, right down to the letter,
it needs a closed license
to feel that much unfettered!
tell ya what's hypocritical: all those klepto mysticals,
best o' luck unfreezing their tongues
when they chomp on my popsicle!
oop ack gar gar gar!
thi
Ron House
Try to pay attention. I think the GPL is a bad, defective licence, but
what I called hypocrisy was the claim that "It is free as in free
speech, not free as in free beer." Because it IS free as in free beer.
The licence prevents you obtaining payment for the software product (and
NO, please don't go around the "but you can charge for support" merry go
round again). It is arguing for the morality of something on the basis
of a fundamental misrepresentation of its nature.
> Even when its the author's choice when to use it?
Groan. The reason I object to GPL but not to ANY OTHER free or open
source licence that I know of is precisely because it prevents someone
wanting to contribute from releasing THEIR code on THEIR terms. That is,
it is exactly and only its tendency to make people release code on OTHER
THAN their own preferred terms that is the problem. Suppose I get a
beaut GPL program, but find it doesn't work exactly how I like. So I
make lots of neat changes and now it's magnificent. My friends see it
and say "Hey, can I have a copy?" I now have the choice of either saying
no, or releasing MY code on terms that are not of MY liking. And most
people, as they are usually well-meaning, kind, and generous, will
decide to release the code with what is to them an unsatisfactory
licence. This is my problem with it: it punishes generosity.
Jay Belanger
Ron House <ho...@usq.edu.au> writes:
> Greg Menke wrote:
> > Ron House <ho...@usq.edu.au> writes:
>
> > So the GPL is hypocrisy simply because you cannot charge for access to
> > sourcecode?
>
> Try to pay attention. I think the GPL is a bad, defective licence, but
> what I called hypocrisy was the claim that "It is free as in free
> speech, not free as in free beer." Because it IS free as in free beer.
> The licence prevents you obtaining payment for the software product (and
> NO, please don't go around the "but you can charge for support" merry go
> round again). It is arguing for the morality of something on the basis
> of a fundamental misrepresentation of its nature.
Would "It is free as in free speech, but not necessarily free as in
free beer" make you happier? The point being, of course, that the
"free as in free beer" is irrelevant when speaking of free software in
the GNU sense. Since the word "free" can be confusing to some, the
precise meaning that is intended is made clear.
> > Even when its the author's choice when to use it?
>
> Groan. The reason I object to GPL but not to ANY OTHER free or open
> source licence that I know of is precisely because it prevents someone
> wanting to contribute from releasing THEIR code on THEIR terms. That is,
> it is exactly and only its tendency to make people release code on OTHER
> THAN their own preferred terms that is the problem. Suppose I get a
> beaut GPL program, but find it doesn't work exactly how I like. So I
> make lots of neat changes and now it's magnificent. My friends see it
> and say "Hey, can I have a copy?" I now have the choice of either saying
> no, or releasing MY code on terms that are not of MY liking.
Only your code? You seem to want the option of releasing other
people's code on terms that aren't of their liking. How hypocritical
of you.
Jay
David Kastrup
> Greg Menke wrote:
> > Ron House <ho...@usq.edu.au> writes:
>
> > So the GPL is hypocrisy simply because you cannot charge for access to
> > sourcecode?
>
> Try to pay attention. I think the GPL is a bad, defective licence,
> but what I called hypocrisy was the claim that "It is free as in
> free speech, not free as in free beer." Because it IS free as in
> free beer. The licence prevents you obtaining payment for the
> software product (and NO, please don't go around the "but you can
> charge for support" merry go round again).
I earn my money with GPLed software. People pay to have it developed
and supported by me. I am not a software distributor, working with
mass reproduction of software. And I am not getting into that
business.
> Groan. The reason I object to GPL but not to ANY OTHER free or open
> source licence that I know of is precisely because it prevents
> someone wanting to contribute from releasing THEIR code on THEIR
> terms.
They can release their code on their terms all they want. But they
won't release my code on their terms.
> Suppose I get a beaut GPL program, but find it doesn't work exactly
> how I like. So I make lots of neat changes and now it's
> magnificent. My friends see it and say "Hey, can I have a copy?" I
> now have the choice of either saying no, or releasing MY code on
> terms that are not of MY liking.
It's not your code. It's the code of someone else with changes of
yours. You demand that you should be the one deciding how the code of
somebody else is rereleased. At the same time you complain when
somebody else demands how your changes to his code should be released.
That's hypocrisy.
--
David Kastrup, Kriemhildstr. 15, 44793 Bochum
T. Max Devlin
>Greg Menke wrote:
> > Ron House <ho...@usq.edu.au> writes:
>
> > So the GPL is hypocrisy simply because you cannot charge for access to
> > sourcecode?
>
>Try to pay attention. I think the GPL is a bad, defective licence, but
>what I called hypocrisy was the claim that "It is free as in free
>speech, not free as in free beer." Because it IS free as in free beer.
Only to end-users. To developers, it is only free as in free speech. You
can't copy someone else's speech.
Somehow, due to the magic of computers, though, you can copy their beer. ;-)
>The licence prevents you obtaining payment for the software product (and
>NO, please don't go around the "but you can charge for support" merry go
>round again).
How about "but you can charge for any software product you want, except
licenses"?
>It is arguing for the morality of something on the basis
>of a fundamental misrepresentation of its nature.
It's nature is a procedural instruction set. How "a software product" relates
to that is an issue of the market, not the nature of software. In a market
where end user software licenses are the issue, it makes sense that you can't
see "a software product" as being comprised of anything else.
> > Even when its the author's choice when to use it?
>
>Groan. The reason I object to GPL but not to ANY OTHER free or open
>source licence that I know of is precisely because it prevents someone
>wanting to contribute from releasing THEIR code on THEIR terms.
Not really. As long as their terms are consistent with the GPL, there's no
problem.
(Okay; so it seems disingenuous to say that, knowing that only the GPL is
generally consistent with the GPL. But seeing as this only ensures source
code is always available, I honestly can't see the problem, so it doesn't seem
all that disingenuous to me.)
>That is,
>it is exactly and only its tendency to make people release code on OTHER
>THAN their own preferred terms that is the problem.
Yes, but this is ONLY true in so far as THEIR code is derived from code OTHER
THAN their own.
>Suppose I get a
>beaut GPL program, but find it doesn't work exactly how I like. So I
>make lots of neat changes and now it's magnificent. My friends see it
>and say "Hey, can I have a copy?" I now have the choice of either saying
>no, or releasing MY code on terms that are not of MY liking.
Who's code?
Yea, I can see how disagreeing with US Copyright law as to what constitutes a
'derivative work' can be a real annoyance. Nevertheless, that's the law, and
without it, you don't have any right to your "intellectual property" to begin
with, so....
>And most
>people, as they are usually well-meaning, kind, and generous, will
>decide to release the code with what is to them an unsatisfactory
>licence. This is my problem with it: it punishes generosity.
Ooh, yeah, ouch; punishes it real bad. Guffaw.
Most people, having derived their code from GPL code, have no problem with
GPLing their code. What kind of troll are you, Ron?
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
Per Abrahamsen
> The licence prevents you obtaining payment for the software product (and
> NO, please don't go around the "but you can charge for support" merry go
> round again).
Can you imagine how a sailer who has just travelled around the world
feel when some land-dweller tells him that the Earth is flat?
I can. I make a living producing GPL'ed software. Or more
accurately, like most employees I make a living solving problems for
my employer. In my case, that usually involves writing some software
to solve the problem, because that is where my skills are. But the
software is not the product, it is a byproduct of solving the
problems.
The vast majority of programmers are in a similar situation, they are
employed to solve problems, the software the produce is just a mean to
that end. It is not surprising that the mundanes fail to realise this,
as the software they see is shrink-wrapped from the local computer
outlet. Just like the world from the fixed point of a land-dwellers
look basically flat, if a bit rough at some places.
I happen to be able to share the software with the world freely under
the GPL because a) I want to, and my employer sees no reason not to
grant me that as long as his problems get solved, b) additions made by
others may benefit us for free, and c) that allows me to link GPL'ed
code in the software (thus saving time) without damaging our ability
to share the software with partners.
For smart employers a) and b) are enough to allow the release of the
code, but for the thick-headed c) may help as well.
Currently, all the large and visible GPL'ed projects have multiple
people who obtain payment for developing the code, usually full-time.
And there are to all likelihood far more people like me working on
less visible vertical GPL'ed applications full-time getting a nice
steady monthly paycheck for our work.
The land-dwellers are wrong. The world is round (more or less). And
you can get paid for writing GPL'ed code. Many do, and the number
seem to be increasing all the time.
Of course, there will always be some people who insisting on ignoring
the evidence will claim the world is flat, never move from their safe
house on land. These people will probably also keep claiming that the
only way to make money on software is to forbid others from
distributing it. Certainly, saying otherwise would require accepting
new ideas.
Dave Leigh
<3F04FAEA...@usq.edu.au>:
> Greg Menke wrote:
> > Ron House <ho...@usq.edu.au> writes:
>
> > So the GPL is hypocrisy simply because you cannot charge for access to
> > sourcecode?
>
> Try to pay attention. I think the GPL is a bad, defective licence, but
> what I called hypocrisy was the claim that "It is free as in free
> speech, not free as in free beer." Because it IS free as in free beer.
> The licence prevents you obtaining payment for the software product (and
> NO, please don't go around the "but you can charge for support" merry go
> round again). It is arguing for the morality of something on the basis
> of a fundamental misrepresentation of its nature.
The license does nothing of the sort. You can charge what you like, but
source code or access to it must be included as part of the deal.
> > Even when its the author's choice when to use it?
>
> Groan. The reason I object to GPL but not to ANY OTHER free or open
> source licence that I know of is precisely because it prevents someone
> wanting to contribute from releasing THEIR code on THEIR terms. That is,
> it is exactly and only its tendency to make people release code on OTHER
> THAN their own preferred terms that is the problem. Suppose I get a
> beaut GPL program, but find it doesn't work exactly how I like. So I
> make lots of neat changes and now it's magnificent. My friends see it
> and say "Hey, can I have a copy?" I now have the choice of either saying
> no, or releasing MY code on terms that are not of MY liking. And most
> people, as they are usually well-meaning, kind, and generous, will
> decide to release the code with what is to them an unsatisfactory
> licence. This is my problem with it: it punishes generosity.
You had two other options before you ever started work. First, if it didn't
work how you like, you -- KNOWING IN ADVANCE that you don't like the GPL's
terms -- could have written competing software of your own. Second, you
could have submitted a feature request to the original authors just as you
would with proprietary software.
Your argument's without merit. The original authors already handed you a
"beaut GPL program"... you said so yourself. You're not concerned over the
bulk of the program, which is somebody else's work from concept to
implementation, but because it didn't work "exactly" as you like. Yet
somehow you feel that your minor changes to this beaut program entitle you
to punish the generosity of the original authors by changing the terms of
THEIR licensing so you can distribute THEIR work on terms not to THEIR
liking. The GPL shields authors from that kind of abuse, so go find a
BSD-licensed program to modify and call 'your code.'
--
Dave Leigh, Consulting Systems Analyst
Cratchit.org
Greg Menke
So your issue is that since it prevents you from restricting someone
else from enjoying the same freedom of access to the source that you
enjoyed the GPL does not match a particular meaning of "free" that you
want it to?
The GPL creates a software commons where everyone has free access to
the sourcecode. If you don't like the terms, then simply do not
modify GPL software. If the GPL inconveniences some business plan of
yours, it hardly makes the GPL "hypocrisy". Your convienence and
pleasure are not the GPL's concern, it is there to stop you from
inconveniencing someone else.
The GPL doesn't stop you from being generous, your desire to get paid
for access to your changes to software that everyone else has freely
contributed to is what stops you from being generous.
Personally, I like the fact that contributions I make to a GPL system
will remain under the same terms. Sure, it'd be nice to get nice, fat
checks- but I'm glad to not be paying them.
Gregm
Brian Masinick
There are quite a few erroneous comments that have been exchanged
here. The GPL doesn't, and has never, prohibited individuals from
charging for their software or even for access to the source code.
RMS himself used to charge $150 when he'd produce and send a tape to
people interested in GNU Emacs. Even today, the Free Software
Foundation funds some of it's work through sales of GNU software and
documentation.
At the same time, the GPL clearly states that source code must be made
available. At no time does it ever prohibit you from charging whatever
fee people are willing to pay for access to that source code or the
application, it merely guarantees the continued availability and
sharing of that source code and it also guarantees YOU as the
continued owner of the rights to that software, unless you choose to
give up those rights yourself.
--
Brian Masinick
mailto:masi...@yahoo.com
Mike
The GPL allows charging for the media and handling. When RMS made
the tape and charged $150 USD he received the $150 USD for the media,
making the tape, shipping the tape, etc. He received no funds for
the GPL code written to the tape.
David Kastrup
> In article <87znjuk...@noname.nodomain.none>, Brian Masinick wrote:
> >
> > There are quite a few erroneous comments that have been exchanged
> > here. The GPL doesn't, and has never, prohibited individuals from
> > charging for their software or even for access to the source code.
> > RMS himself used to charge $150 when he'd produce and send a tape
> > to people interested in GNU Emacs. Even today, the Free Software
> > Foundation funds some of it's work through sales of GNU software
> > and documentation.
>
> The GPL allows charging for the media and handling. When RMS made
> the tape and charged $150 USD he received the $150 USD for the
> media, making the tape, shipping the tape, etc. He received no funds
> for the GPL code written to the tape.
I recommend you read the GPL again. The charges for media and
handling are in the clause that determines how to deal with the
situation that you have only provided binaries. Only in _this_ case
are you required to provide the source code at no more than nominal
cost. You can charge what amount you want for sublicensing/copying
the source code. $150 USD certainly is not "at cost", and the $10000
or whatever it was that the FSF charges for a Deluxe GNU distribution
certainly isn't either.
I just leave the below passage in, since it already explained that:
> > At the same time, the GPL clearly states that source code must be
> > made available. At no time does it ever prohibit you from charging
> > whatever fee people are willing to pay for access to that source
> > code or the application, it merely guarantees the continued
> > availability and sharing of that source code and it also
> > guarantees YOU as the continued owner of the rights to that
> > software, unless you choose to give up those rights yourself.
--
Paul Jarc
> Try to pay attention. I think the GPL is a bad, defective licence, but
> what I called hypocrisy was the claim that "It is free as in free
> speech, not free as in free beer." Because it IS free as in free beer.
Try to pay attention. I explained how you misunderstood "free speech,
not free beer", and regardless, some do indeed make money from free
software. If you want to ignore these facts, please keep your
ignorance to yourself.
paul
Alan Mackenzie
[ .... ]
> Currently, all the large and visible GPL'ed projects have multiple
> people who obtain payment for developing the code, usually full-time.
Really? Who all gets paid to develop Emacs (whether GNU or X)? Just
curious.
--
Alan Mackenzie (Munich, Germany)
Email: aa...@muuc.dee; to decode, wherever there is a repeated letter
(like "aa"), remove half of them (leaving, say, "a").
Thien-Thi Nguyen
> Really? Who all gets paid to develop Emacs (whether GNU or X)? Just
> curious.
i get paid to enhance emacs (on occasion) but not by the fsf, and
probably not near as much what i could be paid were i to succumb to a
day job. (no complaints, however -- knock on wood.)
i must say the GPL keeps particular areas of gig negotiation dead
simple; in most cases no lawyers are necessary, and in those cases where
it is part of the client's standard procedure (unavoidable), the air is
easily cleared by emphasizing the modification nature of the work over
the "write new code" nature. everybody wins, including the lawyer, who
doesn't have to waste time w/ this particular code monkey.
a lot of people who view the GPL unfavorably probably don't realize how
much time they can save not talking to lawyers by sticking w/ the GPL,
after some amount of ramp-up to grok it in fullness, of course. perhaps
they have a secret desire to be lawyers, themselves, in order to apply
their wit and energy towards the influence of other people's take on
reality. me, i just try to convince the computer to listen to me and do
something useful for a change!
thi
T. Max Devlin
That was downright beautiful, Thi.
Ron House
> They can release their code on their terms all they want. But they
> won't release my code on their terms.
Excellent philosophy. I agree completely.
>>Suppose I get a beaut GPL program, but find it doesn't work exactly
>>how I like. So I make lots of neat changes and now it's
>>magnificent. My friends see it and say "Hey, can I have a copy?" I
>>now have the choice of either saying no, or releasing MY code on
>>terms that are not of MY liking.
>
>
> It's not your code. It's the code of someone else with changes of
> yours.
My code can be in files entirely of my writing, and yours ditto, but if
I release your files under your licence and my files under mine, the GPL
purports to say that I have violated something because I have
distributed them together. See the Mozilla licence for an example of how
your rights to your code can be protected without interfering with my
rights to mine.
You should be aware, also, that under copyright law, if I modify your
code with a small change, you still own the copyright, so if changes are
confined to your files, the whole question doesn't arise under any
licence. The problem is the GPL's stickiness, nothing else. Your whole
reply is not to the point.
As an aside, I doubt that the GPL is really sound and can do what it
claims, but that just makes it a bungling villain, not a good guy.
David Kastrup
> David Kastrup wrote:
>
> > They can release their code on their terms all they want. But they
> > won't release my code on their terms.
>
> Excellent philosophy. I agree completely.
>
> >>Suppose I get a beaut GPL program, but find it doesn't work exactly
> >>how I like. So I make lots of neat changes and now it's
> >>magnificent. My friends see it and say "Hey, can I have a copy?" I
> >>now have the choice of either saying no, or releasing MY code on
> >>terms that are not of MY liking.
> > It's not your code. It's the code of someone else with changes of
> > yours.
>
> My code can be in files entirely of my writing, and yours ditto, but
> if I release your files under your licence and my files under mine,
> the GPL purports to say that I have violated something because I
> have distributed them together.
If both are intended to form one inseparable executable, you are
building on my work. It is not the GPL that says anything about this,
it is common copyright law.
> You should be aware, also, that under copyright law, if I modify
> your code with a small change, you still own the copyright, so if
> changes are confined to your files, the whole question doesn't arise
> under any licence.
Wrong. The point is that copyright law _prohibits_ _any_
redistribution of copies, modified or not, unless additional licensing
is agreed upon between distributor and recipient of software. The GPL
could stipulate that you were only allowed to redistribute software
when singing in a high-pitched voice "London bridge is falling down",
and there would be no way for you to navigate around that requirement
except for negotiating different terms, since refusing those terms
leaves you with the defaults according to copyright law, and those
prohibit redistribution.
> The problem is the GPL's stickiness, nothing else. Your whole reply
> is not to the point.
>
> As an aside, I doubt that the GPL is really sound and can do what it
> claims, but that just makes it a bungling villain, not a good guy.
Oh, the GPL can do very well what it claims because every recipient is
completely free to ignore it and not accept its terms. But that does
not allow you to invent different terms. And it will buy you nothing,
since in contrast to other licences, the GPL does not try to restrict
your rights beyond what the law provides as default, but extend them.
If you refuse to have your rights extended, go for it, feel free.
T. Max Devlin
>
>> They can release their code on their terms all they want. But they
>> won't release my code on their terms.
>
>Excellent philosophy. I agree completely.
>
>>>Suppose I get a beaut GPL program, but find it doesn't work exactly
>>>how I like. So I make lots of neat changes and now it's
>>>magnificent. My friends see it and say "Hey, can I have a copy?" I
>>>now have the choice of either saying no, or releasing MY code on
>>>terms that are not of MY liking.
>>
>> It's not your code. It's the code of someone else with changes of
>> yours.
>
>My code can be in files entirely of my writing, and yours ditto, but if
>I release your files under your licence and my files under mine, the GPL
>purports to say that I have violated something because I have
>distributed them together.
If your code in "your files" relies on and requires my code in *libraries*,
then yours is derivative of mine. Sorry; that's the way it works. Write your
own libraries if it's a problem for you.
>See the Mozilla licence for an example of how
>your rights to your code can be protected without interfering with my
>rights to mine.
You are confused. The GPL does not at all interfere with your rights. Only
with your privileges. You have the privilege of seeing GPL code; only if you
promise to distribute any changes likewise are you allowed to improve on it
(and distribute your improvements.)
It sounds like it would be very rare that anyone would bother improving on it,
right? No, wrong; it is the most attractive offer anyone concerned with
software development has ever seen.
Those interested in software ownership, which is to say charging for eulas,
are flabbergasted.
>You should be aware, also, that under copyright law, if I modify your
>code with a small change, you still own the copyright, so if changes are
>confined to your files, the whole question doesn't arise under any
>licence.
What the hell is this supposed to mean?
>The problem is the GPL's stickiness, nothing else. Your whole
>reply is not to the point.
You wish.
>As an aside, I doubt that the GPL is really sound and can do what it
>claims, but that just makes it a bungling villain, not a good guy.
It is sound, it does do what it claims. You are odd.
Linønut
> As an aside, I doubt that the GPL is really sound and can do what it
> claims, ...
I dunno. RMS sounds like a pretty logical and analytical fellow.
--
Rejuventate your hardware with Linux!
Linønut
>>See the Mozilla licence for an example of how your rights to your code
>>can be protected without interfering with my rights to mine.
>
> You are confused. The GPL does not at all interfere with your rights.
> Only with your privileges. You have the privilege of seeing GPL code;
> only if you promise to distribute any changes likewise are you allowed
> to improve on it (and distribute your improvements.)
>
> It sounds like it would be very rare that anyone would bother
> improving on it, right? No, wrong; it is the most attractive offer
> anyone concerned with software development has ever seen.
>
> Those interested in software ownership, which is to say charging for
> eulas, are flabbergasted.
This is a very nice way of putting it.
Rob Thorpe
> David Kastrup wrote:
>
> > They can release their code on their terms all they want. But they
> > won't release my code on their terms.
>
> Excellent philosophy. I agree completely.
>
> >>Suppose I get a beaut GPL program, but find it doesn't work exactly
> >>how I like. So I make lots of neat changes and now it's
> >>magnificent. My friends see it and say "Hey, can I have a copy?" I
> >>now have the choice of either saying no, or releasing MY code on
> >>terms that are not of MY liking.
> >
> >
> > It's not your code. It's the code of someone else with changes of
> > yours.
>
> My code can be in files entirely of my writing, and yours ditto, but if
> I release your files under your licence and my files under mine, the GPL
> purports to say that I have violated something because I have
> distributed them together.
By the same logic I could rewrite the last few pages of "Pavane" by
Keith Roberts (the ending sucks) and publish it myself :-)
mlw
> Greg Menke wrote:
> > Ron House <ho...@usq.edu.au> writes:
>
> > So the GPL is hypocrisy simply because you cannot charge for access to
> > sourcecode?
>
> Try to pay attention. I think the GPL is a bad, defective licence, but
> what I called hypocrisy was the claim that "It is free as in free
> speech, not free as in free beer." Because it IS free as in free beer.
> The licence prevents you obtaining payment for the software product (and
> NO, please don't go around the "but you can charge for support" merry go
> round again). It is arguing for the morality of something on the basis
> of a fundamental misrepresentation of its nature.
The GPL does not force itself on you. You are free not to release your
software as GPL.
>
> > Even when its the author's choice when to use it?
>
> Groan. The reason I object to GPL but not to ANY OTHER free or open
> source licence that I know of is precisely because it prevents someone
> wanting to contribute from releasing THEIR code on THEIR terms.
This is completely false. If you or I write something completely original,
from scratch, we can distribute it any way we wish. Especially if it was
written on Linux.
The problems come up when we base our development on someone else's work. If
I add two chapters to a Harry Potter, it it not a unique work and I may not
sell it. I'm sure I would have Lawyers diving out of planes to sue me.
One must always be careful (on any platform) about which libraries and
external code is used. I could easily create a program under Windows and
use a proprietary library or activeX control. At that point I would have to
license it for distribution. If I do not like the royalty requirements of
that library, I would have to design it out.
If you don't like the "royalty" requirement of GPL, i.e. freedom, design out
the GPL component.
> That is,
> it is exactly and only its tendency to make people release code on OTHER
> THAN their own preferred terms that is the problem. Suppose I get a
> beaut GPL program, but find it doesn't work exactly how I like. So I
> make lots of neat changes and now it's magnificent. My friends see it
> and say "Hey, can I have a copy?" I now have the choice of either saying
> no, or releasing MY code on terms that are not of MY liking.
But YOU modified code that was not yours. You got the benefit of someone
else's work. In other words, someone was generous to give you a program
that did most of what you wanted. You added a few things that made it
better.
Of what value are your additions without the whole? What right do you have
to build on someone else's work and call it your own?
Back to Harry Potter, suppose you rewrote a few chapters in the latest Harry
Potter book. Why should you think that you can sell the version you have
without paying JK Rowling? You can, not morally or legally.
By the same token, why should you think you should have he right to
distribute "your" additions to someone else's code under a different
license, when the license by which you acquired the original code made the
stipulation that any changes must also be GPL?
If you don't like the requirements of the GPL, don't modify GPL code. It
isn't yours.
> And most
> people, as they are usually well-meaning, kind, and generous, will
> decide to release the code with what is to them an unsatisfactory
> licence. This is my problem with it: it punishes generosity.
>
It rewards generosity. If I am generous to release code in the first place,
it protects that my generosity will not be circumvented by some jerk
modifying my code (building on my hard work) and releasing it under a
license I did not authorize.
David Steuber
> While restarting Outlook, Ron House grumbled:
>
> > As an aside, I doubt that the GPL is really sound and can do what it
> > claims, ...
>
> I dunno. RMS sounds like a pretty logical and analytical fellow.
I think there is a very good reason why the GPL has never been tested
in a court. No lawyer has come up with the sophistry to work around
it. It is just that solid.
--
One Editor to rule them all. One Editor to find them,
One Editor to bring them all and in the darkness bind them.
(do ((a 1 b) (b 1 (+ a b))) (nil a) (print a))
David Kastrup
> Linųnut <linųn...@bone.com> writes:
>
> > While restarting Outlook, Ron House grumbled:
> >
> > > As an aside, I doubt that the GPL is really sound and can do
> > > what it claims, ...
> >
> > I dunno. RMS sounds like a pretty logical and analytical fellow.
>
> I think there is a very good reason why the GPL has never been
> tested in a court. No lawyer has come up with the sophistry to work
> around it. It is just that solid.
It does not need to be solid. It could be completely braindead and it
would not matter. Because the recipient is free to completely ignore
it. Which buys him squat and then less, since the GPL only grants
additional freedoms beyond what would apply without it.
Per Abrahamsen
> Per Abrahamsen <abr...@dina.kvl.dk> wrote on Fri, 04 Jul 2003 13:35:03 +0200:
>
> [ .... ]
>
>> Currently, all the large and visible GPL'ed projects have multiple
>> people who obtain payment for developing the code, usually full-time.
>
> Really? Who all gets paid to develop Emacs (whether GNU or X)? Just
> curious.
Gerd Moellmann was paid to be full-time maintainer of Emacs. Jim
Blandy had the same job before him. Currently RMS has the job as a
volunteer (he never gets paid by the FSF), as none of the other lead
developer wanted the responsibility. But apparently the FSF has the
money to hire the right person, should he appear.
I suspect Handa also has i18n of Emacs as a big part of his job, for
most of the others I suspect Emacs maintenance is only a small part of
their day job.
I know less about XEmacs, but when I meet them at a conference years
ago two the lead developers were working full time on XEmacs on a
Japanese grant. Historically, the original development of XEmacs was
paid for by Lucid because they needed an IDE, and later a partial
rewrite was sponsored by Sun (for the same purpose).
My guess is that both projects in average have one full-time paid
developer, but both have dry periods surving on volunteer and bit-part
paid work, as well as active periods with many full-time people.
It is not something that is talked about a lot on the development
lists, so the above is to a large part speculation.
David Kastrup
> Rabid Mets fan, Per Abrahamsen, cheered ...
>
> > Gerd Moellmann was paid to be full-time maintainer of Emacs. Jim
> > Blandy had the same job before him. Currently RMS has the job as
> > a volunteer (he never gets paid by the FSF), as none of the other
> > lead developer wanted the responsibility. But apparently the FSF
> > has the money to hire the right person, should he appear.
>
> It really steams me that people like RHS and Linux work for peanuts
^^^ ^^^^^ those are people?
> all the while conveying great benefits to humanity while slimy
> shitpiles like Bill Gates, Steve Ballmer and Paul Allen gouge the
> world for with billions while at the same time hurting it with
> crappy software.
Your views are a bit lopsided. Strong ethics have always had to be
their own reward mostly. Since they are usually based on some law of
reciprocation, their gains can't be above average. If they gained you
more than that, it would mean that you would be cheating on somebody.
> it's like the cruel mean boyfriend who gets all the sex and the nice
> guy boyfriend who gets all the agony.
Well, it is his choice to be nice, isn't it?
T. Max Devlin
>a gay blade <a...@gay.blade> writes:
>
>> Rabid Mets fan, Per Abrahamsen, cheered ...
>>
>> > Gerd Moellmann was paid to be full-time maintainer of Emacs. Jim
>> > Blandy had the same job before him. Currently RMS has the job as
>> > a volunteer (he never gets paid by the FSF), as none of the other
>> > lead developer wanted the responsibility. But apparently the FSF
>> > has the money to hire the right person, should he appear.
>>
>> It really steams me that people like RHS and Linux work for peanuts
> ^^^ ^^^^^ those are people?
>> all the while conveying great benefits to humanity while slimy
>> shitpiles like Bill Gates, Steve Ballmer and Paul Allen gouge the
>> world for with billions while at the same time hurting it with
>> crappy software.
>
>Your views are a bit lopsided. Strong ethics have always had to be
>their own reward mostly. Since they are usually based on some law of
>reciprocation, their gains can't be above average.
LOL.
Your theory that ethics are based on "some law of reciprocation" is just
speculation. Common, I'll admit; hell, it is a matter which is taken on
dogmatic faith by the vast majority of practitioners in the 'soft sciences'
(psychology, sociology, etc.) and other post-modernists ranging far and wide
in other fields as well. But it is both unproven and incorrect, nevertheless.
>If they gained you
>more than that, it would mean that you would be cheating on somebody.
And we know this because you say so, right?
>> it's like the cruel mean boyfriend who gets all the sex and the nice
>> guy boyfriend who gets all the agony.
>
>Well, it is his choice to be nice, isn't it?
Was; by the time he is nice, that choice is historical.
Ron House
> Ron House <ho...@usq.edu.au> writes:
>
>
>>David Kastrup wrote:
>>
>>
>>>They can release their code on their terms all they want. But they
>>>won't release my code on their terms.
>>
>>Excellent philosophy. I agree completely.
>>
>>
>>>>Suppose I get a beaut GPL program, but find it doesn't work exactly
>>>>how I like. So I make lots of neat changes and now it's
>>>>magnificent. My friends see it and say "Hey, can I have a copy?" I
>>>>now have the choice of either saying no, or releasing MY code on
>>>>terms that are not of MY liking.
>>>
>>>It's not your code. It's the code of someone else with changes of
>>>yours.
>>
>>My code can be in files entirely of my writing, and yours ditto, but
>>if I release your files under your licence and my files under mine,
>>the GPL purports to say that I have violated something because I
>>have distributed them together.
>
>
> If both are intended to form one inseparable executable, you are
> building on my work. It is not the GPL that says anything about this,
> it is common copyright law.
That remark alone proves you know nothing about copyright law. Copyright
controls expression, not function. What something is intended to do is
completely outside the field covered by the law. If you write your code
and I write my code, and my code is intended to be used with your code,
there is nothing you can do about my code (this is why I think the GPL
is ultimately toothless). However, if I distribute the actual
executable, then yes, as it contains your code, it has to satisfy your
licence requirements.
>>You should be aware, also, that under copyright law, if I modify
>>your code with a small change, you still own the copyright, so if
>>changes are confined to your files, the whole question doesn't arise
>>under any licence.
>
> Wrong. The point is that copyright law _prohibits_ _any_
> redistribution of copies, modified or not, unless additional licensing
> is agreed upon between distributor and recipient of software.
This is (a) not contrary to what I just said, and (b) irrelevant to my
concern about what the GPL tries to make me do with MY files, not yours,
about which I am happy for you to do what you like.
>...
> Oh, the GPL can do very well what it claims because every recipient is
> completely free to ignore it and not accept its terms. But that does
> not allow you to invent different terms. And it will buy you nothing,
> since in contrast to other licences, the GPL does not try to restrict
> your rights beyond what the law provides as default, but extend them.
> If you refuse to have your rights extended, go for it, feel free.
You still don't understand that I have NO OBJECTION (indeed, I
positively affirm) your right to distribute your code under your
licence. My problem is that your licence tries to make me put your terms
on my code. Having said that for the dozenth time, I think no further
point is served by continuing to argue with brick walls.
Ron House
> If your code in "your files" relies on and requires my code in *libraries*,
> then yours is derivative of mine. Sorry; that's the way it works. Write your
> own libraries if it's a problem for you.
Baloney. Copyright concerns itself with expression, not function. If my
document is mine, all written by me, then how it is intended to operate
is not a question that copyright law concerns itself with. A combined
executable would be derivative of your code, and that is why licences
such as Mozilla can place restrictions on such executables, but I have
no problem with that at all. If people here had taken the trouble to
respond to my actual criticism of the GPL instead of waxing righteous
about things I don't criticise, a lot more (that is, some) progress
might have been made.
Ron House
>>My code can be in files entirely of my writing, and yours ditto, but if
>>I release your files under your licence and my files under mine, the GPL
>>purports to say that I have violated something because I have
>>distributed them together.
>
> By the same logic I could rewrite the last few pages of "Pavane" by
> Keith Roberts (the ending sucks) and publish it myself :-)
Still missing the point as always. My criticism of the GPL is a _moral_
one, that it tries to strongarm others to distribute their code under
the GPL or not at all, that is, to be ungenerous or else fall into line.
On the other hand, if "together" means in an executable, then there is
no doubt that the law is on the GPL's side, just as it would be on Keith
Roberts' in the above example. All I meant by "together" above, though,
was something like on the same CD or website, available to compile
together by the recipient. I don't believe the GPL is sound in that
area, but it doesn't change the moral point that I don't agree with its
intentions. Now don't come back and say "What if I put Keith Roberts'
novel on the web site?", because you know and I know that the licence
terms for the text of a novel are in themselves very different from the
GPL. There is also the question of the author of a novel's moral rights.
Whoever
> Rob Thorpe wrote:
> > Ron House <ho...@usq.edu.au> wrote in message news:<3F0927B7...@usq.edu.au>...
>
> >>My code can be in files entirely of my writing, and yours ditto, but if
> >>I release your files under your licence and my files under mine, the GPL
> >>purports to say that I have violated something because I have
> >>distributed them together.
> >
> > By the same logic I could rewrite the last few pages of "Pavane" by
> > Keith Roberts (the ending sucks) and publish it myself :-)
>
> Still missing the point as always. My criticism of the GPL is a _moral_
> one, that it tries to strongarm others to distribute their code under
> the GPL or not at all, that is, to be ungenerous or else fall into line.
Not at all: you have choices: use GPL source code and re-distribute under
the GPL, use GPL source code and don't re-distribute at all or don't use
GPL code at all. There is no "strongarm-ing". Compared to CSS where you
can't re-distribute at all, the GPL is very open.
> On the other hand, if "together" means in an executable, then there is
> no doubt that the law is on the GPL's side, just as it would be on Keith
> Roberts' in the above example. All I meant by "together" above, though,
> was something like on the same CD or website, available to compile
> together by the recipient. I don't believe the GPL is sound in that
> area, but it doesn't change the moral point that I don't agree with its
> intentions.
In this case, this would be mere "aggregation" -- the GPL does not impose
additional restrictions on software that is aggregated with GPL software.
That is what SuSE does: aggregates their own s/w with GPL software to
provide a distribution on which SuSE restricts re-distribution of the
non-GPL parts.
David Kastrup
> T. Max Devlin wrote:
>
> > If your code in "your files" relies on and requires my code in
> > *libraries*, then yours is derivative of mine. Sorry; that's the
> > way it works. Write your own libraries if it's a problem for you.
>
> Baloney. Copyright concerns itself with expression, not function. If
> my document is mine, all written by me, then how it is intended to
> operate is not a question that copyright law concerns itself with.
You better check the scope of derived works according to American
copyright law. Actual inclusion is not required.
Peter Köhlmann
Your "actual criticism" of the GPL are bullshit, and that is the reason
you get those answers you don't like
If you had sat down for a moment and just read the GPL, you would have
known that. Instead you try to tell us this utter garbage
--
Microsoft's Guide To System Design:
It could be worse, but it'll take time.
T. Max Devlin
>> Ron House <ho...@usq.edu.au> writes:
>>
>>>David Kastrup wrote:
>>>
>>>>They can release their code on their terms all they want. But they
>>>>won't release my code on their terms.
>>>
>>>Excellent philosophy. I agree completely.
>>>
>>>>>Suppose I get a beaut GPL program, but find it doesn't work exactly
>>>>>how I like. So I make lots of neat changes and now it's
>>>>>magnificent. My friends see it and say "Hey, can I have a copy?" I
>>>>>now have the choice of either saying no, or releasing MY code on
>>>>>terms that are not of MY liking.
>>>>
>>>>It's not your code. It's the code of someone else with changes of
>>>>yours.
>>>
>>>My code can be in files entirely of my writing, and yours ditto, but
>>>if I release your files under your licence and my files under mine,
>>>the GPL purports to say that I have violated something because I
>>>have distributed them together.
>>
>> If both are intended to form one inseparable executable, you are
>> building on my work. It is not the GPL that says anything about this,
>> it is common copyright law.
>
>That remark alone proves you know nothing about copyright law.
What makes you think that?
>Copyright
>controls expression, not function.
And if you base your function on my expression, you're building on my work.
Get it?
>What something is intended to do is
>completely outside the field covered by the law. If you write your code
>and I write my code, and my code is intended to be used with your code,
>there is nothing you can do about my code (this is why I think the GPL
>is ultimately toothless).
Depends on what you mean by "intended to" to be perfectly honest. If by that
you mean that only by using a GPL library will your program work, then your
program is derivative of the library.
It isn't uncommon for programmers to misunderstand this. Lord knows you're
not the first to get confused, nor the first to complain. Nevertheless, you
are mistaken in your understanding.
>However, if I distribute the actual
>executable, then yes, as it contains your code, it has to satisfy your
>licence requirements.
If you code requires my code to be built, then it is derived from my code,
because that is how code works. Deal with it.
>>>You should be aware, also, that under copyright law, if I modify
>>>your code with a small change, you still own the copyright, so if
>>>changes are confined to your files, the whole question doesn't arise
>>>under any licence.
>>
>> Wrong. The point is that copyright law _prohibits_ _any_
>> redistribution of copies, modified or not, unless additional licensing
>> is agreed upon between distributor and recipient of software.
>
>This is (a) not contrary to what I just said, and (b) irrelevant to my
>concern about what the GPL tries to make me do with MY files, not yours,
>about which I am happy for you to do what you like.
They're only YOUR files if you don't base them (make them require by the way
you designed them) on SOMEONE ELSE'S files.
> >...
>> Oh, the GPL can do very well what it claims because every recipient is
>> completely free to ignore it and not accept its terms. But that does
>> not allow you to invent different terms. And it will buy you nothing,
>> since in contrast to other licences, the GPL does not try to restrict
>> your rights beyond what the law provides as default, but extend them.
>> If you refuse to have your rights extended, go for it, feel free.
>
>You still don't understand that I have NO OBJECTION (indeed, I
>positively affirm) your right to distribute your code under your
>licence.
As long as you're allowed to use it if it is in a library, right?
>My problem is that your licence tries to make me put your terms
>on my code.
It is unfortunate that what you consider "your code" is not always "your
code". Intellectual property law is filled with examples of this kind of
problem.
Still, you lose. No, it isn't your code.
>Having said that for the dozenth time, I think no further
>point is served by continuing to argue with brick walls.
You will learn, or you will not, that is true. But I'll keep trying to
explain it as long as you are willing to at least pretend to learn.
Rob Thorpe
> Rob Thorpe wrote:
> > Ron House <ho...@usq.edu.au> wrote in message news:<3F0927B7...@usq.edu.au>...
>
> >>My code can be in files entirely of my writing, and yours ditto, but if
> >>I release your files under your licence and my files under mine, the GPL
> >>purports to say that I have violated something because I have
> >>distributed them together.
> >
> > By the same logic I could rewrite the last few pages of "Pavane" by
> > Keith Roberts (the ending sucks) and publish it myself :-)
>
> Still missing the point as always. My criticism of the GPL is a _moral_
> one
In my view the same applies. Changing a peice of software or adding
to it is the morally similar to changing the last few pages of a
novel. It produces a derivative work.
Paul Jarc
>>> if I release your files under your licence and my files under
>>> mine, the GPL purports to say that I have violated something
>>> because I have distributed them together.
> All I meant by "together" above, though, was something like on the
> same CD or website, available to compile together by the recipient.
The GPL makes no such restriction. Section 2:
# In addition, mere aggregation of another work not based on the
# Program with the Program (or with a work based on the Program) on
# a volume of a storage or distribution medium does not bring the
# other work under the scope of this License.
paul
Albert van der Horst
> > sourcecode?
>
>Try to pay attention. I think the GPL is a bad, defective licence, but
>what I called hypocrisy was the claim that "It is free as in free
>speech, not free as in free beer." Because it IS free as in free beer.
>The licence prevents you obtaining payment for the software product (and
>NO, please don't go around the "but you can charge for support" merry go
>round again). It is arguing for the morality of something on the basis
>of a fundamental misrepresentation of its nature.
This is absolutely and totally incorrect.
I made a special version of the GNU c-compiler and happily charged
all my hours. The customer is of course obliged to distribute the
source code of the customized c-compiler, if they distribute the
compiler. They don't. They distribute embedded system, with programs
compiled by the customized c-compiler.
By the way, this solved a great big problem for them. They didn't
need to hand-change assembler libraries. I tell this, not for you,
but for the edification of those who listen in.
This is the way open source works in the industry, where the real
work is done and the real wealth is produced.
>Ron House ho...@usq.edu.au
> http://www.sci.usq.edu.au/staff/house
>
--
Albert van der Horst,Oranjestr 8,3511 RA UTRECHT,THE NETHERLANDS
One man-hour to invent,
One man-week to implement,
One lawyer-year to patent.
Albert van der Horst
>As an aside, I doubt that the GPL is really sound and can do what it
>claims, but that just makes it a bungling villain, not a good guy.
I don't know what you think GPL claims. I just know that it has done
what it was intended for, empower those who do the actual work in the
industry. The emergency of Linux is directly related to the inspiration
that those who can program, get from the certainty that their intellectual
property disseminate, yet cannot be stolen.
>--
>Ron House ho...@usq.edu.au
> http://www.sci.usq.edu.au/staff/house
>
Lee Wei Shun
> One man-hour to invent,
> One man-week to implement,
> One lawyer-year to patent.
Borrowing your sig, OK? Want to send it to a patent lawyer (not to patent
it!) friend of mine.
Regards,
WS
--
Change to leews to mail.
Linux user #61399
The beginning of the
end
soft-eng
e.g. employees of organizations like RedHat, FSF etc. Even
early on, companies like Apple paid money to FSF because
they viewed themselves as hardware companies, and it was in
their best interests to see cheap/free (as in beer) software.
But the number of people who make money off free software
is a very small number. The problem is that if
all the software work in the world was of this type, it
would support only a very small number of developers.
As a result, not too many students would want to study software.
As a result, many Universities would not teach software. As
a result, research work would dwindle down. As a result,
most progress in software would halt. While not catastrophic,
this doesn't appear to be a very good thing for human
future in general.
In that sense, Free Software developers are basically
cutting off the branch that they are sitting on, they
are working on closing down the very institutions
that taught them how to write software. (I don't
mean employees of RedHat, who are simply using
their talents for something their employers
want. But I do mean people who contribute their
free time and effort to free-software rather than
something either more charitable or more beneficial.)
That's why it appears FSF/GNU is ultimately destructive,
not constructive.
Kin Cho
> As a result, many Universities would not teach software. As
> a result, research work would dwindle down. As a result,
> most progress in software would halt. While not catastrophic,
> this doesn't appear to be a very good thing for human
> future in general.
I think formal computer science instruction is becoming less relevant. Using
nothing but google and *ux or *bsd sources, I imagine a lot of people from all
over the world could study software on their own very well. As a result,
software that people care about will continue to progress.
-kin
D. C. Sessions
> But the number of people who make money off free software
> is a very small number.
Tell that to $EMPLOYER's embedded-systems people.
We recently made a very noisy press announcement
on the subject.
For all practical purposes, /all/ of our software
people are working on /software/ /libre./
> The problem is that if
> all the software work in the world was of this type, it
> would support only a very small number of developers.
Like about 90% of the ones currently employed?
Less than 10% of programmers work on software that
is sold or licensed.
> As a result, not too many students would want to study software.
"Not too many" being at least 90% of the current
number, assuming that improved utility and less
cream-skimming by suppliers doesn't /increase/
the amount spent on bespoken software and other
custom applications.
> As a result, many Universities would not teach software. As
> a result, research work would dwindle down. As a result,
> most progress in software would halt. While not catastrophic,
> this doesn't appear to be a very good thing for human
> future in general.
Congratulations -- you've just explaine why scientific
progress practically ceased when publication took the
place of closed Guild secrets.
--
| Microsoft: "A reputation for releasing inferior software will make |
| it more difficult for a software vendor to induce customers to pay |
| for new products or new versions of existing products." |
end
T. Max Devlin
>There are a number of people making money off free software,
>e.g. employees of organizations like RedHat, FSF etc. Even
>early on, companies like Apple paid money to FSF because
>they viewed themselves as hardware companies, and it was in
>their best interests to see cheap/free (as in beer) software.
>
>But the number of people who make money off free software
>is a very small number. The problem is that if
>all the software work in the world was of this type, it
>would support only a very small number of developers.
The problem seems to be you are assuming that because the number of
professional free software developers is small in proportion to all non-free
software developers, it would remain small if there weren't any non-free
software developers.
>As a result, not too many students would want to study software.
Programmers (at least gifted ones) would probably be paid much more money if
all software work in the world was of this type; it is only corporations that
hire software developers who make money profiteering on the programmer's work
as trade secrets that would make less money, if all software work in the world
was of this type.
>As a result, many Universities would not teach software. [...]
And then eventually the world collapses into a black hole, and....
soft-eng
> In <9fa75d42.03073...@posting.google.com>, soft-eng wrote:
>
> > But the number of people who make money off free software
> > is a very small number.
>
> Tell that to $EMPLOYER's embedded-systems people.
> We recently made a very noisy press announcement
> on the subject.
>
> For all practical purposes, /all/ of our software
> people are working on /software/ /libre./
So your company has a lot of software people
but no work for them. What's your point?
> > The problem is that if
> > all the software work in the world was of this type, it
> > would support only a very small number of developers.
>
> Like about 90% of the ones currently employed?
> Less than 10% of programmers work on software that
> is sold or licensed.
Surely not on free software?
Are you claiming in-house devlopment accounts for
90% of all work? While in the current economy (partially
generated by the free software ethics) it is quite likely,
I don't believe that's been the case during most of software
development's history.
> > As a result, not too many students would want to study software.
>
> "Not too many" being at least 90% of the current
> number, assuming that improved utility and less
> cream-skimming by suppliers doesn't /increase/
> the amount spent on bespoken software and other
> custom applications.
"Cream-skimming by suppliers" works in all professions, but cutting
your nose to spite your face doesn't sound like a good idea.
> > As a result, many Universities would not teach software. As
> > a result, research work would dwindle down. As a result,
> > most progress in software would halt. While not catastrophic,
> > this doesn't appear to be a very good thing for human
> > future in general.
>
> Congratulations -- you've just explaine why scientific
> progress practically ceased when publication took the
> place of closed Guild secrets.
Wrong comparison. Software progress is not restricted
to free-software, and in fact software methodologies
progressed quite fine before FSF was born. All that
got added into free-software are low-level techniques
that are of little relevance to overall progress.
For actual progress, it's sufficient that people make
available their research in the form of algorithms,
at a reasonably high level of communication.
Traditionally, this has worked fine. Putting it
down into low-level C code is a requirement that's
very recent, that's come into being because of
free-software, and that adds nothing to the
actual advance.