ANNOUNCE: GNU TeX 0.80 (beta)
Robert Kiesling
GNU TeX, version 0.80 (beta) is available for testing, and should be
stable and complete enough for use. It is an implementation of TeX
and LaTeX, and is based on the standard Web2c source distribution.
GNU TeX follows the TeX Directory Standard (TDS) and includes these
packages:
TeX and LaTeX (Web2c)
AMSTeX and AMSLaTeX
PDFTeX
Metafont and Metapost
Texinfo utilities
Makeindex
Xdvi
Dvips
Dvilj
Xpdf
Gsftopk
Computer Modern Fonts, Bitmapped and Type 1
EC Bitmapped Fonts
Bitstream Charter Type 1 Fonts
URW Nimbus, Antigua, and Grotesk Type 1 Fonts
Much of the work before Version 1.0 involves writing documentation and
help texts, coding virtual fonts, and general clean-up of the TEXMF
library and the source code for the binaries.
For later releases, we would like to include free replacements for
some commonly used non-free TeX and LaTeX packages. (We cannot
include those packages in this free software collection.) If you are
an expert TeX hacker and you are seriously interested in writing a
replacement for a non-free package, please send me mail.
The distribution is archived it two Gzipped tar files, totalling
approximately 16 MB of disk space. The distribution requires about
80MB of disk space to build and install the binaries.
Please mail me for details of how to obtain the GNU TeX source and
library archives. Binaries for Intel x86 GNU/Linux and Sparc/Solaris
are planned for version 1.0. Binaries compiled for other hardware
platforms will become part of the distribution as they become
available
Robert Kiesling
--
Robert Kiesling
kies...@ix.netcom.com
Reinhard Zierke
> GNU TeX, version 0.80 (beta) is available for testing, and should be
> stable and complete enough for use. It is an implementation of TeX
> and LaTeX, and is based on the standard Web2c source distribution.
What are the main differences to teTeX 0.9 ?
Reinhard
--
Reinhard Zierke Universitaet Hamburg, FB Informatik
zie...@informatik.uni-hamburg.de Vogt-Koelln-Strasse 30, D-22527 Hamburg
postm...@informatik.uni-hamburg.de Tel.: (040) 5494-2295/2276 Fax: -2241
Herman Bruyninckx
> GNU TeX, version 0.80 (beta) is available for testing, and should be
> stable and complete enough for use. It is an implementation of TeX
> and LaTeX, and is based on the standard Web2c source distribution.
> GNU TeX follows the TeX Directory Standard (TDS) and includes these
> packages:
Judging from the appearance of the word `GNU' in the package name, I guess
this is a open source effort to produce high-quality software. This is of
course a laubible initiative, but I thought Tex and derivatives were *the*
example of high-quality open software. So what exactly is the reason to add
`another' distribution? What non-free software are you thinking of?
Herman
--
Herman.B...@mech.kuleuven.ac.be (Ph.D.) Fax: +32-(0)16-32 29 87
Dept. Mechanical Eng., Div. PMA, Katholieke Universiteit Leuven, Belgium
<http://www.mech.kuleuven.ac.be/~bruyninc>
Rolf Marvin B|e Lindgren
| GNU TeX, version 0.80 (beta) is available for testing, and should be
| stable and complete enough for use.
is it just me, or isn't there a word about where to get it?
--
Rolf Lindgren
ro...@ask.uio.no
David Kastrup
> On Sat, 16 Jan 1999, Robert Kiesling wrote:
>
> > GNU TeX, version 0.80 (beta) is available for testing, and should be
> > stable and complete enough for use. It is an implementation of TeX
> > and LaTeX, and is based on the standard Web2c source distribution.
> > GNU TeX follows the TeX Directory Standard (TDS) and includes these
> > packages:
>
> Judging from the appearance of the word `GNU' in the package name, I guess
> this is a open source effort to produce high-quality software. This is of
> course a laubible initiative, but I thought Tex and derivatives were *the*
> example of high-quality open software. So what exactly is the reason to add
> `another' distribution? What non-free software are you thinking of?
Just look at a few licenses, like that of foilTeX (which may not be
used for anything serious, according to its license).
teTeX contains a lot of packages, some of which cannot really be put
onto a CD and be sold for profit, according to their license.
Of course, it would have been a better solution to try to co-work with
Thomas Esser, and to try to carefully split the non-free from the free
packages.
One, of course, also has to note that the development organization of
teTeX has, despite its high quality, has not made use of human
ressources optimally in the more recent past. For example, there has
been no official release of teTeX for about 2 years (although crucial
components of it, like web2c, have had several releases in the
meantime).
One will see whether a GNU TeX distribution will be able to better
organize the necessary ressources. While I even think that Thomas
Esser would not be too unhappy if the principal responsibility for a
Unix TeX distribution resting on his shoulders was levied somewhat, I
guess that it would have been much saner to try to arrange for smooth
change instead of a separate effort. One reason, of course, being
that the TeX crowd does not have as widespread high quality people
floating around as, say, Linux has.
--
David Kastrup Phone: +49-234-700-5570
Email: d...@neuroinformatik.ruhr-uni-bochum.de Fax: +49-234-709-4209
Institut für Neuroinformatik, Universitätsstr. 150, 44780 Bochum, Germany
Herman Bruyninckx
> Herman Bruyninckx <bruy...@mail.mech.kuleuven.ac.be> writes:
>
> > On Sat, 16 Jan 1999, Robert Kiesling wrote:
> >
> > > GNU TeX, version 0.80 (beta) is available for testing, and should be
> > > stable and complete enough for use. It is an implementation of TeX
> > > and LaTeX, and is based on the standard Web2c source distribution.
> > > GNU TeX follows the TeX Directory Standard (TDS) and includes these
> > > packages:
> >
> > Judging from the appearance of the word `GNU' in the package name, I guess
> > this is a open source effort to produce high-quality software. This is of
> > course a laubible initiative, but I thought Tex and derivatives were *the*
> > example of high-quality open software. So what exactly is the reason to add
> > `another' distribution? What non-free software are you thinking of?
>
> Just look at a few licenses, like that of foilTeX (which may not be
> used for anything serious, according to its license).
>
> teTeX contains a lot of packages, some of which cannot really be put
> onto a CD and be sold for profit, according to their license.
>
[...]
Maybe it's a good idea then that the GnuTex people use this newsgroup to
regularly post a list of `to do' things, so that the TeX community can
help. Now, we just don't know what GnuTeX needs.
Matthew Vernon
> [ Robert Kiesling
>
> | GNU TeX, version 0.80 (beta) is available for testing, and should be
> | stable and complete enough for use.
>
> is it just me, or isn't there a word about where to get it?
www.gnu.org/ftp.gnu.org/alpha.gnu.org and mirrors are all probably
good starts (though I haven't looked for it)
Matthew
--
Elen Sila Lumenn' Omentielvo
Matthew Vernon, Steward of the Cambridge Tolkien Society
Selwyn College Computer support
http://pick.sel.cam.ac.uk http://www.cam.ac.uk/CambUniv/Societies/tolkien/
Robert Kiesling
Reinhard Zierke <zie...@dante.de> wrote:
>Robert Kiesling <kies...@ix.netcom.com> wrote:
>
>> GNU TeX, version 0.80 (beta) is available for testing, and should be
>> and LaTeX, and is based on the standard Web2c source distribution.
>
Right now:
1. Completely free source code (we would like a TeX distribution that can be
included with GNU software).
2. Easier installation, smaller disk space requirements.
3. Free documentation (in progress).
The development is being done with the goal of merging once again with
teTeX or TeX Live in the future. I've spoken with the maintainers of
those distributions, and have told that this distribution will provide
a development track that is closer to the intent and spirit of GNU
software, which, due to the popularity of free operating systems, has
a much greater installed base than in the past.
--
Robert Kiesling
kies...@ix.netcom.com
Robert Kiesling
Rolf Marvin B|e Lindgren <ro...@morgoth.uio.no> wrote:
>[ Robert Kiesling
>
>| GNU TeX, version 0.80 (beta) is available for testing, and should be
>| stable and complete enough for use.
>
I wasn't sure how public I should make the announcement, due to the
way in which the GNU archives are arranged. But now, I don't think
there's much harm in saying that it's archived at:
as:
gnutex-0.80.INSTALL
gnutex-0.80.tar.gz
gnutexmf-0.80.tar.gz
If you have trouble accessing the files, let me know.
--
Robert Kiesling
kies...@ix.netcom.com
Robert Kiesling
Herman Bruyninckx <bruy...@mail.mech.kuleuven.ac.be> wrote:
>On Sat, 16 Jan 1999, Robert Kiesling wrote:
>
>> and LaTeX, and is based on the standard Web2c source distribution.
>> packages:
>
>Judging from the appearance of the word `GNU' in the package name, I guess
>this is a open source effort to produce high-quality software. This is of
>course a laubible initiative, but I thought Tex and derivatives were *the*
>example of high-quality open software. So what exactly is the reason to add
>`another' distribution? What non-free software are you thinking of?
There are certainly commercial TeX distributions, not only on unices,
but on MS and Apple platforms. More than that, however, many of the
macro packages that come with the TeX distributions are non-free in
their licensing. Another issue is fonts. That area probably will
remain contentious--all you need to do is read comp.fonts for a while
to be convinced--and with that consideration in mind, the distribution
includes only free fonts.
None of this, of course, keeps individual users from adding the
TeX or LaTeX packages that are available on CTAN, or any commercial
add-ons.
--
Robert Kiesling
kies...@ix.netcom.com
Matthew Vernon
For information, your nearest sunsite mirror carries alpha in
/public/Mirrors/alpha.gnu.org
Matthew
kies...@ix.netcom.com (Robert Kiesling) writes:
> In article <lbz67a7...@morgoth.uio.no>,
> Rolf Marvin B|e Lindgren <ro...@morgoth.uio.no> wrote:
> >[ Robert Kiesling
> >
> >| GNU TeX, version 0.80 (beta) is available for testing, and should be
> >| stable and complete enough for use.
> >
> >is it just me, or isn't there a word about where to get it?
>
> I wasn't sure how public I should make the announcement, due to the
> way in which the GNU archives are arranged. But now, I don't think
> there's much harm in saying that it's archived at:
>
> ftp://alpha.gnu.org/gnu
>
> as:
>
> gnutex-0.80.INSTALL
> gnutex-0.80.tar.gz
> gnutexmf-0.80.tar.gz
>
> If you have trouble accessing the files, let me know.
--
Robert Kiesling
Herman Bruyninckx <bruy...@mail.mech.kuleuven.ac.be> wrote:
>
>> Herman Bruyninckx <bruy...@mail.mech.kuleuven.ac.be> writes:
>>
>> >
>> > > and LaTeX, and is based on the standard Web2c source distribution.
>> > > GNU TeX follows the TeX Directory Standard (TDS) and includes these
>> > > packages:
>> >
>> > Judging from the appearance of the word `GNU' in the package name, I guess
>> > this is a open source effort to produce high-quality software. This is of
>> > course a laubible initiative, but I thought Tex and derivatives were *the*
>> > example of high-quality open software. So what exactly is the reason to add
>> > `another' distribution? What non-free software are you thinking of?
>>
>> used for anything serious, according to its license).
>>
>> teTeX contains a lot of packages, some of which cannot really be put
>> onto a CD and be sold for profit, according to their license.
>>
>[...]
>
>Maybe it's a good idea then that the GnuTex people use this newsgroup to
>regularly post a list of `to do' things, so that the TeX community can
>help. Now, we just don't know what GnuTeX needs.
Just about everything that is not part of LaTeX2e proper. I've
written some macros for font encoding, but the utilities, and
everything in the macros/contrib directory of CTAN, is not in the
distribution.
I don't have a list of specific goals right now. I first would like
to hear the responses of other people who use the distribution.
--
Robert Kiesling
kies...@ix.netcom.com
Donald Arseneau
>macro packages that come with the TeX distributions are non-free in
>their licensing.
I'll point out, for people not so familiar with GNU and FSF that
their use of "free" is, err, somewhat non-standard english.
http://www.gnu.org probably has an extensive definition.
Donald Arseneau as...@triumf.ca
Jay Belanger
> kies...@ix.netcom.com writes...
> >macro packages that come with the TeX distributions are non-free in
> >their licensing.
>
> I'll point out, for people not so familiar with GNU and FSF that
> their use of "free" is, err, somewhat non-standard english.
> http://www.gnu.org probably has an extensive definition.
Actually, it is standard english, the problem is that the adjective
"free" means different things in different situations. The first
meaning that probably comes to mind is "at no cost", but that isn't
always the correct meaning. In some situations it clearly means
"unconstrained" (which is the first meaning in my dictionary). In
some situations, the word "free" needs further explanation; e.g, my dog
was given to me at no cost... is the dog free? Software seems ot fall
into this third category, and so by all means see the gnu website for
more explanation.
Jay
Timothy Murphy
>One, of course, also has to note that the development organization of
>teTeX has, despite its high quality, has not made use of human
>ressources optimally in the more recent past.
I don't know what this means.
teTeX seems to me to be very well organised,
certainly much better than most GNU products.
>For example, there has
>been no official release of teTeX for about 2 years (although crucial
>components of it, like web2c, have had several releases in the
>meantime).
But there have been innumerable (say, weekly) unofficial releases of teTeX
for those interested.
If TE prefers to be ultra-conservative about official releases,
that is his prerogative.
It has nothing to do with the quality of the product.
--
Timothy Murphy
e-mail: t...@maths.tcd.ie
tel: +353-1-2842366
s-mail: School of Mathematics, Trinity College, Dublin 2, Ireland
David Kastrup
> David Kastrup <d...@mailhost.neuroinformatik.ruhr-uni-bochum.de> writes:
>
> >One, of course, also has to note that the development organization of
> >teTeX has, despite its high quality, has not made use of human
> >ressources optimally in the more recent past.
>
> I don't know what this means.
> teTeX seems to me to be very well organised,
> certainly much better than most GNU products.
>
> >For example, there has
> >been no official release of teTeX for about 2 years (although crucial
> >components of it, like web2c, have had several releases in the
> >meantime).
>
> But there have been innumerable (say, weekly) unofficial releases of teTeX
> for those interested.
> If TE prefers to be ultra-conservative about official releases,
> that is his prerogative.
> It has nothing to do with the quality of the product.
I was not denigrating the quality of the product (although one problem
is that there is not a fixed product to speak of. teTeX 0.9 by now
means anything from a set of different versions distributed over 2
years). Still, the management of the release seems to be a single
person effort, mainly (there is no CVS site or other means of sharing
work) and only work on components (like web2c) is delegated. Yet
Thomas Esser is by now in a personal situation (job, married, etc.)
which does not leave him enough time to manage the work needed for
making a proper release (basically, installation scripts for Windows
were what he said was the last missing thing for a release. And he is
just managing to keep up with putting updated components into teTeX,
but does not seem to have the time to do the things he wants to, as
this has been the state for about a year at last).
So it seems that the management of teTeX is putting too many
responsibilities on Thomas Esser and is trying to delegate too little.
While this ensures that what *does* get out is of the high quality
standards TE demands from himself, it does also seem to mean that it
never gets to the point of being released.
Timothy Murphy
>Thomas Esser is by now in a personal situation (job, married, etc.)
>which does not leave him enough time to manage the work needed for
>making a proper release
Surely marriage leaves one more time to do what one wants, not less.
>basically, installation scripts for Windows
>were what he said was the last missing thing for a release.
It was my impression that TE did not do any Windows stuff himself,
but just facilitated those who did.
>So it seems that the management of teTeX is putting too many
>responsibilities on Thomas Esser and is trying to delegate too little.
Surely this is up to TE to decide.
I'm sure he would ask for help if he needed it.
>While this ensures that what *does* get out is of the high quality
>standards TE demands from himself, it does also seem to mean that it
>never gets to the point of being released.
The question is whether gnuTeX is likely to assist.
I shall be very much surprised (flabberghasted, in fact)
if it approaches teTeX in quality.
Will (or does) it use kpathsea?
If not, do we all have to re-arrange our TeX files?
If it does, will the gnuTeX "developers" have the same mutually assisting
relationship with the web2c person as TE seems to have?
As I see it, TE is basically an implementation of web2c
which has become pre-eminent solely because it is very good.
Robert Kiesling
Timothy Murphy <t...@maths.tcd.ie> wrote:
[text deleted]
>The question is whether gnuTeX is likely to assist.
>I shall be very much surprised (flabberghasted, in fact)
>if it approaches teTeX in quality.
>Will (or does) it use kpathsea?
It does.
>If not, do we all have to re-arrange our TeX files?
GNU TeX uses the TeX Directory Standard.
>If it does, will the gnuTeX "developers" have the same mutually assisting
>relationship with the web2c person as TE seems to have?
The author(s) of Web2c and texk have been active in the GNU TeX
project from the beginning and have contributed many useful
suggestions for its development.
--
Robert Kiesling
kies...@ix.netcom.com
Patrick TJ McPhee
Jay Belanger <j...@vh213601.truman.edu> wrote:
% as...@reg.triumf.ca (Donald Arseneau) writes:
% > I'll point out, for people not so familiar with GNU and FSF that
% > their use of "free" is, err, somewhat non-standard english.
% > http://www.gnu.org probably has an extensive definition.
%
% Actually, it is standard english, the problem is that the adjective
% "free" means different things in different situations. The first
% meaning that probably comes to mind is "at no cost", but that isn't
% always the correct meaning. In some situations it clearly means
% "unconstrained" (which is the first meaning in my dictionary). In
% some situations, the word "free" needs further explanation; e.g, my dog
For instance, when not using a definition from standard English. I recall
the fiasco surrounding ispell when FSF decided it wasn't free in just the
way they wanted it to be.
--
Patrick TJ McPhee
East York Canada
pt...@interlog.com
Louis Vosloo
One point of reference: the copyleft text is substantially longer
than the typical EULA (and harder to understand).
Jay Belanger
> Jay Belanger <j...@vh213601.truman.edu> wrote:
[snip]
> % In some situations, the word "free" needs further explanation; e.g, my dog
>
> For instance, when not using a definition from standard English. I recall
> the fiasco surrounding ispell when FSF decided it wasn't free in just the
> way they wanted it to be.
Or sometimes even when using a definition from standard English. Such
as "free". I've heard it said that one of the strenghths and
weaknesses of languages is ambiguity -- at any rate, it is a fact of
life that some words have more than one meaning. In the case of
ambiguity, it makes sense to explain what is meant.
Jay
David Carlisle
This seems like a most unhelpful and misleading development.
After many years of confusing anarchy, at last tetex and texlive
distributions have provided a real chance to have a consistent
unix tex setup. Introducing a new distribution at this time can have
no practical benefits.
There are at most two or three packages in common use that have
license conditions that would be likely to cause problems with gnu
distributions, it would make more sense to just negotiate with the
authors of those packages to try to come up with a minor modification
to tetex if that is required.
Calling it a gnu TeX distribution gives the false impression that it
is all under gnu license. Most of the TeX distribution is freely
distributed but NOT under GPL.
TeX is not under GPL
Computer Modern fonts are not under GPL
LaTeX is not under GPL
The majority of contributed packages are not under GPL.
The kpathsea path searching library which makes up an important part
of the unix tex distribution is under GPL, but that is by no means
the whole of the distribution.
Until these points are clarified, please do not include my packages
in the distribution.
David Carlisle
David Kastrup
> Calling it a gnu TeX distribution gives the false impression that it
> is all under gnu license. Most of the TeX distribution is freely
> distributed but NOT under GPL.
A lot of a typical GNU system is not under the GPL. To be considered
a component of a GNU system, software must be freely redistributable
and changeable also in commercial settings. That's about all.
> TeX is not under GPL
But you could GPL an arbitrary version of it, as it is in the Public
Domain.
> Computer Modern fonts are not under GPL
Same here.
> LaTeX is not under GPL
It *is* freely redistributable.
> The majority of contributed packages are not under GPL.
Some of them are *not* freely distributable. Those would be missing
from GNU TeX, presumably. Do you have any idea how many packages
(like foilTeX) are distributed illegally on Linux CDs? Quite a few
state that they may not be distributed for money. The market volume
here are millions of copies. About every Linux distributor currently
could be sued out of his wits just for distributing teTeX. There was
even a time when the DC fonts were not allowed to be distributed in
that way, but few poeple cared.
A GNU distribution would not include such software, so would be rather
safe to distribute.
Of course, splitting teTeX into freely redistributable parts and
others would also have been a choice.
> The kpathsea path searching library which makes up an important part
> of the unix tex distribution is under GPL, but that is by no means
> the whole of the distribution.
It does not need to be. The GNU Hurd, for example, uses a non-GPLed
Microkernel.
> Until these points are clarified, please do not include my packages
> in the distribution.
If you refuse to let your packages be distributed in certain freely
redistributable distributions, you have to put them under a less free
license.
Of course, you can *ask* for people not to include those packages for
which you have chosen a free license, but that's against the spirit of
the license in the first place.
Still, it would have been nice of one would have heard any attempts of
coordination with Thomas Esser before doing this separate effort.
David Carlisle
> > in the distribution.
>
> If you refuse to let your packages be distributed in certain freely
> redistributable distributions, you have to put them under a less free
> license.
>
No. If the distribution is modified so that it meets the condtitions in
latex's legal.txt (specifically that it includes all the latex sources
if it includes any) then I am sure that anyone has a full right to make
a distribution. However rights are one thing, I can still _ask_ that my
files are not included (temporarily) until what I see as some potential
problems are sorted out.
> Still, it would have been nice of one would have heard any attempts of
> coordination with Thomas Esser before doing this separate effort.
quite.
David
Robin Fairbairns
>David Carlisle <da...@dcarlisle.demon.co.uk> writes:
>> Calling it a gnu TeX distribution gives the false impression that it
>> is all under gnu license. Most of the TeX distribution is freely
>> distributed but NOT under GPL.
>
>A lot of a typical GNU system is not under the GPL. To be considered
>a component of a GNU system, software must be freely redistributable
>and changeable also in commercial settings. That's about all.
but tex, latex, etc., aren't changeable in any settings whatever. (at
least, any settings where they continue to be called by their original
names.)
>> TeX is not under GPL
>
>But you could GPL an arbitrary version of it, as it is in the Public
>Domain.
so long as you didn't imply that the result was a tex implementation.
>> Computer Modern fonts are not under GPL
>
>Same here.
>
>> LaTeX is not under GPL
>
>It *is* freely redistributable.
but not modifiable under its own name.
>> The majority of contributed packages are not under GPL.
>
>Some of them are *not* freely distributable. Those would be missing
>from GNU TeX, presumably. Do you have any idea how many packages
>(like foilTeX) are distributed illegally on Linux CDs?
it's clear that linux cds can't be relied on for satisfactory versions
of tex (vide the business a while back where one could buy such cds
with corrupt cm fonts on them). anyone who assumes otherwise is
already making a considerable leap in the dark.
>Quite a few
>state that they may not be distributed for money. The market volume
>here are millions of copies. About every Linux distributor currently
>could be sued out of his wits just for distributing teTeX. There was
>even a time when the DC fonts were not allowed to be distributed in
>that way, but few poeple cared.
>
>A GNU distribution would not include such software, so would be rather
>safe to distribute.
>
>Of course, splitting teTeX into freely redistributable parts and
>others would also have been a choice.
thomas should do a bunch of work because someone else wants to make
money from it? i boggle.
>> The kpathsea path searching library which makes up an important part
>> of the unix tex distribution is under GPL, but that is by no means
>> the whole of the distribution.
>
>It does not need to be. The GNU Hurd, for example, uses a non-GPLed
>Microkernel.
you mean, people are still banging away at that? i boggle.
>> Until these points are clarified, please do not include my packages
>> in the distribution.
>
>If you refuse to let your packages be distributed in certain freely
>redistributable distributions, you have to put them under a less free
>license.
anything of mine that claims to offer support, does so in the case of
*unmodified* versions of the package. aiui, unless i take special
steps to the contrary, my stuff is now going to appear on gnu archive
sites, which will encourage people to assume they can hack around with
it, redistribute modified versions, and so on. i don't like this.
why can't this gnu crowd bring itself to do what sebastian rahtz has
done re. the tex live distributions, viz., to contact package authors
for their approval, rather than to suggest that package authors need
to contact the crowd?
>Of course, you can *ask* for people not to include those packages for
>which you have chosen a free license, but that's against the spirit of
>the license in the first place.
tosh. many of us (i'm certainly one) object to our software being
distributed in such a way that people are likely to make random
modifications and then moan at us because the result doesn't work.
it doesn't mean that we don't want our software to be redistributed,
merely that we want to manage our own workload.
--
Robin Fairbairns, Cambridge
David Kastrup
> > > Until these points are clarified, please do not include my packages
> > > in the distribution.
> >
> > If you refuse to let your packages be distributed in certain freely
> > redistributable distributions, you have to put them under a less free
> > license.
> >
>
> No. If the distribution is modified so that it meets the condtitions in
> latex's legal.txt (specifically that it includes all the latex sources
> if it includes any) then I am sure that anyone has a full right to make
> a distribution. However rights are one thing, I can still _ask_ that my
> files are not included (temporarily) until what I see as some potential
> problems are sorted out.
So you see it as a potential problem that the distribution is calling
itself GNU-TeX. Big deal.
Do you know what the difference between GNU/Linux and Linux is
according to the FSF? Linux is just a kernel and is called just
Linux, even though it is GPLed in its entirety. GNU/Linux is a
complete operating system built around this kernel and is called
GNU/Linux even though the GPLed parts make up about 30% of it (but
everything is freely redistributable). This is because something like
GNU/Linux was already concocted by them before the time of Linux
Torvalds, and he took all the stuff and gave it a kernel to revolve
around.
So I cannot see a particular problem with the name. The FSF has in
the past made a lot of enemies with its naming games (particularly in
Linux camps) for products for which they are not responsible apart
from delivering components, but if a team of them or others is going
to name something "GNU" all by their own, I don't think there is going
to be trouble. "GNU" in some name always meant that it was intended
to be a component of the GNU system. The GNU system has nothing to do
with the GPL, except that the FSF has decided that all components to
that system it would be managing by themselves would be under the GPL,
and that they are encouraging others to follow this model.
BTW, if they abide by your request, it means that they are not going
to distribute LaTeX at all, since distributing LaTeX without your
files is prohibited by its license.
Personally, I don't think people wanting to make a freely
redistributable distribution should be asked to cripple it by leaving
out the entire LaTeX, when you have no problem with commercial
distributions containing it.
Whatever the motivations of the distributors are, I think this is a
bit of an overreaction.
Actually, teTeX has been without any serious competitor in the Unix
scene for quite long. With Linux, we have half a dozen main
distributions to choose from and dozens of others. This has lead to a
lot of development and competition with regard to ease-of-use and
installation, and a lot of improvement.
The crucial point is that with freely distributable software (and
GNU-TeX will be concerned only with that), there is no necessity to
reinvent the wheel, because all the blueprints are already available.
When a particular freely redistributable distributions has a
"killer-component" in it, every other freely redistributable
distribution can just take and include it.
So it is not that good ideas get wasted with this approach. Good
ideas will end up in all distributions eventually, not so good ones in
fewer or just a single one.
In short, I can see little to no harm done by the advent of a
"GNU-TeX" distribution, even though it remains to be seen whether the
effort will lead to added value or not.
David Carlisle
> So you see it as a potential problem that the distribution is calling
> itself GNU-TeX. Big deal.
Not really (not much of a problem, and not much of a deal)
The name could have been better chosen. But there are other
distributions with worse names (truetex for instance) and
I wouldn't really object to the name if that's all there is.
> BTW, if they abide by your request, it means that they are not going
> to distribute LaTeX at all, since distributing LaTeX without your
> files is prohibited by its license.
Yes, I don't really expect that that will be a final outcome, but most
people who are starting to put together a new distribution do contact
the authors of the main components first, to see if they have any
comments or views. They don't just agressively make announcements
that affectively say it will be done the way we like it or we will
attempt to take over by producing `rival' software. It just doesn't seem
a very friendly approach.
I didn't say they could not include my files ever, I just asked that
they did not do that until certain issues have been resolved. It turns
out since then that one of the issues is that they anyway do not follow
the latex distribution conditions as they only selectively distribute
files.
> Whatever the motivations of the distributors are, I think this is a
> bit of an overreaction.
If you go to the trouble of writing a license and distribution file
and then someone ignores it, then surely it is not an overreaction to
say that files should not be distributed until the conditions are
adhered to. Especially as in this case if the conditions were especially
worded to meet the concerns of the GNU project.
> So it is not that good ideas get wasted with this approach. Good
> ideas will end up in all distributions eventually, not so good ones in
> fewer or just a single one.
and until that time, those of us answering questions on c.t.t have to
go back in time to the days when we had to guess what kind of
installation a unix tex user might have and how that particular setup
might be configured before we can give any advice on what a user is
supposed to do.
I do respect the GNU project's position on free software, and I can see
why they might have a problem with some parts of tetex, but working to
get tetex modularised in such a way that free and non free files may be
easily separated would seem like a far more useful way for anyone to
spend the time, than just setting up a new distribution with its own
configuration commands.
David
David Kastrup
> David Kastrup <d...@mailhost.neuroinformatik.ruhr-uni-bochum.de> wrote:
> >David Carlisle <da...@dcarlisle.demon.co.uk> writes:
>
> >> TeX is not under GPL
> >
> >But you could GPL an arbitrary version of it, as it is in the Public
> >Domain.
>
> so long as you didn't imply that the result was a tex implementation.
Not at all. All that is required to use the name "TeX" is to adhere
to the conditions the AMS requires for the use of their trademarked
name "TeX". Not licensing TeX under the GPL is not part of these
conditions. Passing the Trip test is.
David Kastrup
> >> Computer Modern fonts are not under GPL
> >
> >Same here.
> >
> >> LaTeX is not under GPL
> >
> >It *is* freely redistributable.
>
> but not modifiable under its own name.
Obviously, GNU-TeX will not be distributed with all of its components
under the GPL (they cannot change the license of something for which
they don't have copyright). So what problem do you have?
> >> The majority of contributed packages are not under GPL.
> >
> >Some of them are *not* freely distributable. Those would be missing
> >from GNU TeX, presumably. Do you have any idea how many packages
> >(like foilTeX) are distributed illegally on Linux CDs?
>
> it's clear that linux cds can't be relied on for satisfactory versions
> of tex (vide the business a while back where one could buy such cds
> with corrupt cm fonts on them). anyone who assumes otherwise is
> already making a considerable leap in the dark.
And you want it to remain that way? You insist that no distribution
should be allowed that would be safe for Linux distributors to
include? Why?
> >Of course, splitting teTeX into freely redistributable parts and
> >others would also have been a choice.
>
> thomas should do a bunch of work because someone else wants to make
> money from it? i boggle.
Ok, you think it objectionable that someone should take the work to
split teTeX into freely redistributable parts and non-freely
redistributable. You want that the only person working on teTeX ever
should be Thomas Esser (you do not even consider the option that
somebody else might help him with that). You do object to a
*separate* effort to make a freely redistributable TeX distribution.
In short, you object to a freely redistributable TeX distribution
for Unix regard less of who does it. Why?
> >> The kpathsea path searching library which makes up an important part
> >> of the unix tex distribution is under GPL, but that is by no means
> >> the whole of the distribution.
> >
> >It does not need to be. The GNU Hurd, for example, uses a non-GPLed
> >Microkernel.
>
> you mean, people are still banging away at that? i boggle.
Yes, they are, for the reason that they thought as far as they have
got already, they might as well finish it. See you nearest GNU
mirror.
> >> Until these points are clarified, please do not include my packages
> >> in the distribution.
> >
> >If you refuse to let your packages be distributed in certain freely
> >redistributable distributions, you have to put them under a less free
> >license.
>
> anything of mine that claims to offer support, does so in the case of
> *unmodified* versions of the package.
If you license your package in that way, the GNU folks cannot change
the license, nor will they (at least if "GNU folks" means the FSF
which are *very* anal about slavishly adhering to a license or not
using the software at all).
> aiui, unless i take special steps to the contrary, my stuff is now
> going to appear on gnu archive sites, which will encourage people to
> assume they can hack around with it, redistribute modified versions,
> and so on. i don't like this.
If people are stupid, people are stupid. There is other stuff on the
GNU mirrors with licenses other than the GPL. If you don't want
stupid people to use your software, put a minimum IQ requirement in
your license.
> why can't this gnu crowd bring itself to do what sebastian rahtz has
> done re. the tex live distributions, viz., to contact package
> authors for their approval, rather than to suggest that package
> authors need to contact the crowd?
If the license of the software conflicts with the distribution terms
of GNU-TeX, they must do so. If not, I do not see the necessity.
After all, one of the main advantages of free software is *not* to
have to phone around for ages before basing a project on it.
> >Of course, you can *ask* for people not to include those packages for
> >which you have chosen a free license, but that's against the spirit of
> >the license in the first place.
>
> tosh. many of us (i'm certainly one) object to our software being
> distributed in such a way that people are likely to make random
> modifications and then moan at us because the result doesn't work.
What do you object to? If it is distributed according to its license,
you get all the copyright headers in it that tell you you cannot
modify the file and report bugs. If people are too stupid to read,
they will be too stupid to read regardless whether their distribution
is named GNU-TeX or teTeX.
> it doesn't mean that we don't want our software to be redistributed,
> merely that we want to manage our own workload.
By prohibiting a project to further the distribution of TeX-related
software by distributing your software under the terms of the license
you yourself put on it?
Sorry, I don't get it.
While I agree that it might have been more fruitful to aid with teTeX
in order to get it into a state where the GNU people might use it
unmodified, I am simply surprised by all people crying bloody murder.
When your software is distributed violating your license, you have all
right to complain and even to sue.
If this is not the case, the clamour is hypocritical. If you don't
want everybody to use your software under your license, then don't
license it that way.
David Carlisle
> Not at all. All that is required to use the name "TeX" is to adhere
> to the conditions the AMS requires for the use of their trademarked
> name "TeX". Not licensing TeX under the GPL is not part of these
> conditions. Passing the Trip test is.
If tex were distributed under GPL (only) then anyone else would be able
to take that code, modify it and redistribute the modification as tex.
It is precisely to stop this that tex and latex are distributed under
conditions rather different to the GPL. However these conditions have
been accepted as `Free' by people involved in GNU, so the conditions
on tex and latex are in no way a bar to them appearing in a GNU system.
David
David Kastrup
> > Not at all. All that is required to use the name "TeX" is to adhere
> > to the conditions the AMS requires for the use of their trademarked
> > name "TeX". Not licensing TeX under the GPL is not part of these
> > conditions. Passing the Trip test is.
>
> If tex were distributed under GPL (only) then anyone else would be able
> to take that code, modify it and redistribute the modification as
> tex. It is precisely to stop this that tex and latex are distributed under
> conditions rather different to the GPL.
With regard to TeX, I would say that you are confused. TeX, the
Program, is Public Domain. You may change it in any way you like and
distribute it around. You may not call it TeX, however, since TeX is
a trademark of the AMS and you would violate the trademark if you
would call it TeX.
With LaTeX, the condition is that you have to change the file name as
well when changing the file. This is mainly a nuisance, but does not
preclude yourself adapting the code to do what you want. Which is the
reason that it has a grudging "ok" with GNU people.
> However these conditions have been accepted as `Free' by people
> involved in GNU, so the conditions on tex and latex are in no way a
> bar to them appearing in a GNU system.
So what particular problem are you referring to if you say that the
GNU-TeX distribution would have to be in violation of the LaTeX
license?
David Carlisle
> So what particular problem are you referring to if you say that the
> GNU-TeX distribution would have to be in violation of the LaTeX
> license?
It doesn't have to be, it just is. (So they tell me).
It does not contain the full latex sources. I would have thought the
requirement to include full sources would not have been a problem to
anyone used to the GNU license.
To repeat there is one technical infingement (not including full
sources) which they can easily put right. That is as far as I know the
only reason that it would break the license conditions.
More of a problem is to answer the question as to whether it is helpful
to get in to the buisness of competing unix tex distributions.
This I wouldn't normally have responded to publicly even if I had a
private view on it, if it hadn't been for the rather inappropriatly
aggressive nature of the gnutex announcements. Publicly asking for
volunteers to help `replace' software that has been freely given (even
if with slightly inconvenient licenses) does not seem to be a very polite
way to introduce yourself to the TeX community.
David
David Kastrup
> > BTW, if they abide by your request, it means that they are not going
> > to distribute LaTeX at all, since distributing LaTeX without your
> > files is prohibited by its license.
>
> Yes, I don't really expect that that will be a final outcome, but most
> people who are starting to put together a new distribution do contact
> the authors of the main components first, to see if they have any
> comments or views. They don't just agressively make announcements
> that affectively say it will be done the way we like it or we will
> attempt to take over by producing `rival' software. It just doesn't seem
> a very friendly approach.
The announcement I read on this group (please tell me if you had
something different) was:
GNU TeX, version 0.80 (beta) is available for testing, and should
be stable and complete enough for use. It is an implementation of
TeX and LaTeX, and is based on the standard Web2c source
distribution. GNU TeX follows the TeX Directory Standard (TDS)
and includes these packages:
[...]
Much of the work before Version 1.0 involves writing documentation
and help texts, coding virtual fonts, and general clean-up of the
TEXMF library and the source code for the binaries.
For later releases, we would like to include free replacements for
some commonly used non-free TeX and LaTeX packages. (We cannot
include those packages in this free software collection.) If you
are an expert TeX hacker and you are seriously interested in
writing a replacement for a non-free package, please send me mail.
teTeX or any other distribution is not even mentioned (see footnote).
The packages they list are all very *standard* packages. In no way do
they say that they want to develop alternatives to them. They do say,
however, that they would like people to contribute free packages in
addition to the ones they already have. Instead of taking a
distribution they could not distribute according to their rules and
pestering every single author of it that has some non-free component
in it to reconsider its license, they have decided to start with a set
that fulfills their criteria and add to it as required and possible.
This has the advantage of taking probably years less to arrive at
anything they can distribute. This, of course, has disadvantages as
well.
They don't talk of "takeover", they don't talk of "rival software" or
anything. I can discern no "aggressiveness" in the announcement
except that they dare to announce a project with a different focus in
an area where up to now mostly a single other product has been
prevalent.
They do not "take over" anything, but simply start a fresh project
that meets their requirements from scratch.
Like everyone else, they will have to adhere to the licenses of the
components they are including. If they find this infeasible, they
will have to write replacements, ask others to do so, or give up.
As long as they play by the rules, I see no reason to treat them worse
than any commercial distributor of TeX-related products.
Footnote: The Free Software Foundation has the policy not to talk
about any products that do not meet their redistributability criteria.
At times, this leads to situations one could consider a bit weird.
For example, Aladdin Software produces Ghostscript which can be freely
copied by people under its license, but may not be commercially
reproduced for an arbitrary fee, thus do not meet the criteria for
free software that the FSF applies. Aladdin continually donate
versions that are about a year old to the GNU project. The
documentation of GNU Ghostscript is checked and censored in order to
remove any trace of information about the Ghostscript versions not
released under GPL, and is then released in its cleaned-up form.
Similarly, the FSF Web pages never mention Aladdin Ghostscript or its
existence.
The principle behind this is understandable: they don't want to act as
an advertisement agency for software distributed under a license they
don't feel acceptable. Yet the strict enforcement of their policies
might seem exaggeratedly Puritan at times. Other people would prefer
the term "raving lunatic" for it. But though this be madness, there
is system to it. Which means that one does them an injustice if one
considers them equal to every other madman. It is a very peculiar,
consistent and predictable madness. As opposed to other belief
systems dressing as moral, you can take or leave it, however, so it is
a rather harmless form of madness not particularly dangerous to the
general public.
Phil Edwards
David Carlisle <dav...@nag.co.uk> wrote:
+
+ Yes, I don't really expect that that will be a final outcome, but most
+ people who are starting to put together a new distribution do contact
+ the authors of the main components first, to see if they have any
+ comments or views.
At the very least! Has TE at least been notified?
+ They don't just agressively make announcements
+ that affectively say it will be done the way we like it or we will
+ attempt to take over by producing `rival' software. It just doesn't seem
+ a very friendly approach.
This very reason is why I will strongly resist using "GNU-TeX" (or
whatever its name eventually becomes). Of /course/ they never announce
these intentions as such, but even /suggest/ to a GNU discussion list that
"there might be a more productive way," and you are forever anathema.
I have my own problems with tetex-0.9, but it /works/ and works very,
very, very well. With one exception, a GNU distribution of TeX offers
little benefits besides the FSF being able to say, "See, even the science
community relies on us!"
+ and until that time, those of us answering questions on c.t.t have to
+ go back in time to the days when we had to guess what kind of
+ installation a unix tex user might have and how that particular setup
+ might be configured before we can give any advice on what a user is
+ supposed to do.
This is the (possible) GNU benefit. While /using/, say, LaTeX, is child's
play, installing/configuring/debugging is a nightmare. My co-workers
(fellow TeX lovers, all of them) have repeatedly observed that all
problems in configuring TeX are treated as trivial by the newsgroup;
that's because they all /are/ trivial, once you already know the answer!
One of the things I've always appreciated about the GNU products I use
is that instructions for setup (apart from the generic INSTALL file)
are almost always up-to-date. When I tried to install the "current"
release of tetex-0.4, it exploded miserably. I then tried the "fragile
beta, do not use unless yer a developer" tetex-0.9 that had been out for
several months, and it worked fine. (I was flamed out of existence for
suggesting that 0.4 should be trashed, and that 0.9 should have been
made the "current" release a long time ago, since in reality it was.
See DejaNews if you're terminally bored.)
I've never had any problems like that with a GNU product. On the
flip side, I've rarely met any developers/maintainers as arrogant as
(unfortunately) many of the FSF are. I'll stick with tetex/MiKTeX,
but that's just my pointless opinion.
Luck++;
=> If you post a followup, PLEASE don't email a copy to me. I read news
often enough that replying to things twice is annoying. But thanks.
Sven Utcke
> In article <m390f24...@vh213601.truman.edu>,
> Jay Belanger <j...@vh213601.truman.edu> wrote:
> % as...@reg.triumf.ca (Donald Arseneau) writes:
>
> % > I'll point out, for people not so familiar with GNU and FSF that
> % > their use of "free" is, err, somewhat non-standard english.
> % > http://www.gnu.org probably has an extensive definition.
> %
> % Actually, it is standard english, the problem is that the adjective
> % "free" means different things in different situations. The first
> % meaning that probably comes to mind is "at no cost", but that isn't
> % always the correct meaning. In some situations it clearly means
> % "unconstrained" (which is the first meaning in my dictionary). In
> % some situations, the word "free" needs further explanation; e.g, my dog
>
> For instance, when not using a definition from standard English. I recall
> the fiasco surrounding ispell when FSF decided it wasn't free in just the
> way they wanted it to be.
Uhu. It's not even a week that I had to tell someone not to get
ispell 4, but rather the newer ispell 3.1.20. So now we get the same
for TeX?
Sven
--
_ _ Lehrstuhl fuer Mustererkennung und Bildverarbeitung
| |_ __ | |__ Sven Utcke
| | ' \| '_ \ phone: +49 761 203 8274 Am Flughafen 17
|_|_|_|_|_.__/ fax : +49 761 203 8262 79110 Freiburg i. Brsg.
mailto:ut...@informatik.uni-freiburg.de www.informatik.uni-freiburg.de/~utcke
Simon Cozens
> GNU TeX, version 0.80 (beta) is available for testing, and should be
> stable and complete enough for use.
Regardless of the legal and moral implications of using GNU software,
I completely fail to see the *gain* of this. Why do we *need* another
UNIX TeX distribution? The ones we have are perfectly adequate; I
personally use tetex at home, and I am not beholden to anyone.
Should I feel the distribution itself somehow binds me and restricts my
sense of `freedom', I'm perfectly at liberty (as it were) to assemble my
own working TeX system from the sources supplied on CTAN.
I'm sure there is a good reason why the GNU project, who I usually
wholeheartedly support, should want to march along and create a distribution
to suit their own licensing arrangements. I'd just like to know what it is.
--
So remember when you're feeling very small and insecure
How amazingly unlikely is your birth,
And pray that there's intelligent life somewhere up in space,
'Cause there's bugger-all down here on Earth. (Monty Python)
David Kastrup
> Robert Kiesling (comp.text.tex):
> > GNU TeX, version 0.80 (beta) is available for testing, and should be
> > stable and complete enough for use.
>
> Regardless of the legal and moral implications of using GNU software,
> I completely fail to see the *gain* of this. Why do we *need* another
> UNIX TeX distribution? The ones we have are perfectly adequate; I
> personally use tetex at home, and I am not beholden to anyone.
> Should I feel the distribution itself somehow binds me and restricts my
> sense of `freedom', I'm perfectly at liberty (as it were) to assemble my
> own working TeX system from the sources supplied on CTAN.
That's exactly what they did and now you complain about it. Why are
you perfectly at liberty to do so, and they not?
> I'm sure there is a good reason why the GNU project, who I usually
> wholeheartedly support, should want to march along and create a
> distribution to suit their own licensing arrangements. I'd just like
> to know what it is.
I suppose they want a distribution that can be used for a version of
their GNU system. That means they need a distribution which
exclusively contains components that are freely redistributable even
on commercial CD-Roms.
teTeX is not. The criterion Thomas Esser uses for putting something
into teTeX is whether it is in general use and publicly available, so
that he might make it available per ftp.
Since this is not sufficient for the needs of the GNU project, I guess
they have decided to roll their own distribution that they can legally
include on CD-ROMs for a fee.
That way they have a TeX distrib they can put into the free section of
CD-ROMs without legal trouble, and in particular the GNU project has
been very paranoid about legal matters.
Of course, this sort of surprise attack on the TeX community is
another example of Gnus in a chinashop. The FSF is never going to get
a price for good timing, tone and diplomacy.
That the project is put forward in this way does not mean that it is
not worthwhile. It only casts some suspicions on the ability of the
project organisators to have the communication skills needed for doing
such a project.
Let's just ignore this initial blunder and give them the benefit of
doubt at first.
Jan Vroonhof
> Ok, you think it objectionable that someone should take the work to
> split teTeX into freely redistributable parts and non-freely
> redistributable.
I just think it silly given the fact that work is a already (being)
done, c.f. the debian teTeX package which splits teTeX into free and
non-free parts (in the DFSG sense of course).
Jan
David Kastrup
> David Kastrup <d...@mailhost.neuroinformatik.ruhr-uni-bochum.de> writes:
>
> > Ok, you think it objectionable that someone should take the work to
> > split teTeX into freely redistributable parts and non-freely
> > redistributable.
>
> I just think it silly given the fact that work is a already (being)
> done, c.f. the debian teTeX package which splits teTeX into free and
> non-free parts (in the DFSG sense of course).
This work needs to be redone for every new release of teTeX you want
to do. This might become inconvenient after some time. I can
understand that instead of having to rely on snapshots from a Debian
treated version of teTeX, they might think it more convenient to start
an own distribution.
I do agree that up to now this is all guesswork as the posters of the
GNU-TeX effort have left us entirely in the dark with regard to their
objects and why they think it has been unfeasible to try to achieve
them in cooperation with others.
Simon Cozens
>> sense of `freedom', I'm perfectly at liberty (as it were) to assemble my
>> own working TeX system from the sources supplied on CTAN.
> That's exactly what they did and now you complain about it. Why are
> you perfectly at liberty to do so, and they not?
Because I'm not claiming that I'm putting together a system for distribution.
> Since this is not sufficient for the needs of the GNU project, I guess
> they have decided to roll their own distribution that they can legally
> include on CD-ROMs for a fee.
> That way they have a TeX distrib they can put into the free section of
> CD-ROMs without legal trouble, and in particular the GNU project has
> been very paranoid about legal matters.
This now makes sense. Thank you for responding in a polite and positive
manner! I do start to wonder, though, where it will all end - will there
ever be an Open Source / PD / `Free' software enterprise that will be
allowed to stand on its own two feet without needing to be GNUified?
Granted, TeX is a fairly standard thing to want to include on a Linux
distribution, GNU or otherwise, and this one is understandable. But whatever
next? When will we see GNU Netscape, I wonder...?
--
Everybody wants to go to heaven, but nobody wants to die.
David Carlisle
> Thank you for responding in a polite and positive manner!
But that's half the trouble: David K is the polite and positive face
of this project *but* he knows nothing about it except the announcment
that appeared in this group, and a general knowledge about the kind of
concerns the gnu project has.
It would be nice to hear from people who are involved what they are up
to:-)
> When will we see GNU Netscape, I wonder...?
You mean you don't use gnu emacs w3-mode as your web browser?
David
David Kastrup
> David Kastrup (comp.text.tex):
> >> sense of `freedom', I'm perfectly at liberty (as it were) to assemble my
> >> own working TeX system from the sources supplied on CTAN.
>
> > That's exactly what they did and now you complain about it. Why are
> > you perfectly at liberty to do so, and they not?
>
> Because I'm not claiming that I'm putting together a system for
> distribution.
So it is ok if one assembles ones own working TeX system from the
sources supplied on CTAN as long as one does not plan to make a
distribution from it? So why are the authors of TeX-live allowed to
do a distribution? Why is Thomas Esser allowed to do one? Why is
Fabrice allowed to roll one?
No, I guess that cannot be the problem with the announcement. The
problem is merely that they have not explained their particular aim,
have not explained why none of the existing distros can be made to fit
their purpose satisfactorily.
> > Since this is not sufficient for the needs of the GNU project, I guess
> > they have decided to roll their own distribution that they can legally
> > include on CD-ROMs for a fee.
>
> > That way they have a TeX distrib they can put into the free section of
> > CD-ROMs without legal trouble, and in particular the GNU project has
> > been very paranoid about legal matters.
>
> This now makes sense. Thank you for responding in a polite and positive
> manner!
Well, not being involved with this particular project, I do not feel
compelled to be impolite. I have on numerous occasions been
confronted with GNU people, and there public relation skills are more
often than not non-existent. Which does not mean that the things they
are working on are wildebeest manure.
Unfortunately, we all are still lacking first-hand knowledge about the
particular this project tries to satisfy.
> I do start to wonder, though, where it will all end - will there
> ever be an Open Source / PD / `Free' software enterprise that will be
> allowed to stand on its own two feet without needing to be GNUified?
How about XFree86? sendmail? Apache? The BSD name server? The
entire free BSD stuff, much of which is used in GNU projects? Those
things are sufficiently free already.
But about your question: the GNU proponents will *not* be satisfied
with products where only versions that cannot be freely redistributed
and reused at least in the manner the GPL license permits are
available. It will end when a completely usable exists, that is
freely redistributable and available in source form.
> Granted, TeX is a fairly standard thing to want to include on a Linux
> distribution, GNU or otherwise, and this one is understandable. But whatever
> next? When will we see GNU Netscape, I wonder...?
The sufficiently free Netscape is Mozilla and is developed at
www.mozilla.org (or was that com?).
Currently no other Web browser project is actively propagated by the
FSF, for that reason.
Christopher Browne
<d...@mailhost.neuroinformatik.ruhr-uni-bochum.de> wrote:
>Simon Cozens <si...@brecon.co.uk> writes:
>
>> David Kastrup (comp.text.tex):
>> >> sense of `freedom', I'm perfectly at liberty (as it were) to assemble my
>> >> own working TeX system from the sources supplied on CTAN.
>>
>> > That's exactly what they did and now you complain about it. Why are
>> > you perfectly at liberty to do so, and they not?
>>
>> Because I'm not claiming that I'm putting together a system for
>> distribution.
>
>So it is ok if one assembles ones own working TeX system from the
>sources supplied on CTAN as long as one does not plan to make a
>distribution from it? So why are the authors of TeX-live allowed to
>do a distribution? Why is Thomas Esser allowed to do one? Why is
>Fabrice allowed to roll one?
Indeed. If these people are allowed to build distributions of TeX, why
*not* another?
>No, I guess that cannot be the problem with the announcement. The
>problem is merely that they have not explained their particular aim,
>have not explained why none of the existing distros can be made to fit
>their purpose satisfactorily.
There is the legitimate issue that teTeX doesn't partition itself into
freely redistributable components with sufficient clarity for FSF
purposes.
And it's not just an issue for "FSF purists;" Linux distributions (that
the "FSF purists" prefer to call "GNU systems") *do* have to concern
themselves with the status of software. Kermit is a good example of a
program that is "free by some standards" that cannot without some pretty
special action be legally included as part of a Linux distribution. And
there are pieces of teTeX for which licensing is unclear.
>> > Since this is not sufficient for the needs of the GNU project, I guess
>> > they have decided to roll their own distribution that they can legally
>> > include on CD-ROMs for a fee.
>>
>> > That way they have a TeX distrib they can put into the free section of
>> > CD-ROMs without legal trouble, and in particular the GNU project has
>> > been very paranoid about legal matters.
>>
>> This now makes sense. Thank you for responding in a polite and positive
>> manner!
>
>Well, not being involved with this particular project, I do not feel
>compelled to be impolite. I have on numerous occasions been
>confronted with GNU people, and there public relation skills are more
>often than not non-existent. Which does not mean that the things they
>are working on are wildebeest manure.
The luminaries such as RMS add, to any technical considerations, the
fact that they are extremists. Back at the start of the GNU Project,
the participants were thought a bit "crazy," and had to persist through
considerable opposition.
Unfortunately, the "scars" from the process, along with the
consideration that they have thought "FSF thoughts" so long that they
have become basic assumptions (that to others are beliefs that may be
suspended at need) certainly does make them appear more than a little
extreme.
>> I do start to wonder, though, where it will all end - will there
>> ever be an Open Source / PD / `Free' software enterprise that will be
>> allowed to stand on its own two feet without needing to be GNUified?
>
>How about XFree86? sendmail? Apache? The BSD name server? The
>entire free BSD stuff, much of which is used in GNU projects? Those
>things are sufficiently free already.
There is an unfortunate tendancy of the FSF to call much of this sort of
stuff "GNU Software" even if the authors of the software do not favor
that moniker. (See any issue of the GNUS Bulletin for ample examples.)
>But about your question: the GNU proponents will *not* be satisfied
>with products where only versions that cannot be freely redistributed
>and reused at least in the manner the GPL license permits are
>available. It will end when a completely usable exists, that is
>freely redistributable and available in source form.
Indeed.
>> Granted, TeX is a fairly standard thing to want to include on a Linux
>> distribution, GNU or otherwise, and this one is understandable. But whatever
>> next? When will we see GNU Netscape, I wonder...?
>
>The sufficiently free Netscape is Mozilla and is developed at
>www.mozilla.org (or was that com?).
>
>Currently no other Web browser project is actively propagated by the
>FSF, for that reason.
I don't think that the "MPL" license of Mozilla proved to be quite
satisfactory to the FSF "denizens."
There has not been a sufficient population of people interested in
replicating Mozilla functionality as to allow for much of a project on
the FSF side of the world.
Note that there *cannot* be a "GNU Netscape" as such:
- Netscape Communications Corporation owns the trademark "Netscape," and
the release of source code does not change that.
- The MPL is not directly compatible with the GPL (most source code
licenses that are "more restrictive" than BSD licenses are not
particularly compatible with the GPL), which means that the FSF can't
combine Mozilla code into GPL code, and vice-versa.
--
"take USABLE from UNSTABLE and you get NT"
cbbr...@hex.net- <http://www.hex.net/~cbbrowne/lsf.html>
Robin Fairbairns
>>sources supplied on CTAN as long as one does not plan to make a
>>distribution from it? So why are the authors of TeX-live allowed to
>>do a distribution? Why is Thomas Esser allowed to do one? Why is
>>Fabrice allowed to roll one?
>
>Indeed. If these people are allowed to build distributions of TeX, why
>*not* another?
thomas and fabrice (who are doing complementary things with the same
basic setup, aiui) don't sell anything at all.
sebastian (producing tex live) has gone to some lengths to ensure that
copyright issues in re. anything on the disc are addressesed.
tex-live is `formally' not sold (i.e., it's supposed only to go for
production and distribution costs, not for profit).
>>No, I guess that cannot be the problem with the announcement. The
>>problem is merely that they have not explained their particular aim,
>>have not explained why none of the existing distros can be made to fit
>>their purpose satisfactorily.
>
>There is the legitimate issue that teTeX doesn't partition itself into
>freely redistributable components with sufficient clarity for FSF
>purposes.
but that doesn't explain why the fsf effort isn't (like the debian
group's already is) devoted to that end in the first place.
we already have enough, stable enough, tex distributions for un*x-
alikes. the fsf people, with their "anything you can do, we can do
better" approach are merely putting up the backs of people who have
worked hard over many years to get to the point we are now at.
there *is* point to partitioning things, but not to starting all over
again.
>And it's not just an issue for "FSF purists;" Linux distributions (that
>the "FSF purists" prefer to call "GNU systems") *do* have to concern
>themselves with the status of software. Kermit is a good example of a
>program that is "free by some standards" that cannot without some pretty
>special action be legally included as part of a Linux distribution. And
>there are pieces of teTeX for which licensing is unclear.
kermit's another of these weird setups. a guy who did his work with
money from defence budgets, but now refuses to allow his work to be
used in defence projects. (i would love to do the latter myself, but
i compromised myself >20 years ago by working for a firm that took
defence money for software development.)
>>Well, not being involved with this particular project, I do not feel
>>compelled to be impolite. I have on numerous occasions been
>>confronted with GNU people, and there public relation skills are more
>>often than not non-existent. Which does not mean that the things they
>>are working on are wildebeest manure.
>
>The luminaries such as RMS add, to any technical considerations, the
>fact that they are extremists. Back at the start of the GNU Project,
>the participants were thought a bit "crazy," and had to persist through
>considerable opposition.
back at the start, most people of my acquaintance said "it's all very
well, but what do we do for a living if no-one's allowed to charge for
software?". but we all had a lot of sympathy with him, because the us
patent office (in particular) and intellectual property lawyers (in
general) were making such a complete hash of software protection.
many of us persuaded our firms to support the fsf, by buying things
from them (uk firms couldn't take advantage of usan tax breaks by
making donations at the time).
in the mean time, some of us started on the path that has led us to
spending a lot of our time supporting the tex community world wide.
and now we find ourselves feeling that the outfit, which we once
supported, is turning around and spitting in our collective faces.
sigh.
--
Robin Fairbairns, Cambridge
David Kastrup
> Christopher Browne <cbbr...@hex.net> wrote:
> >On 19 Jan 1999 21:57:07 +0100, David Kastrup
> >>So it is ok if one assembles ones own working TeX system from the
> >>sources supplied on CTAN as long as one does not plan to make a
> >>distribution from it? So why are the authors of TeX-live allowed to
> >>do a distribution? Why is Thomas Esser allowed to do one? Why is
> >>Fabrice allowed to roll one?
> >
> >Indeed. If these people are allowed to build distributions of TeX, why
> >*not* another?
>
> thomas and fabrice (who are doing complementary things with the same
> basic setup, aiui) don't sell anything at all.
>
> sebastian (producing tex live) has gone to some lengths to ensure that
> copyright issues in re. anything on the disc are addressesed.
> tex-live is `formally' not sold (i.e., it's supposed only to go for
> production and distribution costs, not for profit).
So you obviously *do* see that currently there is no distribution
that, say, RedHat Linux could put on one of their Linux CD-ROMs (which
they sell) without being liable to breach of copyright charges. All
of the existing distros have taken pains not to get involved in
commercial settings. You effectively say that it is a bad idea to
make a Unix distribution that could be included in something like
RedHat Linux legally and without further negotiations. Why?
> >There is the legitimate issue that teTeX doesn't partition itself into
> >freely redistributable components with sufficient clarity for FSF
> >purposes.
>
> but that doesn't explain why the fsf effort isn't (like the debian
> group's already is) devoted to that end in the first place.
I have seen nothing in the announcement that would have stated a
different goal. It seems to me very plausible that they are devoted
to exactly that end, as this is the principal aim of almost every GNU
effort, and in particular the GNU project in toto.
> we already have enough, stable enough, tex distributions for un*x-
> alikes.
We don't have any freely redistributable that could be distributed for
a profit. Where does profit start? With recouping costs? A
corporation can do nothing more than recouping costs, because the
corporation itself has no way of spending money for its enjoyment,
only its employees can. Where do you draw the line?
> the fsf people, with their "anything you can do, we can do
> better" approach are merely putting up the backs of people who have
> worked hard over many years to get to the point we are now at.
They don't say "anything you can do, we can do better". Their stance
always has been: "we need freely distributed software. Regardless of
the quality of existing not freely distributable software, we will
start our own projects in order to provide this if necessary". There
does not exist a reasonably complete Unix distribution that meets
their ends. So they started their own.
> there *is* point to partitioning things, but not to starting all over
> again.
As most of the work done already *is* freely redistributable, they
don't have to start over all again. They just have to sort out
things.
> back at the start, most people of my acquaintance said "it's all very
> well, but what do we do for a living if no-one's allowed to charge for
> software?". but we all had a lot of sympathy with him, because the us
> patent office (in particular) and intellectual property lawyers (in
> general) were making such a complete hash of software protection.
>
> many of us persuaded our firms to support the fsf, by buying things
> from them (uk firms couldn't take advantage of usan tax breaks by
> making donations at the time).
>
> in the mean time, some of us started on the path that has led us to
> spending a lot of our time supporting the tex community world wide.
> and now we find ourselves feeling that the outfit, which we once
> supported, is turning around and spitting in our collective faces.
Whatever names you want to call the FSF, they *never* *ever* turned
around in any way. You can accuse them of being absolutely rigid and
immobile, but they don't turn around. From the start, they have
always said that the only acceptable solution to them is free
software, taht can be freely redistributed and modified. A lot of
people try to feel somewhat close to their ideals by cranking out
shareware software, or software not to be distributed commercially, or
other forms of software that are in the end closed to the dynamics one
would want for spreading free software effectively.
They also have not "spit in your face" in any way. They have said
that they still have a need for a distribution that meets their
criteria.
If IBM chose to sue RedHat Linux for damages because they were
including foiltex on their CD-ROMs, they would win. If anybody
presented slides he made while using the multicol.sty included in the
RedHat distributions at a company meeting, and Frank Mittelbach were
to be present, he could sue the presenter for breach of copyright.
This is a situation that free software should not find itself in. If
I buy free software and use it or distribute it, even for a profit, I
should not be liable to damages. This very much can harm corporate
acceptance of free software. If people use something like RedHat
Linux, they want to know they are in a secure position.
Having a Unix TeX distribution that will try meeting this aim actively
will help towards this position. If it gets obsoleted some day
because other existing distributions apply the same high set of
standards, so much the better. Currently this is not to be seen.
Timothy Murphy
>Do you know what the difference between GNU/Linux and Linux is
>according to the FSF? Linux is just a kernel and is called just
>Linux, even though it is GPLed in its entirety. GNU/Linux is a
>complete operating system built around this kernel and is called
>GNU/Linux even though the GPLed parts make up about 30% of it (but
>everything is freely redistributable). This is because something like
>GNU/Linux was already concocted by them before the time of Linux
>Torvalds, and he took all the stuff and gave it a kernel to revolve
>around.
I guess most people reading this group are familiar with Linux,
but it ought to be said that the above is gobbledygook.
Stallman and FSF tried to get the Linux community to use the term GNU/Linux
in place of Linux.
The suggestion was unanimously rejected.
[Linus Torvalds gave a marvellous demonstration
of how to carry on a rational discussion with fundamentalists.]
GNU/Linux does not exist outside the overheated imaginations of FSF fanatics.
My worry about GNU-TeX is that it sounds depressingly like the same scenario.
Hopefully I am wrong.
--
Timothy Murphy
e-mail: t...@maths.tcd.ie
tel: +353-1-2842366
s-mail: School of Mathematics, Trinity College, Dublin 2, Ireland
David Carlisle
> There does not exist a reasonably complete Unix distribution that
> meets their ends. So they started their own.
Having now looked at the files, and also given the original
announcement's statement about smaller disk space requirements,
I'd say that this gnutex, at least in its present form, is not aiming to
be a `reasonably complete Unix distribution' in the sense of tetex Ntex
and texlive. It is more aiming to be a well documented minimal starter
kit in the spirit of the texmflib tar file that Karl Berry always used
to put in the core web2c distribution.
> As most of the work done already *is* freely redistributable, they
> don't have to start over all again. They just have to sort out
> things.
But that isn't what they are doing. There is no sign of this setup using
the tetex configuration scripts, so it just makes it that much harder to
answer the questions on this newsgroup `how do I configure ....'.
> This is a situation that free software should not find itself in.
I agree there. (Without wanting to comment on the particular examples
you gave).
> Having a Unix TeX distribution that will try meeting this aim actively
> will help towards this position.
But currently the impression (and as far as I can see, the actual fact)
is that this distribution helps the distributors (as they need not worry
about legal issues) but does so at the expense of the tex users.
When compared to say tetex it is not just the non free stuff that it is
missed out; it is almost all of it. No `contrib' latex packages at all,
not one. No hyperref, no natbib, no pstricks, no nothing.
Perhaps this is due to the `alpha' status of the release, but
who knows, if they don't say what they are up to. If the final version
comes out looking like this and linux distributions contain this as
their tex distribution, what is a latex user to do, figure out which
packages he needs and install them all individually, or just get tetex
or texlive and do a single install command to replace the whole lot?
It seems like the second choice will be preferable to most people.
David
Stefan Monnier
> Stallman and FSF tried to get the Linux community to use the term GNU/Linux
> in place of Linux.
Yes. The reason quite clearly was "to give credit where credit is due".
I.e. to include the word "GNU" to make it clear that it critically depends
on many tools developped with the help of the FSF.
> My worry about GNU-TeX is that it sounds depressingly like the same scenario.
> Hopefully I am wrong.
Reading my paragraph above I feel quite confident to say you are indeed utterly
wrong. What they are doing here is to setup a distribution that follows the
FSF guidelines so that the FSF can then support TeX.
I don't think this is fundamentally bad, although it's clearly been done is the
most awkward way, ensuring a strong (and justified I have to say) opposition
from the very beginning. The lack of participation of the GNU TeX group in
this very thread is even more disturbing.
Maybe the Debian-teTeX maintainer should just go ahead and quickly release
FreeTeTeX making GNU TeX pointless.
Stefan
Jay Belanger
> David Kastrup <d...@mailhost.neuroinformatik.ruhr-uni-bochum.de> writes:
>
> >Do you know what the difference between GNU/Linux and Linux is
> >according to the FSF? Linux is just a kernel and is called just
> >Linux, even though it is GPLed in its entirety. GNU/Linux is a
> >complete operating system built around this kernel and is called
> >GNU/Linux even though the GPLed parts make up about 30% of it (but
> >everything is freely redistributable). This is because something like
> >GNU/Linux was already concocted by them before the time of Linux
> >Torvalds, and he took all the stuff and gave it a kernel to revolve
> >around.
>
> I guess most people reading this group are familiar with Linux,
> but it ought to be said that the above is gobbledygook.
Hmm? Which part is gobbledygook? Are you saying that the above is
*not* the difference according to the FSF? What, then, is the
difference between Linux and GNU/Linux according to the FSF? Does the
FSF think that there is a difference?
I thought that David Kastrup nicely summarized the FSF's position, but
perhaps you have some specific corrections?
> Stallman and FSF tried to get the Linux community to use the term GNU/Linux
> in place of Linux.
> The suggestion was unanimously rejected.
> [Linus Torvalds gave a marvellous demonstration
> of how to carry on a rational discussion with fundamentalists.]
> GNU/Linux does not exist outside the overheated imaginations of FSF fanatics.
The suggestion of using "lignux" was, quite appropriately in my
opinion, nearly unanimously rejected. However I see "GNU/Linux"
referred to quite a bit, even outside the FSF. Granted, using "Linux"
to refer to the whole operating system seems to me to be by far more
common than "GNU/Linux", but both are out there.
Jay
Louis Vosloo
> their licensing. Another issue is fonts. That area probably will
> remain contentious--all you need to do is read comp.fonts for a while
> to be convinced--and with that consideration in mind, the distribution
> includes only free fonts.
Which are those?
Are you planning on including the AMS/BSR/Y&Y CM and AMS fonts in Type 1 format?
How does the AMS copyright stand on that?
If you include free fonts, will they come with "source code"?
And the right for any one to modify them anyway they please?
What about Knuth's dictum about not calling them CM fonts if the metrics change?
Seems like a potential can of worms to me, somehow.
David Kastrup
> > There does not exist a reasonably complete Unix distribution that
> > meets their ends. So they started their own.
>
> Having now looked at the files, and also given the original
> announcement's statement about smaller disk space requirements,
> I'd say that this gnutex, at least in its present form, is not aiming to
> be a `reasonably complete Unix distribution' in the sense of tetex Ntex
> and texlive. It is more aiming to be a well documented minimal starter
> kit in the spirit of the texmflib tar file that Karl Berry always used
> to put in the core web2c distribution.
It might be that it is intended to become something larger later on.
Without them deigning to inform us of their intentions, we may never
know. I suggest we rest this discussion until we are kindly informed
what they are planning to do eventually.
If this proves to be above them, I guess we need not worry much about
a split of efforts. Not unlikely. Remarkably few comments by them
appeared in the newsgroup, and the only personal communication I
received scoffed me for calling their announcement an "attack" which I
cannot remember to have made. Probably they confused me with David
Carlisle, and even he quickly reverted to civilized discourse after
initial anger which it would rather have been their task to qualm
rather than mine.
> > As most of the work done already *is* freely redistributable, they
> > don't have to start over all again. They just have to sort out
> > things.
>
> But that isn't what they are doing. There is no sign of this setup using
> the tetex configuration scripts, so it just makes it that much harder to
> answer the questions on this newsgroup `how do I configure ....'.
Actually, mst of the questions are of the sort "LaTeX does not run"
without mentioning the versin of LaTeX, the system it is running on
and so on.
It might also be unfair to forget that there is the NTeX distro
around, the TeX live distro, the web2c basic distro. You don't ask to
prohibit them, do you?
> > This is a situation that free software should not find itself in.
>
> I agree there. (Without wanting to comment on the particular examples
> you gave).
Well, if enough people agree to work this out, *and* if Thomas Esser
is willing to split teTeX into free and non-free parts and the
management of teTeX allows this to happen in reasonable time, or maybe
this is already achieved at the basic CTAN organization level and
mirrored into the distros, this will all become academic.
> > Having a Unix TeX distribution that will try meeting this aim actively
> > will help towards this position.
>
> But currently the impression (and as far as I can see, the actual fact)
> is that this distribution helps the distributors (as they need not worry
> about legal issues) but does so at the expense of the tex users.
> When compared to say tetex it is not just the non free stuff that it is
> missed out; it is almost all of it. No `contrib' latex packages at all,
> not one. No hyperref, no natbib, no pstricks, no nothing.
Well, perhaps Debians efforts to start from teTeX and properly sort it
out will turn out more useful.
> Perhaps this is due to the `alpha' status of the release, but
> who knows, if they don't say what they are up to.
Yes, yes, yes.
I guess we can let this case rest until they tell us. Until then, no
use guessing and getting excited.
Timothy Murphy
>The author(s) of Web2c and texk have been active in the GNU TeX
>project from the beginning and have contributed many useful
>suggestions for its development.
As far as I can make out, GNU-TeX is an implementation of web2c (like teTeX).
It seems then that the binaries will be identical to teTeX,
except that some (unspecified) programs will be omitted
because they do not conform to FSF's definition of free.
Similarly the texmf tree will differ from that of teTeX
in that certain fonts will be omitted on the same grounds.
Is that a reasonable description of the product?
Timothy Murphy
>So you obviously *do* see that currently there is no distribution
>that, say, RedHat Linux could put on one of their Linux CD-ROMs (which
>they sell) without being liable to breach of copyright charges.
As a matter of interest, RedHat _does_ include tetex-0.9-6.i386.rpm
on its 5.2 distribution.
(Also tetex-latex-0.9-6.i386.rpm and various other tetex-related RPMs.)
Timothy Murphy
>If the final version
>comes out looking like this and linux distributions contain this as
>their tex distribution, what is a latex user to do, figure out which
>packages he needs and install them all individually, or just get tetex
>or texlive and do a single install command to replace the whole lot?
>It seems like the second choice will be preferable to most people.
I don't think there is any danger/chance of Linux distributions
containing GNU/TeX as opposed to tetex.
I think all distributions currently include tetex.
Certainly RedHat-5.2 contains a complete tetex distribution.
The chances of this being replaced by GNU/TeX are nil, I imagine.
Timothy Murphy
>>>>>> "Timothy" == Timothy Murphy <t...@maths.tcd.ie> writes:
>> Stallman and FSF tried to get the Linux community to use the term GNU/Linux
>> in place of Linux.
>Yes. The reason quite clearly was "to give credit where credit is due".
>I.e. to include the word "GNU" to make it clear that it critically depends
>on many tools developped with the help of the FSF.
I wasn't arguing whether this was good or bad;
you compared GNU/Linux with Linux,
and I simply pointed out that one of these terms
describes a computer operating system
while the other is only used in philosophical discussions.
>> My worry about GNU-TeX is that it sounds depressingly like the same scenario.
>> Hopefully I am wrong.
>Reading my paragraph above I feel quite confident to say you are indeed utterly
>wrong.
Probably.
What worried me was that you drew the parallel with GNU/Linux---Linux.
>so that the FSF can then support TeX.
Is this terribly important?
GNU is a Good Thing,
but FSF sometimes seems like a fundamentalist religion.
[It's a bit like waiting to see if the SWP
supports the Good Friday agreement in NI.]
David Carlisle
> and even he [me] quickly reverted to civilized discourse
don't sound so surprised:-)
David (C not K)
Timothy Murphy
>Hmm? Which part is gobbledygook? Are you saying that the above is
>*not* the difference according to the FSF?
I'm not privy to the thought-processes of the FSF,
but in my view the statement that
>> >GNU/Linux is a
>> >complete operating system built around this kernel and is called
>> >GNU/Linux even though the GPLed parts make up about 30% of it (but
>> >everything is freely redistributable).
is gobbledygook, since there is no operating system called GNU/Linux,
except as an alias for Linux.
>Granted, using "Linux"
>to refer to the whole operating system seems to me to be by far more
>common than "GNU/Linux", but both are out there.
Like, 10000 times as common.
Rodger Donaldson
<monnier+comp/text/tex/news/@tequila.cs.yale.edu> wrote:
>>>>>> "Timothy" == Timothy Murphy <t...@maths.tcd.ie> writes:
>> Stallman and FSF tried to get the Linux community to use the term GNU/Linux
>> in place of Linux.
>
>Yes. The reason quite clearly was "to give credit where credit is due".
>I.e. to include the word "GNU" to make it clear that it critically depends
>on many tools developped with the help of the FSF.
Of course, despite the fact that the HURD uses large swathes of Linux kernel
code to do anything useful, such as networking, the FSF have not seen fit to
apply these rules to themselves and referring to the Linux/HURD. Which make
their arguments look a little hypocritical.
--
Rodger Donaldson rod...@ihug.co.nz
"Windows is the answer, but only if the question was 'what is the
intellectual equivalent of being a galley slave?'"
--Larry Smith, in comp.os.linux.misc
Christopher Browne
wrote: n
>> Stallman and FSF tried to get the Linux community to use the term GNU/Linux
>> in place of Linux.
>> [Linus Torvalds gave a marvellous demonstration
>> of how to carry on a rational discussion with fundamentalists.]
>> GNU/Linux does not exist outside the overheated imaginations of FSF fanatics.
>
>The suggestion of using "lignux" was, quite appropriately in my
>opinion, nearly unanimously rejected. However I see "GNU/Linux"
>to refer to the whole operating system seems to me to be by far more
>common than "GNU/Linux", but both are out there.
On the one hand, the FSF "demand for recognition," as assortedly
expressed in the (happily soundly rejected) "lignux" debacle and, more
recently, with the promotion of "GNU/Linux," seems to me to be a
somewhat sad thing.
On the other hand, it *does* correctly point out that a whole lot of the
software used in "Linux systems" happens not to be either:
a) Only for use with the Linux kernel
(great gobs of it run on many "systems that resemble UNIX," and even
other places like OS/2, VMS, Win32, and even MVS...), or
b) Produced by people that might identify themselves as associated
with Linux.
TeX wasn't developed by "Linux folk." Nor was Perl, or Python, or GCC,
or GhostScript, or XFree86, or any other number of other things that
have been appropriated for use with "Linux systems." All may have "Linux
participants," but aren't "Linux" projects.
The FSF folk propose the use of the term "GNU System," due to the
similarities in intent and in licensing, and since there probably is
more "Linux software" that is associated with "GNU projects" than with
any other source of code. In effect, they suggest naming the system
"democratically" with votes being generated by where the code came from,
and since they win on the basis of plurality, they get dibs on the
name.
It's quite fair for people that use Linux to recognize that there are
many code sources to which gratefulness is owed. Unfortunately,
*demanding* gratefulness is not in terribly good taste, and the
perception of that demand causes folks to want to "push back."
--
"MS Windows is the computer equivalent of burger chains and bowling
lanes. It is software that "works" only if you lower your expectations
to a point where you have essentially buried them." -- Sam
cbbr...@hex.net- <http://www.ntlug.org/~cbbrowne/lsf.html>
David Kastrup
> David Kastrup <d...@mailhost.neuroinformatik.ruhr-uni-bochum.de> writes:
>
> >So you obviously *do* see that currently there is no distribution
> >that, say, RedHat Linux could put on one of their Linux CD-ROMs (which
> >they sell) without being liable to breach of copyright charges.
>
> As a matter of interest, RedHat _does_ include tetex-0.9-6.i386.rpm
> on its 5.2 distribution.
> (Also tetex-latex-0.9-6.i386.rpm and various other tetex-related RPMs.)
Yes, and if IBM wanted to sue them for it, they could.
David Kastrup
> Jay Belanger <j...@vh213601.truman.edu> writes:
>
> >Hmm? Which part is gobbledygook? Are you saying that the above is
> >*not* the difference according to the FSF?
>
> I'm not privy to the thought-processes of the FSF,
> but in my view the statement that
>
> >> >GNU/Linux is a
> >> >complete operating system built around this kernel and is called
> >> >GNU/Linux even though the GPLed parts make up about 30% of it (but
> >> >everything is freely redistributable).
>
> is gobbledygook, since there is no operating system called GNU/Linux,
> except as an alias for Linux.
Please be more careful in your namecalling, and please refrain from
creative editing in order to make me appear like an idiot. The
paragraph in question was:
> >Do you know what the difference between GNU/Linux and Linux is
> >according to the FSF? Linux is just a kernel and is called just
> >Linux, even though it is GPLed in its entirety. GNU/Linux is a
> >complete operating system built around this kernel and is called
> >GNU/Linux even though the GPLed parts make up about 30% of it (but
> >everything is freely redistributable). This is because something like
> >GNU/Linux was already concocted by them before the time of Linux
> >Torvalds, and he took all the stuff and gave it a kernel to revolve
> >around.
So, I *did* state the view according to the FSF, not my own. And the
*facts* on which the FSF bases their demand that the system be called
GNU/Linux are pretty much correct (although dropping the kernel in
sounds too simple. A lot of system utilities specific to this kernel
also needed to be created and/or adapted).
So one can hardly call this gobbledygook. It *was* the FSF that has
set out to make a free system like Linux, and it *was* the FSF that
has in painstaking work done a lot of crucial work (like writing gcc
and emacs and organizing and doing quite a lot of other work, lobbying
for X11 to be released under a free license and so on and so on) that
made the Linux kernel a useful thing in the first place.
Without the GNU project to work on, Linux would have mostly stayed a
toy project printing letters to a terminal.
These are *facts*, not speculation. We would all be using FreeBSD, at
least we would be using it if it not as well relied on work of the
FSF.
If the FSF had restrained themselves to pointing this out, and
praising Linux as a truly free GNU system, nobody would have
objected. Indeed, people more likely than not would have been proud.
In a tradition of doing the most contraproductive and confrontational
public relation step possible, the FSF however decided to unilaterally
change the name of something which was not their responsibility, even
if mostly assembled from parts they provided for this purpose.
*This* deserves your criticism, but not the facts behind it.
And to call my presentation of the FSF's view gobbledygook is impolite
if you can offer no better summary.
Robert Kiesling
"GNU Distribution of TeX, LaTeX, AMSTeX, and PDFTeX," or some
shortened form, would be more appropriate.
By "LaTeX 2e," I mean that portion of LaTeX 2e which is documented in
the source2e compilation. That seems to me the minimum LaTeX 2e code
required to properly run a LaTeX 2e.
I also apologize for the strident and political tone of this
statement, but I can't put it any more succinctly:
1. LaTeX 2e is too important of a software system not to be included
with GNU software;
2. The GNU GPL, and licenses similar to it, have proven successful
in facilitating the distribution of advanced software to the
greatest number of people with the minimum expense and restriction;
3. A significant part of currently extant LaTeX 2e code is licensed
in a manner that is too restrictive to be consistent with the GPL;
4. That portion of the currently extant LaTeX 2e code necessary to
operate a minimal LaTeX 2e installation is licensed in a manner
consistent with the GPL. Therefore;
5. The GNU Project cannot maintain and distribute a significant part
of the LaTeX 2e code in existence without compromising its integrity;
6. The GNU Project can, consistent with its principles, maintain and
distribute that part of the extant LaTeX 2e code that is necessary
to operate a minimal LaTeX 2e installation; and,
7. The GNU Project believes that maintaining and distributing a LaTeX 2e
under a license consistent with the GPL is of benefit to the LaTeX
community and the GNU Project.
--
Robert Kiesling
kies...@ix.netcom.com
Steven Blunt
(Christopher Browne) wrote:
>The FSF folk propose the use of the term "GNU System," due to the
>similarities in intent and in licensing,
That's the problem I have with GNU/Linux. Fair enough for Debian(1)
perhaps, but GNU is supposed to be a completely free system, so GNU/Linux
shouldn't really apply to say Suse.
(1) Debian probably has more non-free software that any other distro, but
they claim moral superiority by putting it on a separate CD, perhaps GNU
shouldn't apply to them either.
cya
--
Steven Blunt
spb...@ozemail.com.au
http://enterfornone.simplenet.com/
Antti-Juhani Kaijanaho
lately than following c.t.t.]
David Carlisle <dav...@nag.co.uk> writes:
> working to get tetex modularised in such a way that free and non
> free files may be easily separated would seem like a far more useful
> way for anyone to spend the time
The Debian packaging of teTeX has done this. We have a tetex-nonfree
package (in the non-free section of the Debian distribution), which
includes all the packages that have DFSG-incompatible licenses.
However, we do occasionally find (by chance) non-free files in the
free packages, which is not good - technically, we've been breaking
their licenses all this time, as the main section (including the
supposedly free teTeX packages) gets burned to CDs which are sold for
profit by others.
So I personally applaud the idea of a GNU TeX distribution, since I
know the GNU project is pedantic about licensing (which cannot be said
about most other software packagers), technical debates aside. Be it
a GNUified teTeX, teTeX proper with greater care on licensing issues,
or a brand new GNU TeX, I don't really have a preference, as long as
the licensing problems *are* resolved and the result is a working
system.
Antti-Juhani
a Debian developer - not the Debian tetex maintainer, though
- and speaking for myself only
--
%%% Antti-Juhani Kaijanaho % ga...@iki.fi % http://www.iki.fi/gaia/ %%%
EMACS, n.: Emacs May Allow Customised Screwups
(unknown origin)
Antti-Juhani Kaijanaho
got all the articles. If this message raises issues that have been
hashed out in other messages, please accept my apololgies.]
David Carlisle <da...@dcarlisle.demon.co.uk> writes:
> Calling it a gnu TeX distribution gives the false impression that it
> is all under gnu license.
This is not true (I'd call it FUD if it didn't come from David
Carlisle whom I've learned to admire). There are several GNU packages
with licenses that are not the GNU GPL or the GNU LGPL (flex comes
first to mind). If someone has the impression that GNU is GPL is GNU,
then he or she should learn the facts.
The goal of the GNU project is to produce a complete free system.
Most (but not all) of the original GNU software is under the GPL or
the LGPL, but the system includes much third-party software (such as X
and TeX), most of which is *not* under the GPL or the LGPL.
The "GNU" in the name of the GNU TeX distribution implies that it was
made by or for the GNU project. It does not imply anything about
licensing (apart from the fact that it will be free in its entirety).
> Until these points are clarified, please do not include my packages
> in the distribution.
This is most unfortunate. Perhaps you'd reconsider?
Antti-Juhani
who has no connection to the GNU TeX effort
David Carlisle
> Be it
> a GNUified teTeX, teTeX proper with greater care on licensing issues,
> or a brand new GNU TeX, I don't really have a preference, as long as
> the licensing problems *are* resolved
But it turns out that gnutex is not like tetex at all. It is just
absolutely minimal core tex and latex distrib. So unless you only want
to write in English, in `Lamport Book' latex with no extensions, it is
essentially useless. People get so used to arguing about GPL it is easy
to assume that that is the problem, because of the `gnu' in gnutex. That
is a side issue here. The main issue is that this distribution tries to
undo all the work done over the last few years in ensuring that users
get a reasonable system `out of the box' and go back to the system where
you basically have to have good connectivity and know where to look
before you can get the pieces off the archives to process a typical document.
> and the result is a working system.
There is the rub. The only simple advice one could give to get from an
installed gnutex to a working system would be to delete gnutex and
install tetex. So, what's the point?
David
Robin Fairbairns
>[I'm a bit late with this thread, sorry. I've had other things to do
>lately than following c.t.t.]
>
>David Carlisle <dav...@nag.co.uk> writes:
>
>> working to get tetex modularised in such a way that free and non
>> free files may be easily separated would seem like a far more useful
>> way for anyone to spend the time
>
>The Debian packaging of teTeX has done this. We have a tetex-nonfree
>package (in the non-free section of the Debian distribution), which
>includes all the packages that have DFSG-incompatible licenses.
>However, we do occasionally find (by chance) non-free files in the
>free packages, which is not good - technically, we've been breaking
>their licenses all this time, as the main section (including the
>supposedly free teTeX packages) gets burned to CDs which are sold for
>profit by others.
so you've started a good piece of work, but haven't apparently
finished it yet.
>So I personally applaud the idea of a GNU TeX distribution, since I
>know the GNU project is pedantic about licensing (which cannot be said
>about most other software packagers), technical debates aside. Be it
>a GNUified teTeX, teTeX proper with greater care on licensing issues,
>or a brand new GNU TeX, I don't really have a preference, as long as
>the licensing problems *are* resolved and the result is a working
>system.
this takes the breath away, a bit. there's been a start made on the
work by debian, but you're happy that all that should be thrown away
and a complete new effort be started? am i misunderstanding you? why
wouldn't debian be eager to have their work extended and improved?
i can understand this point about free vs. non-free, though i abhor
the gpl as much as i abhor iso 9001 (because they both impose
transitive closure on things they affect). what i _can't_ understand
is the apparent desire to take wing and fly out of the window because
what's there just now isn't quite right.
if thomas esser had taken this attitude, he'ld have written an entire
new web2c system -- in fact, he wrote some modifications and fed them
back into the web2c development process. what's so magical about
distributions that this isn't the way to proceed with them?
--
Robin Fairbairns, Cambridge
David Carlisle
> correct comments about `GNU/GPL' deleted.
true, I wouldn't have moaned about the name if I wasn't cross with them
generally. I don't claim that Thomas Esser is wrong to call it tetex or
similarly EM TeX or FP TeX etc. I only really moaned about GNUTeX
as they made such a fuss over gnu/linux, wanting to get visible credit.
well, if they want to have gnu prominent when GNU has done a large part
of the work, they should make it less prominent when most of the stuff
is not gnu. But I don't really care. gnutex is a good enough name I
suppose.
> This is most unfortunate. Perhaps you'd reconsider?
well I did not (still do not) really expect that to be the final
outcome. But there are (as this thread shows) several issues to be
sorted out, and if me saying they can't distribute my packages (yet)
means that this distribution will not find itself into every linux
distribution until these issues _are_ looked at, then so be it.
I do hope things get sorted out. Despite the fact that I have
been moaning in this group, Robert Kiesling and I have been having a
friendly enough email correspondence in parallel. Actually I don't see
any evidence that the GNU project understands yet why they have messed
this up so badly, but, live in hope....
David
Robert Kiesling
They are in the distribution. The Metafont sources are, of course,
modifiable, and requesting that the name be changed if the designs are
modified is the same licensing constraint that much of the LaTeX 2e
code has.
--
Robert Kiesling
kies...@ix.netcom.com
Timothy Murphy
>2. The GNU GPL, and licenses similar to it, have proven successful
> in facilitating the distribution of advanced software to the
> greatest number of people with the minimum expense and restriction;
It doesn't seem to me to have been any more (or less) successful
than TeX or Linux.
I realise this is probably an impeachable offence,
but I never read licences (do Americans spell that wrongly?)
-- life is too short.
I wonder if I am in a very small minority?
I knew DEC was going down the tubes
when someone told me they employed more lawyers than programmers.
r d s.q u e e n s u.c a
>They are in the distribution. The Metafont sources are, of course,
>modifiable, and requesting that the name be changed if the designs are
>modified is the same licensing constraint that much of the LaTeX 2e
>code has.
It would be nice if you could list what *isn't* acceptable to GNU and
why. Perhaps some authors would modify their licensing if they
realized there was a problem with it.
Bob T.
Patrick TJ McPhee
Antti-Juhani Kaijanaho <gaia...@iki.fi> wrote:
% Carlisle whom I've learned to admire). There are several GNU packages
% with licenses that are not the GNU GPL or the GNU LGPL (flex comes
We've gone into another stupid gpl discussion, but the author of flex
used to throw fits when people referred to flex as a GNU package. I
think he felt that FSF was trying to claim credit for work that had
nothing to do with them, and I think he was right. I assume that nothing
has happened to change his opinion and that people who call flex GNU flex
are doing so against the express wishes of the author.
% The "GNU" in the name of the GNU TeX distribution implies that it was
% made by or for the GNU project. It does not imply anything about
% licensing (apart from the fact that it will be free in its entirety).
I agree with Carlisle. When you put your name on something, you're
taking credit for it.
--
Patrick TJ McPhee
East York Canada
pt...@interlog.com
Robert Kiesling
Thanks for asking. I could make an exhaustive list of all of the
LaTeX packages, but then some people would feel singled out,
inevitably, I think. Rather than draw unwanted attention to any of
the authors, it would probably be better if they approached us.
The distinction that we've been talking about is that some of the
LaTeX code is freely re-distributable, whether modified or not, and
some of it isn't. The only request the LaTeX 2e licensing agreement
makes wrt/redistributing the files is that any modified versions use
different file names than the original versions. So even though LaTeX
2e is not licensed under the GPL, its licensing terms are very
similar, with only the file name constraint, which, I would think, is
there to improve maintainability.
--
Robert Kiesling
kies...@ix.netcom.com
Antti-Juhani Kaijanaho
the project (actually, none of us do most of the time). In this
message and in all my previous messages I have been expressing my
*presonal* opinion.
r...@cl.cam.ac.uk (Robin Fairbairns) writes:
> so you've started a good piece of work, but haven't apparently
> finished it yet.
You're right. I guess a complete audit of the copyrights in teTeX
would be in order, but I don't know who has the time and energy to do
that (it is a large and above all tedious job). It would probably
have been easier if the copyrights would have been sorted out in the
beginning - when the distro was being assembled in the first place.
> there's been a start made on the work by debian, but you're happy
> that all that should be thrown away and a complete new effort be
> started? am i misunderstanding you?
You are, in part. I'm not proposing to throw away teTeX at this time.
I was thinking of a finished GNU product, not of a beta 0.xx version.
I prefer the GNU project since I know they can be trusted to have
license issues right the first time. But - and I did mention this - I
don't want a crippled system. If the GNU TeX distribution is a
stripped-down TeX system, I don't want it.
Note my words in the previous message: "technical issues aside". My
comments on GNU TeX versus teTeX re licensing was under the assumption
they will become technically equivalent some time in the future.
> why wouldn't debian be eager to have their work extended and
> improved?
(Note that I'm not speaking for the Debian project as an entity.)
IMHO, if we face a situation where we have teTeX, where we
occasionally find license violations, and a more-or-less equivalent
system from the GNU project, my preference would be on the GNU
version. It is shame to throw away good work, but if in return we
reduce our future workload, I think it would be worth it.
The Debian project did this once (it was before my time, BTW): the
project switched from (what I believe to be) a roll-your-own Web2C
system to teTeX since the latter was easier to maintain, regardless of
the fact that much work had been invested in the Debian packaging of
Web2C.
But, as David Carlisle pointed out to me, the GNU TeX distribution is
not (apparently) intended to become a teTeX equivalent. So all this
talk is essentially pointless except as an intellectual exercise.
> though i abhor the gpl as much as i abhor iso 9001 (because they
> both impose transitive closure on things they affect).
I don't know what ISO 9001 is. However, if you can point to me a
license that fulfills the three requirements below, I'd be happy to
consider it as a replacement for the GNU GPL in my own code (including
the one GPL'd thing I have submitted to CTAN, plari).
1) It must satisfy the Debian Free Software Guidelines (DFSG).
2) All derivative works must have a license that satisfies the DFSG.
3) It must be compatible with the GPL.
These are my personal preferences. I cannot think of any other
license than GPL/LGPL which satisfies these requirements.
Rationale for the requirements:
1) I want to be able to distribute my software as part of Debian.
2) I do not want to support proprietary software development,
neither directly nor indirectly.
3) I want to use the big GPL'd codebase in many of my projects.
> what i _can't_ understand is the apparent desire to take wing and
> fly out of the window because what's there just now isn't quite
> right.
I was working from the assumption that the GNU project does what it
does regardless of what we write here in ctt; and my comments assumed
a production-quality GNU TeX (as he comparison system, teTeX 0.9,
although nominally beta, is of production quality).
Antti-Juhani
David Carlisle
> reasonable points about debian linux tetex removed.
But there is a big difference between the debian linux tetex and this
gnutex distribution.
The first aims to solve the problem by working with authors to change
their licenses, or if necessary ommit packages (The wording of the
default header of docstrip in the base latex distribution changed last
year after input from that project).
So the outcome is I assume something that fits well in a GNU system and
is useable by the TeX user.
So everyone applauds that effort (and probably there is a real chance
something like that can be merged back into the original tetex/texlive
so the work of sorting out does not need to be done afresh every
release)
However gnutex aproaches the problem by simply making a distribution
that is unusable. _This_ is the main objection to it. It is OK for the
GNU project to say they need gnutex because it only contains files
with licenses they are happy with, but what is the point of them
distributing it if it can not process even a tiny fraction of tex
documents, and is so cut down it is in contravention of the
distribution conditions on the base latex release which say that all
documentation must be included.
David
David Carlisle
> I realise this is probably an impeachable offence,
> but I never read licences (do Americans spell that wrongly?)
> -- life is too short.
> I wonder if I am in a very small minority?
well certainly you are in company with the creator of gnutex, who never
read the latex license.
David
David Kastrup
> kies...@ix.netcom.com (Robert Kiesling) writes:
>
> >2. The GNU GPL, and licenses similar to it, have proven successful
> > in facilitating the distribution of advanced software to the
> > greatest number of people with the minimum expense and restriction;
>
> It doesn't seem to me to have been any more (or less) successful
> than TeX or Linux.
TeX is not licensed at all, but Public Domain. The Linux kernel is
licensed under the GNU GPL.
David Kastrup
> I agree with Carlisle. When you put your name on something, you're
> taking credit for it.
So you would say that a car manufacturer should not be allowed to put
his name on a car because it includes parts manufactured by others,
like Bosch? No, it's the architect of the assembly that determines its
name.
For that reason I call Linux Linux (even though it includes a large
number of GNU components), and I find nothing wrong with the GNU
project rolling their own distribution which they call GNU-TeX or even
their own Linux distribution which they call GNU/Linux. After all, I
have not complained about Thomas Esser rolling a distribution called
teTeX, either.
Antti-Juhani Kaijanaho
> I agree with Carlisle. When you put your name on something, you're
> taking credit for it.
Credit has nothing to do with licensing - so you are not addressing my
point nor (what I believe was) Carlisle's. And as to /your/ point: Is
Red Hat taking credit for Linux? No, I'd say: when you put your name
on something, you're assuming /responsibility/ for it; this is quite
different.
We're way off topic now. Better move elsewhere.
r d s.q u e e n s u.c a
>In article <788rkc$1dn$2...@knot.queensu.ca>,
>Bob Tennent <r d t@c s.q u e e n s u.c a> wrote:
>>On Thu, 21 Jan 1999 23:53:54 GMT, Robert Kiesling wrote:
>>
>> >They are in the distribution. The Metafont sources are, of course,
>> >modifiable, and requesting that the name be changed if the designs are
>> >modified is the same licensing constraint that much of the LaTeX 2e
>> >code has.
>>
>>It would be nice if you could list what *isn't* acceptable to GNU and
>>why. Perhaps some authors would modify their licensing if they
>>realized there was a problem with it.
>
>Thanks for asking. I could make an exhaustive list of all of the
>LaTeX packages, but then some people would feel singled out,
>inevitably, I think. Rather than draw unwanted attention to any of
>the authors, it would probably be better if they approached us.
>
that upset me: it claimed lilypond was free and musixtex was not.
When I queried what it was about musixtex that wasn't acceptably free,
it turned out that no-one had really bothered to read the license;
they had just contacted the author, who rebuffed their no doubt heavy-handed
overtures because he wasn't interested in their concerns, and the GNU people
concluded that musixtex wasn't free.
I don't know if this is typical but perhaps it wouldn't hurt to be open
about exactly what you people think is a problem. Perhaps you're the problem.
Bob T.
Robin Fairbairns
>A few months ago I saw something on the lilypond site (the GNU music
>typesetter) that upset me: it claimed lilypond was free and musixtex
>
>I don't know if this is typical but perhaps it wouldn't hurt to be
>open about exactly what you people think is a problem. Perhaps
>you're the problem.
*nothing* about musixtex is typical. it's entirely possible that the
lilypond people have reasons even i would consider acceptable for
their judgement.
[i've mailed bob separately, in more detail]
--
Robin Fairbairns, Cambridge
Sven Utcke
> In article <788rkc$1dn$2...@knot.queensu.ca>,
> Bob Tennent <r d t@c s.q u e e n s u.c a> wrote:
> >On Thu, 21 Jan 1999 23:53:54 GMT, Robert Kiesling wrote:
> >
> > >They are in the distribution. The Metafont sources are, of course,
> > >modifiable, and requesting that the name be changed if the designs are
> > >modified is the same licensing constraint that much of the LaTeX 2e
> > >code has.
> >
> >It would be nice if you could list what *isn't* acceptable to GNU and
> >why. Perhaps some authors would modify their licensing if they
> >realized there was a problem with it.
>
> Thanks for asking. I could make an exhaustive list of all of the
> LaTeX packages, but then some people would feel singled out,
> inevitably, I think. Rather than draw unwanted attention to any of
> the authors, it would probably be better if they approached us.
Probably not a good idea. Most authors might not even realise that
their packages aren't free, especially gven the FSF's somewhat special
interpretation of the term free.
Sven
--
_ _ Lehrstuhl fuer Mustererkennung und Bildverarbeitung
| |_ __ | |__ Sven Utcke
| | ' \| '_ \ phone: +49 761 203 8274 Am Flughafen 17
|_|_|_|_|_.__/ fax : +49 761 203 8262 79110 Freiburg i. Brsg.
mailto:ut...@informatik.uni-freiburg.de www.informatik.uni-freiburg.de/~utcke
Aaron M. Renn
> The FSF folk propose the use of the term "GNU System," due to the
> more "Linux software" that is associated with "GNU projects" than with
> any other source of code. In effect, they suggest naming the system
> "democratically" with votes being generated by where the code came from,
> and since they win on the basis of plurality, they get dibs on the
> name.
That is completely wrong.
The GNU project was started with the goal of creating a 100% free software
operating system. An operating system is made up of many components: the
kernel; compilers, debuggers, linkers, libraries, and other development
tools; editors and other utilities; windowing systems; etc. Every one of
these components needed to have a free version in order to have a free
operating system. Since the GNU project was interested in a free operating
system, not in writing each in every component itself, it adopted whatever
free components it could find that already existed, or were written during
system development. This included TeX, the X Window System, Perl, etc. It
then set out to write the rest. This included many unglamourous programs
that otherwise might never have gotten a free version, such as m4, the C
library, echo, etc. Each component filled in a gap that brought the vision
of a totally free operating system closer to reality. As it turned out, the
final component (the kernel) needed to have a free operating system was
written by Linus Torvalds, not by GNU people. But nevertheless, this act
completed the free operating system that was envisioned by the Richard
Stallman back in 1983. Without that vision, there would be no "Linux"
system today. Whereas without any other particular component, a free
operating system would still come to be because the GNU project was making
one. Any given component would eventually get built. Who wrote component X,
Y, or Z is not important. What is important is the overall vision to make
the product (a free operating system) a reality.
If I use GNU MP and Bison to create a desktop calculator, I won't call it
"MP/Bison/RennCalc" even if a majority of the code is GNU components.
Instead I would just call it "RennCalc" because it was my vision to create
the desktop calculator. That requires many components including an input
parser, a calculation engine, a display module, etc. If I find some of
these already written for me, that is just so much the better because it
lets me get my calculator done faster, but it is still my vision of the
calculator that made it a reality. (I would probably still give credit to
the people who wrote major components of the calculator, but wouldn't name
the whole calculator after them).
Similarly, the reason to call it the GNU/Linux system is that it was the GNU
project's vision of a free operating system that make that free operating
system possible. How many components of it were created by the GNU project
or anyone else is irrelevant.
--
Aaron M. Renn (ar...@urbanophile.com) http://www.urbanophile.com/arenn/
Simon Cozens
> They are in the distribution. The Metafont sources are, of course,
> modifiable, and requesting that the name be changed if the designs are
> modified is the same licensing constraint that much of the LaTeX 2e
> code has.
Oh, God, no. *Please* tell me this isn't what's being planned.
LaTeXGNUe? CompGNUter Modern? Blegh.
--
Intel CPUs are not defective, they just act that way.
-- Henry Spencer
Peter Seebach
Aaron M. Renn <ar...@urbanophile.com> wrote:
>If I use GNU MP and Bison to create a desktop calculator, I won't call it
>"MP/Bison/RennCalc" even if a majority of the code is GNU components.
>Instead I would just call it "RennCalc" because it was my vision to create
>the desktop calculator.
But what if the person who developed GNU MP was planning to develop a
desktop calculator, and maybe was already working on it? Does the fact that
the other developer had that vision make your calculator "GNU/RennCalc"?
If not, then why does the fact that the FSF had plans for an operating system
make a system that someone else envisioned and built, using some of their
components, "GNU/foo"?
>Similarly, the reason to call it the GNU/Linux system is that it was the GNU
>project's vision of a free operating system that make that free operating
>system possible. How many components of it were created by the GNU project
>or anyone else is irrelevant.
Except that, so far as I can tell, it was Linus's vision of writing an
operating system (using components from wherever he found them) that made
Linux happen. So it should be called Linux.
Or perhaps BSD/Linux, since it was the vision of the BSD engineers that
they could use TCP/IP routines as a fundemental part of an operating system,
so, an operating system following "their vision" (i.e., incorporating TCP/IP)
should be called BSD/foo.
I don't buy it. The FSF didn't do Linux. Linux would have come about whether
the FSF's ultimate goal was a good toolchain or an operating system, because
it uses the toolchain and utilities. FreeBSD uses a comparable amount of FSF
code, but it's not GNU/FreeBSD.
Essentially, the problem is, you can't wait until someone fulfills your goals
and then say "ah-hah! That's my goal, therefore, I can put my name on it."
Now, GNU Hurd is a GNU operating system. But Linux is a Linus Torvalds
operating system, with components from wherever they're handy.
-s
--
Copyright 1999, All rights reserved. Peter Seebach / se...@plethora.net
C/Unix wizard, Pro-commerce radical, Spam fighter. Boycott Spamazon!
Send me money - get cool programs and hardware! No commuting, please.
Visit my new ISP <URL:http://www.plethora.net/> --- More Net, Less Spam!
Richard Kaszeta
> >Thanks for asking. I could make an exhaustive list of all of the
> >LaTeX packages, but then some people would feel singled out,
> >inevitably, I think. Rather than draw unwanted attention to any of
> >the authors, it would probably be better if they approached us.
> >
> A few months ago I saw something on the lilypond site (the GNU music typesetter)
> that upset me: it claimed lilypond was free and musixtex was not.
> When I queried what it was about musixtex that wasn't acceptably free,
> it turned out that no-one had really bothered to read the license;
> they had just contacted the author, who rebuffed their no doubt heavy-handed
> overtures because he wasn't interested in their concerns, and the GNU people
> concluded that musixtex wasn't free.
>
> I don't know if this is typical but perhaps it wouldn't hurt to be open
> about exactly what you people think is a problem. Perhaps you're the problem.
Indeed, how you approach an author greatly influences how they
respond. Last year I spent a week going through most of teTeX's texmf
tree (aside from the fonts), and identifying the license for each
package, and identifying a number of packages which presented
distribution problems for teTeX (i.e. can't include it on a cdrom,
can't distribute the .sty file[1] at all, non-commercial use only, no
license statement whatsoever) or for people like linux distribution
developers that need to know which portions of teTeX they can and
cannot discover. Sadly, a significant portion of teTeX fell into this
category, enough that I took it upon myself to contact all the authors
I could, explain to them what I believed their license to be, what I
perceived to be difficulties with the license, and request that they
consider using a more open license that would allow redistribution.
I just made sure I was polite, non-confrontational, and willing to
acknowledge that they the decision was their's and I wasn't trying to
coerce them[2].
Of authors that responded to my mail, *every* one of them expressed
support of free software, and more than a dozen authors uploaded new
versions of their packages with less-restrictive licenses.
[1] There are still a lot of .sty files which used the boilerplate
license from older version of docstrip which explicity says "you are
not allowed to distribute this file", making projects like the TeX
Live CD hard to implement legally. Thankfully, once this was pointed
out to the LaTeX3 team, the default license info inserted by docstrip
has been changed.
[2] I know a handful of authors of fairly prominent open source
software packages. One of programs is released under the GPL and is
owned by him... despite being GPL various folks keep bugging him to
turn it over to the FSF. The other uses another prominent open
source non-GPL license and regularly gets abusive email about not
being GPL. Needless to say, this pisses them off...
--
Richard W Kaszeta Graduate Student/Sysadmin
bo...@me.umn.edu University of MN, ME Dept
http://www.menet.umn.edu/~kaszeta
Rob Partington
> FreeBSD uses a comparable amount of FSF
>code, but it's not GNU/FreeBSD.
I thought the point about FreeBSD was that it didn't use anywhere
near as much GNU software as Linux did and that it is possible to
have FreeBSD without any GNU software and still have a working system?
>Now, GNU Hurd is a GNU operating system. But Linux is a Linus Torvalds
>operating system, with components from wherever they're handy.
Aren't they both just kernels, and not operating systems?
--
rob partington personal: <URL: mailto:r...@browser.org >
Aaron M. Renn
> Essentially, the problem is, you can't wait until someone fulfills your goals
> and then say "ah-hah! That's my goal, therefore, I can put my name on it."
Did you just beam into this planet? The publicly stated goal of the GNU
project has always been to create a free, Unix compatible operating system.
I suggest you try reading the GNU manifesto. Nobody suddenly jumped up with
the goal of writing a free operating system after it became a reality.
Peter Seebach
Rob Partington <r...@browser.org> wrote:
>On Fri, 22 Jan 1999 16:55:31 GMT, Peter Seebach <se...@plethora.net> wrote:
>> FreeBSD uses a comparable amount of FSF
>>code, but it's not GNU/FreeBSD.
>I thought the point about FreeBSD was that it didn't use anywhere
>near as much GNU software as Linux did and that it is possible to
>have FreeBSD without any GNU software and still have a working system?
No; the big point is that the *KERNEL* has no GPL'd code. Ditto for the
other *BSD's. They still use gcc or egcs, etc etc.
>>Now, GNU Hurd is a GNU operating system. But Linux is a Linus Torvalds
>>operating system, with components from wherever they're handy.
>Aren't they both just kernels, and not operating systems?
Based on common usage, I'm referring to "Linux" as the thing you get on a
CD from one of however many vendors. It includes an X distribution, it
includes all sorts of stuff. Distinct from "the Linux kernel".
The point being, that's a system you can use because Linus had a vision of
having a complete O/S that was free software.
Aaron M. Renn
> > FreeBSD uses a comparable amount of FSF
> >code, but it's not GNU/FreeBSD.
>
> I thought the point about FreeBSD was that it didn't use anywhere
> near as much GNU software as Linux did and that it is possible to
> have FreeBSD without any GNU software and still have a working system?
The point about FreeBSD is that is was never the free operating system the
GNU project set out to build. FreeBSD grew out of a desire by different
people to take the existing BSD system and get it released as free software,
which is what they did. This required replacing some components (such as
the compiler), which the FreeBSD project did. They do use GNU components
and I presume other non-BSD software such as perl. But that doesn't change
where FreeBSD came from, which is someplace completely different from GNU.
This was a different project with a different vision and goal.
Aaron M. Renn
> The point being, that's a system you can use because Linus had a vision of
> having a complete O/S that was free software.
Could you please provide me with a reference to back up that statement?
From what I have read, Linus started writing his kernel to get around
limitations in Minix. To the best of my knoweldge, Linus has never created
or distributed a full operating system. Nor has he ever stated a desire to
be in that business.
Peter Seebach
>GNU project set out to build. FreeBSD grew out of a desire by different
>people to take the existing BSD system and get it released as free software,
>which is what they did. This required replacing some components (such as
>the compiler), which the FreeBSD project did. They do use GNU components
>and I presume other non-BSD software such as perl. But that doesn't change
>where FreeBSD came from, which is someplace completely different from GNU.
>This was a different project with a different vision and goal.
And Linux came from someplace completely different than GNU.
I don't buy the argument that if what I envision and build happens to have
been thought of by someone else, they can stick their name on it.
Especially since Hurd is so much different from Linux.
Peter Seebach
>> The point being, that's a system you can use because Linus had a vision of
>> having a complete O/S that was free software.
>Could you please provide me with a reference to back up that statement?
>From what I have read, Linus started writing his kernel to get around
>limitations in Minix. To the best of my knoweldge, Linus has never created
>or distributed a full operating system. Nor has he ever stated a desire to
>be in that business.
He has created a project which can be found as a distribution. The GNU
project didn't make Red Hat any more than Linus did - but Linus was the one
who made Red Hat possible.
Yes, GNU makes them all possible - but it also makes NetBSD possible, and it
makes NeXTStep possible. gcc is a very powerful tool. But it doesn't make
the O/S using it a GNU operating system, and it's not reasonable to
retroactively claim that this was the GNU operating system all along.
Greg Lindahl
> But what if the person who developed GNU MP was planning to develop a
> desktop calculator, and maybe was already working on it? Does the fact that
> the other developer had that vision make your calculator "GNU/RennCalc"?
Uh, seebs, can you please stop repeating old flamewars? Maybe this is
new for comp.lang.tex, but this went around about 5 times on
gnu.misc.discuss already, at great length. I suspect that most of the
readers of comp.text.tex just don't give a crap, either. And even
Linus asked people to stop flaming about GNU/Linux. So shut up already.
-- g
Peter Seebach
Sorry... I just saw an argument I found particularly flawed, and felt the
need to debate it. I don't really care that much about the terminology - I
just object to that rationalization for it.
Greg Lindahl
> Sorry... I just saw an argument I found particularly flawed, and felt the
> need to debate it. I don't really care that much about the terminology - I
> just object to that rationalization for it.
The best part is that you are, at the same time:
1) complaining that rms is posting off-topic material to gnu.misc.discuss
and
2) continuing to cross-post this flamewar to comp.text.tex, when it no
longer mentions TeX in any way, shape, or form.
-- g
Peter Flynn
> I realise this is probably an impeachable offence,
> but I never read licences (do Americans spell that wrongly?)
I think to be impeached you need to "reach a much higher standard
of dishonesty" as someone once said...
> I knew DEC was going down the tubes
> when someone told me they employed more lawyers than programmers.
There's hope for Microsoft, then...
///Peter
--
DTDs are not common knowledge because programming students are not
taught markup. A markup language is not a programming language.
Timothy Murphy
>TeX is not licensed at all, but Public Domain.
Is this true?
Can a program be copyright and PD?
--
Timothy Murphy
e-mail: t...@maths.tcd.ie
tel: +353-1-2842366
s-mail: School of Mathematics, Trinity College, Dublin 2, Ireland
David Kastrup
> In article <36A8B800...@urbanophile.com>,
> Aaron M. Renn <ar...@urbanophile.com> wrote:
> >Peter Seebach wrote:
> >> The point being, that's a system you can use because Linus had a vision of
> >> having a complete O/S that was free software.
>
> >Could you please provide me with a reference to back up that statement?
> >From what I have read, Linus started writing his kernel to get around
> >limitations in Minix. To the best of my knoweldge, Linus has never created
> >or distributed a full operating system. Nor has he ever stated a desire to
> >be in that business.
>
> He has created a project which can be found as a distribution. The GNU
> project didn't make Red Hat any more than Linus did - but Linus was the one
> who made Red Hat possible.
Excuse me, but in this case you are really belittling the importance
of the GNU project. *All* the standard tools you type to and interact
with on a Linux system are GNU utilities. The text utilities, the
file utilities, practically all of the development tools. Not to
speak of the things (like X11) that became freely available due to FSF
lobbying.
Without the GNU project, Linux would be nothing. Of course, without
the Linux kernel, the situation would be quite the same, even though
we might in a few years see the first Hurd-based systems come along.
More likely, we would be using FreeBSD.
Without the GNU project, we would not be using a free system at all.
Without the Linux kernel, we would by now have a different free system
(alternatives are pretty visible by now), although probably the free
software movement would be 10 years behind. Yes, I know that Linux
itself exists just 7 years and some, but it has given a lot of impetus
to other projects as well.
David Kastrup
> David Kastrup <d...@mailhost.neuroinformatik.ruhr-uni-bochum.de> writes:
>
> >TeX is not licensed at all, but Public Domain.
>
> Is this true?
> Can a program be copyright and PD?
Of course an author can reliquish his copyright. Which Knuth did.
Just read "TeX, the Program".
So of course every WP producer could have just taken TeX's typesetting
code legally and incorporate it in his software. There is no reason
things like Word need to be as crappy as they still are.
The use of the *name* TeX, however, is restricted by a trademark AMS
carries.