Re: gnuplot (was: Re: zsh outside of Emacs)

Re: gnuplot (was: Re: zsh outside of Emacs)

[Emanuel Berg]

Robert Thorpe <rt@robertthorpeconsulting.com> writes:

>> for fast ASCII-only checks. Speaking of gnuplot, I
>> had to to get gnuplot-mode from MELPA - that was
>> super easy, but why isn't such a basic mode
>> included in vanilla Emacs, to offer font lock,
>> indentation, and invocation? For the few of you who
>> don't know gnuplot
>
> Perhaps because the authors haven't copyright
> assigned it to the FSF? The name "gnuplot" is a
> coincidence, the program and it's authors have
> nothing to do with the GNU project. That's not to
> say it's not a great program, it is.

Do you mean gnuplot-mode has a missing/conflicting
license, or gnuplot has? With gnuplot that should be
OK as there are built-in modes for God knows what
technologies that most likely have licenses all over
the place, including none at all.

-- 
underground experts united

Re: gnuplot

[Javier Fernandez]

Emanuel Berg <embe8573@student.uu.se> wrote:
> Do you mean gnuplot-mode has a missing/conflicting
> license, or gnuplot has? With gnuplot that should be
> OK as there are built-in modes for God knows what
> technologies that most likely have licenses all over
> the place, including none at all.

gnuplot despite its name is not 'free' software in the sense of
freedom.  You can distribute gnuplot's source, but you cannot
distribute modified versions of the source.

http://gnuplot.cvs.sourceforge.net/gnuplot/gnuplot/Copyright?view=markup

The license I think is pretty similar to the old Pine licence, and
that led to problems between the FSF and the Univeristy of Washington
for distributing modified versions of Pine.  Now a 'libre' clone
exists: Alpine (Alternatively Licensed Program for Internet News and
Email)

http://en.wikipedia.org/wiki/Pine_(email_client)
http://en.wikipedia.org/wiki/Alpine_(email_client)

The moral of the story is that software that is not 'libre' has
converted to 'libre' or died.  In the case of scientific plotting
pgplot, a once popular library for scientifc plotting in compiled
programs (Fortran, C, C++) died (last release 15 years ago) since
Caltech didn't allow distributing modified sources.

http://en.wikipedia.org/wiki/PGPLOT

gnuplot seems to be an exception to the rule and is pretty healthy
(5.0 release was 1 month ago).  I know nothing about the authors, but
they do a great work and have to be working very hard to keep gnuplot
active.  I think at some point they will release it under a more
liberal license.  Otherwise gnuplot will stagnate.  There are GPL
programs with similar functionality

http://en.wikipedia.org/wiki/Pyxplot

Re: gnuplot

[Eli Zaretskii]

> From: Javier Fernandez <jafernan@faeroes.freeshell.org>
> Date: Wed, 11 Feb 2015 16:11:10 +0000 (UTC)
> 
> gnuplot despite its name is not 'free' software in the sense of
> freedom.  You can distribute gnuplot's source, but you cannot
> distribute modified versions of the source.
> 
> http://gnuplot.cvs.sourceforge.net/gnuplot/gnuplot/Copyright?view=markup

But you _can_ distribute the source modifications as patches to
released sources, so what's the difference?

Re: gnuplot

[Javier Fernandez]

Eli Zaretskii <eliz@gnu.org> wrote:
> But you _can_ distribute the source modifications as patches to
> released sources, so what's the difference?

The problems are explained in this ubuntu bug mailing list post

https://bugs.launchpad.net/ubuntu/+source/gnuplot/+bug/195111/comments/9

Henrik Nilsen Omma (henrik) wrote on 2008-11-17: 

> You could fork it in theory but it would have to be forever
> distributed as a combination of the original source at the
> version you forked and an ever-growing set of patches.

> It does allow you to make changes and distribute them but in an
> awkward form.

> it IS GPL-incompatible (e.g. it may not be distributed in binary
> form with the GNU readline library linked to it), because of that
> additional restriction.

> That said, debian-legal do consider this to be DFSG-free but
> consider it to be a compromise. From the linked Debian bug:

Re: gnuplot

[Javier Fernandez]

Perhaps I was a bit extreme in my examples before.  Pine license was a bit 
more restrictive since it didn't allow for distributing binary modified 
works and pine was in debian/non-free.

My meaning was just that gnuplot license poses problems for forking.
Even if you can fork you need to distribute it in an annoying way.

Re: gnuplot (was: Re: zsh outside of Emacs)

[Robert Thorpe]

Emanuel Berg <embe8573@student.uu.se> writes:
...
>
> Do you mean gnuplot-mode has a missing/conflicting
> license, or gnuplot has? With gnuplot that should be
> OK as there are built-in modes for God knows what
> technologies that most likely have licenses all over
> the place, including none at all.

I don't know why it's not included.

Possible reasons:
* The Emacs developers don't think it's important enough.
* It has a non GPL license.
* The Emacs developers can't get copyright assignment for it.

The FSF generally require that Emacs coders assign copyright to them.
Unless it's a small change it's not considered good enough to just use
the GPLv3 license.  As far as I understand it, there are two reasons for
that.  Firstly, it enables them to easily re-license the code.  That
could happen if a legal flaw is found in a license, for example GPLv3
was created to rectify problems with GPLv2.  The other advantage is that
it prevents programmers from being able to relicense their contributions
and request that they're removed.

What sometimes happens is that an elisp package is initially unpopular.
But, the author gathers contributions from many people.  Then the
package becomes popular and people want it in the core or ELPA.  But by
that time the author can't find all the people to contact to ask for
copyright paperwork, or there are so many of them that it isn't
practical.

You could try asking for it to be included in the core it may be
possible.

BR,
Robert Thorpe

Re: gnuplot

[Emanuel Berg]

Javier Fernandez <jafernan@faeroes.freeshell.org>
writes:

> gnuplot despite its name is not 'free' software in
> the sense of freedom. You can distribute gnuplot's
> source, but you cannot distribute modified versions
> of the source.

Again, I'm not saying *gnuplot* should be included in
Emacs, how now that would work :) I'm talking about
the gnuplot-mode. In gnuplot-mode.el I read

    ;; Copyright (C) 2010-2013 Mike McCourt
    ;;
    ;; Authors: Mike McCourt <...>
    ;; URL: https://github.com/mkmcc/gnuplot
    ;; Version: 20131203.2159
    ;; X-Original-Version: 1.2.0
    ;; Keywords: gnuplot, plotting

    ;; This file is not part of GNU Emacs.

It doesn't say anything about what (if any) licence
is at work, but I would conclude from all this
Mr McCourt have his reasons which we may or may not
agree with, but again it doesn't matter because MELPA
is here and gnuplot-mode is in MELPA and he who finds
both Emacs and gnuplot will find MELPA, I'm confident.

By the way, I CC this to the author: let's see what
happens.

> gnuplot seems to be an exception to the rule and is
> pretty healthy (5.0 release was 1 month ago). I know
> nothing about the authors, but they do a great work
> and have to be working very hard to keep gnuplot
> active. I think at some point they will release it
> under a more liberal license. Otherwise gnuplot will
> stagnate.

Of course, I hope they continue work on gnuplot but I
kind of disagree in the sense that I think gnuplot is
just like a programming language: imagine from now on,
no one ever did anything to modernize C - it could
(and would) still be used to do a thousand super-cool
present and future projects. I don't think the world
of math, stats, and visualization will *ever* change
that much as to make gnuplot with its current
functionality level obsolete.

> There are GPL programs with similar functionality

And those are...?

-- 
underground experts united

Re: gnuplot

[Stefan Monnier]

> Again, I'm not saying *gnuplot* should be included in
> Emacs, how now that would work :) I'm talking about
> the gnuplot-mode. In gnuplot-mode.el I read

FWIW, I'd be happy to see gnuplot-mode in GNU ELPA.


        Stefan

Re: gnuplot

[Marcin Borkowski]

On 2015-02-11, at 17:11, Javier Fernandez <jafernan@faeroes.freeshell.org> wrote:

> [...]  I think at some point they will release it under a more
> liberal license.  Otherwise gnuplot will stagnate.  There are GPL

Just my 2 cents: GPL is not necessarily what should be called
a "liberal" license -- personally I find it rather restrictive, in
a sense.

Regards,

-- 
Marcin Borkowski
http://octd.wmi.amu.edu.pl/en/Marcin_Borkowski
Faculty of Mathematics and Computer Science
Adam Mickiewicz University

Re: gnuplot

[Stefan Monnier]

>> [...]  I think at some point they will release it under a more
>> liberal license.  Otherwise gnuplot will stagnate.  There are GPL
> Just my 2 cents: GPL is not necessarily what should be called
> a "liberal" license -- personally I find it rather restrictive, in
> a sense.

He didn't say the GPL is liberal.  Just that it's *more* liberal than
gnuplot's current license.


        Stefan

Re: gnuplot

[Marcin Borkowski]

On 2015-02-12, at 15:14, Marcin Borkowski <mbork@wmi.amu.edu.pl> wrote:

> On 2015-02-11, at 17:11, Javier Fernandez <jafernan@faeroes.freeshell.org> wrote:
>
>> [...]  I think at some point they will release it under a more
>> liberal license.  Otherwise gnuplot will stagnate.  There are GPL
>
> Just my 2 cents: GPL is not necessarily what should be called
> a "liberal" license -- personally I find it rather restrictive, in
> a sense.

BTW: I do not like to sound trollish, but there was recently a heated
discussion with exceptionally rude posts on Org-mode mailing list about
so-called non-"free" (free-as-in-Free-Software-Foundation, to be
precise) software.

It was stated there that GNU mailing lists do not allow promotion of
non-"free" software.  So - if gnuplot is apparently not kosher - what is
the official stance on this very thread?  I'm asking because it was
stated in that discussion that

> [Redacted] is non-free software, and people should not install it, or
> suggest installing it, or even tell people it exists.

and in another post

> Please stop using the GNU mailing lists to promote proprietary
> software.

(To be fair, not all participants seemed to agree with the above -
rather extreme - quotes, and there was also a polite explanation of why
informing about [redacted] was not necessarily a good idea on Org-mode
mailing list.)

I'd really like to understand what is good and what is evil according to
the FSF, and - also from sheer curiosity - why is the current thread
even allowed here (not to mention that there's even a discussion about
a *possibility* of including gnuplot-mode in Emacs itself, which -
according to what was said in the mentioned thread - sounds
sacrilegious).

This thread is another hint for me that there's something fishy going on
with all this "free-software" talk.  I'm afraid that - as is often the
case - when some organization (three-letter or not) says that it aims to
"promote [somebody's] freedom and to defend the rights of all
[somebodies]", it's really some politics and not anybody's freedom it's
all about.  (If there is another explanation for this seemingly unfair
treatment of various software projects, please do enlighten me!)  Even
though I do not personally agree with RMS or FSF, I have to say that
*if* this is indeed the case, I would be rather saddened and seriously
disappointed - until recently, I really believed in good intentions
(however mistaken their morality) behind the FSF.

Regards,

-- 
Marcin Borkowski
http://octd.wmi.amu.edu.pl/en/Marcin_Borkowski
Faculty of Mathematics and Computer Science
Adam Mickiewicz University

Re: gnuplot

[Marcin Borkowski]

On 2015-02-12, at 15:54, Marcin Borkowski <mbork@wmi.amu.edu.pl> wrote:

> [a few rather harsh thoughts of a very disappointed man]

After sending my previous email (which I've been composing for more than
30 minutes, mostly cutting out a few things I decided would be better
withdrawn - maybe still not enough...), I thought that I didn't make one
important thing clear.

I *do not* want to be harsh against Emacs developers.  I consider their
work to be great, and IMHO Emacs is one of the two best applications
written *ever* (even though it has its quirks).  I also *do not* want to
be harsh to the Emacs community, which I consider a very nice and
welcoming one.  I really, really regret that I am not able to give back
the huge debt I incurred by using Emacs in my personal life, my studies
and my job, by contributing to Emacs itself.  I do try to repay this
debt by contributing in other ways (mailing list, blog, to name a few).
What angers me is the injustice I wrote about (and the spreading of
false moral views by the FSF).  What angers me is people being rude
toward some people and *at the same time* accepting very similar things
from other people.

I guess this would probably be EOT for me, though I really looking
forward to reasonable explanations of the situation I wrote about.

Best,

-- 
Marcin Borkowski
http://octd.wmi.amu.edu.pl/en/Marcin_Borkowski
Faculty of Mathematics and Computer Science
Adam Mickiewicz University

Re: gnuplot

[Stefan Monnier]

>> Please stop using the GNU mailing lists to promote proprietary
>> software.

This is indeed the accepted rule.  The FSF was founded to fight against
proprietary software, so it seems only normal that it wouldn't want
people to use its mailing lists to promote it (GNU mailing lists use
the resources of the FSF).

> I'd really like to understand what is good and what is evil according to
> the FSF,

http://www.fsf.org/about/ is a good start.

> and - also from sheer curiosity - why is the current thread
> even allowed here (not to mention that there's even a discussion about
> a *possibility* of including gnuplot-mode in Emacs itself, which -
> according to what was said in the mentioned thread - sounds
> sacrilegious).

AFAIK gnuplot is not proprietary software.  Its license is a bit
constraining but I think it is acceptable as Free Software (and that's
also the interpretation of Debian).  I see it's not listed in the Free
Software Directory, so maybe the FSF treats it as just on the other side
of the fence, but I wouldn't take this to mean that the FSF considers
gnuplot as proprietary.

> I'm afraid that - as is often the case - when some organization
> (three-letter or not) says that it aims to "promote [somebody's]
> freedom and to defend the rights of all [somebodies]", it's really
> some politics and not anybody's freedom it's all about.

Of course it's political by nature.  And given how "freedom" works, you
can't provide freedom somewhere without restricting it elsewhere.

But note that "don't promote <foo> on this list" does not prevent anyone
from using <foo>.

> (If there is another explanation for this seemingly unfair treatment
> of various software projects, please do enlighten me!)

Not sure which software projects you think were treated unfairly (nor
when/where), so I can't say anything useful here.


        Stefan

Re: gnuplot

[Emanuel Berg]

Marcin Borkowski <mbork@wmi.amu.edu.pl> writes:

> It was stated there that GNU mailing lists do not
> allow promotion of non-"free" software. So - if
> gnuplot is apparently not kosher - what is the
> official stance on this very thread? I'm asking
> because it was stated in that discussion that
>
>> [Redacted] is non-free software, and people should
>> not install it, or suggest installing it, or even
>> tell people it exists.
>
> and in another post
>
>> Please stop using the GNU mailing lists to promote
>> proprietary software.

"Promote" in the sense hitting the big drum (almost as
in advertisement if a commercial product), that would
be a bit strange, and definitely unwelcome if done in
isolation, but none of that was the case here.

OT: What I really wanted to tell you: when you quote,
don't use the '>' (or nested '>>' etc.) unless it has
appeared previously in the very thread. Take a look at
this screenshot [1] to understand why.

If you want to make it look nice, do it like this:

    [Redacted] is non-free software, and people should
    not install it, or suggest installing it, or even
    tell people it exists.

That's [C-u C-x TAB] (and C-q, if needed).

[1] http://user.it.uu.se/~embe8573/quote.png

-- 
underground experts united

Re: gnuplot

[Emanuel Berg]

Marcin Borkowski <mbork@wmi.amu.edu.pl> writes:

> After sending my previous email (which I've been
> composing for more than 30 minutes, mostly cutting
> out a few things I decided would be better withdrawn
> - maybe still not enough...), I thought that I
> didn't make one important thing clear.

C'mon, relax.

> I *do not* want to be harsh against Emacs
> developers. I consider their work to be great, and
> IMHO Emacs is one of the two best applications
> written *ever* (even though it has its quirks). I
> also *do not* want to be harsh to the Emacs
> community, which I consider a very nice and
> welcoming one. I really, really regret that I am not
> able to give back the huge debt I incurred by using
> Emacs in my personal life, my studies and my job, by
> contributing to Emacs itself. I do try to repay this
> debt by contributing in other ways (mailing list,
> blog, to name a few).

Don't worry about it! You don't owe Emacs anything:
"repay" by being productive with it in whatever field
or business or activity you are in. Do you think a
bike mechanic worries how to repay whoever came up
with the Torx 25 wrench? Oh no. Just keep it up.

-- 
underground experts united

Re: gnuplot

[Emanuel Berg]

Stefan Monnier <monnier@iro.umontreal.ca> writes:

> AFAIK gnuplot is not proprietary software. Its
> license is a bit constraining but I think it is
> acceptable as Free Software (and that's also the
> interpretation of Debian).

Indeed, it is in the (free) math section.

Try 'aptitude show gnuplot'.

-- 
underground experts united

Re: gnuplot

[Emanuel Berg]

Stefan Monnier <monnier@iro.umontreal.ca> writes:

>> Again, I'm not saying *gnuplot* should be included
>> in Emacs, how now that would work :) I'm talking
>> about the gnuplot-mode.
>
> FWIW, I'd be happy to see gnuplot-mode in GNU ELPA.

I received a mail from the author. He said "Hi all,"
so I suspect it was intended for the newsgroup (I CC
this post to him as well). Here is what he wrote:

    this e-mail  arrived without much context,  so I'm
    kind  of  guessing at  its  meaning.  but, I'd  be
    thrilled to see  gnuplot-mode included with emacs!
    I have no clue how that process works, however.

    I didn't give any thought to the license at all...
    I just wrote gnuplot-mode because I needed it. And
    I intended for  anyone to use it as he  or she saw
    fit. I'd  be happy to  add a license to  the file,
    though.

-- 
underground experts united

Re: gnuplot

[Thien-Thi Nguyen]

[-- Attachment #1: Type: text/plain, Size: 598 bytes --]

() Emanuel Berg <embe8573@student.uu.se>
() Fri, 13 Feb 2015 00:22:24 +0100

   Don't worry about it! You don't owe Emacs anything:

Emacs is but open-paren, aesthetics call for close,
or functions neat, but incomplete, drive one to dream/compose.
 but no one is metering
 the rate of the petering:
only machines really "care" (if that) who does and who owes.

-- 
Thien-Thi Nguyen
   GPG key: 4C807502
   (if you're human and you know it)
      read my lisp: (responsep (questions 'technical)
                               (not (via 'mailing-list)))
                     => nil

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 197 bytes --]