[23.0.90] No sources for lisp/international/uni-*.el?

[23.0.90] No sources for lisp/international/uni-*.el?

[Ulrich Mueller]

The files in lisp/international/uni-*.el are object files, generated
from UnicodeData.txt.

Is it intentional that the corresponding source files (which are in
admin/unidata/ in CVS) are not included in the tarball?

ulrich

Re: [23.0.90] No sources for lisp/international/uni-*.el?

[Glenn Morris]

Ulrich Mueller wrote:

> The files in lisp/international/uni-*.el are object files, generated
> from UnicodeData.txt.
>
> Is it intentional that the corresponding source files (which are in
> admin/unidata/ in CVS) are not included in the tarball?

I guess so, otherwise they should not be in the admin/ directory.

Re: [23.0.90] No sources for lisp/international/uni-*.el?

[Ulrich Mueller]

>>>>> On Mon, 02 Feb 2009, Glenn Morris wrote:

> Ulrich Mueller wrote:
>> The files in lisp/international/uni-*.el are object files,
>> generated from UnicodeData.txt.
>> 
>> Is it intentional that the corresponding source files (which are
>> in admin/unidata/ in CVS) are not included in the tarball?

> I guess so, otherwise they should not be in the admin/ directory.

But this will make it impossible to rebuild these files from the
tarball (e.g., for updating to a newer Unicode version).

Ulrich

No sources for lisp/international/uni-*.el?

[Ulrich Mueller]

I had already asked the following two years ago (see
<http://lists.gnu.org/archive/html/emacs-devel/2009-02/msg00083.html>),
and didn't get an answer then:

>>> The files in lisp/international/uni-*.el are object files,
>>> generated from UnicodeData.txt.
>>> 
>>> Is it intentional that the corresponding source files (which are
>>> in admin/unidata/ in CVS) are not included in the tarball?

>> I guess so, otherwise they should not be in the admin/ directory.

> But this will make it impossible to rebuild these files from the
> tarball (e.g., for updating to a newer Unicode version).

Could somebody please explain how this issue is different from the
recent one with the CEDET grammar files?

Ulrich

Re: No sources for lisp/international/uni-*.el?

[Richard Stallman]

    Could somebody please explain how this issue is different from the
    recent one with the CEDET grammar files?

According to my not totally reliable memory, we concluded that both
forms of these files were equally valid as the "source code", because
they were equally handy for making any changes.

Can someone else confirm this?

-- 
Dr Richard Stallman
President, Free Software Foundation
51 Franklin St
Boston MA 02110
USA
www.fsf.org  www.gnu.org
Skype: No way! That's nonfree (freedom-denying) software.
  Use free telephony http://directory.fsf.org/category/tel/

No sources for lisp/international/uni-*.el?

[Stephen J. Turnbull]

Ulrich Mueller writes:
 > I had already asked the following two years ago (see
 > <http://lists.gnu.org/archive/html/emacs-devel/2009-02/msg00083.html>),
 > and didn't get an answer then:
 > 
 > >>> The files in lisp/international/uni-*.el are object files,
 > >>> generated from UnicodeData.txt.
 > >>> 
 > >>> Is it intentional that the corresponding source files (which are
 > >>> in admin/unidata/ in CVS) are not included in the tarball?
 > 
 > >> I guess so, otherwise they should not be in the admin/ directory.
 > 
 > > But this will make it impossible to rebuild these files from the
 > > tarball (e.g., for updating to a newer Unicode version).
 > 
 > Could somebody please explain how this issue is different from the
 > recent one with the CEDET grammar files?

I don't know if it is still true, but I seem to recall that although
you are allowed to use Unidata freely-as-in-free-software, the Unicode
Consortium wants you (with some degree of legal coercion) to refer
downstream users to their canonical originals, and not distribute them
yourself.

Re: No sources for lisp/international/uni-*.el?

[Eli Zaretskii]

> From: "Stephen J. Turnbull" <stephen@xemacs.org>
> Date: Mon, 08 Aug 2011 11:04:49 +0900
> Cc: emacs-devel@gnu.org
> 
> I don't know if it is still true, but I seem to recall that although
> you are allowed to use Unidata freely-as-in-free-software, the Unicode
> Consortium wants you (with some degree of legal coercion) to refer
> downstream users to their canonical originals, and not distribute them
> yourself.

IANAL, but it sounds like this have changed, see

  http://www.unicode.org/copyright.html#Exhibit1

It says, inter alia:

 Permission is hereby granted, free of charge, to any person obtaining
 a copy of the Unicode data files and any associated documentation
 (the "Data Files") or Unicode software and any associated
 documentation (the "Software") to deal in the Data Files or Software
 without restriction, including without limitation the rights to use,
 copy, modify, merge, publish, distribute, and/or sell copies of the
 Data Files or Software, and to permit persons to whom the Data Files
 or Software are furnished to do so, provided that (a) the above
 copyright notice(s) and this permission notice appear with all copies
 of the Data Files or Software, (b) both the above copyright notice(s)
 and this permission notice appear in associated documentation, and
 (c) there is clear notice in each modified Data File or in the
 Software as well as in the documentation associated with the Data
 File(s) or Software that the data or software has been modified.

Re: No sources for lisp/international/uni-*.el?

[Ulrich Mueller]

>>>>> On Sun, 07 Aug 2011, Richard Stallman wrote:

>     Could somebody please explain how this issue is different from
>     the recent one with the CEDET grammar files?

> According to my not totally reliable memory, we concluded that both
> forms of these files were equally valid as the "source code",
> because they were equally handy for making any changes.

Hm, for example lisp/international/uni-name.el doesn't look like "the
preferred form of the work for making modifications to it". The file
rather has the appearence of line noise. ;-)

Re: No sources for lisp/international/uni-*.el?

[David Kastrup]

Ulrich Mueller <ulm@gentoo.org> writes:

>>>>>> On Sun, 07 Aug 2011, Richard Stallman wrote:
>
>>     Could somebody please explain how this issue is different from
>>     the recent one with the CEDET grammar files?
>
>> According to my not totally reliable memory, we concluded that both
>> forms of these files were equally valid as the "source code",
>> because they were equally handy for making any changes.
>
> Hm, for example lisp/international/uni-name.el doesn't look like "the
> preferred form of the work for making modifications to it". The file
> rather has the appearence of line noise. ;-)

Pretty much so.  Starting comment:

;; Copyright (C) 1991-2009 Unicode, Inc.
;; This file was generated from the Unicode data files at
;; http://www.unicode.org/Public/UNIDATA/.
;; See lisp/international/README for the copyright and permission notice.

Followed by something looking like line noise indeed.

The respective README says

The following files in this directory are derived from the Unicode
Data File at http://www.unicode.org/Public/UNIDATA/UnicodeData.txt,
downloaded on 2006-05-23:

  charprop.el uni-bidi.el uni-category.el uni-combining.el
  uni-comment.el uni-decimal.el uni-decomposition.el uni-digit.el
  uni-lowercase.el uni-mirrored.el uni-name.el uni-numeric.el
  uni-old-name.el uni-titlecase.el uni-uppercase.el

These files were generated from UnicodeData.txt using unidata-gen.el,
which is not included in the Emacs distribution; it can be found in
the admin/unidata directory of the Emacs source repository at
https://savannah.gnu.org/projects/emacs/

The file UnicodeData.txt is used under the terms of the following
Copyright and Permission Notice:


Copyright (C) 1991-2009 Unicode, Inc. All rights reserved. Distributed
under the Terms of Use in http://www.unicode.org/copyright.html.

Permission is hereby granted, free of charge, to any person obtaining
a copy of the Unicode data files and any associated documentation (the
"Data Files") or Unicode software and any associated documentation
(the "Software") to deal in the Data Files or Software without
restriction, including without limitation the rights to use, copy,
modify, merge, publish, distribute, and/or sell copies of the Data
Files or Software, and to permit persons to whom the Data Files or
Software are furnished to do so, provided that (a) the above copyright
notice(s) and this permission notice appear with all copies of the
Data Files or Software, (b) both the above copyright notice(s) and
this permission notice appear in associated documentation, and (c)
there is clear notice in each modified Data File or in the Software as
well as in the documentation associated with the Data File(s) or
Software that the data or software has been modified.

THE DATA FILES AND SOFTWARE ARE PROVIDED "AS IS", WITHOUT WARRANTY OF
ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE
WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
NONINFRINGEMENT OF THIRD PARTY RIGHTS. IN NO EVENT SHALL THE COPYRIGHT
HOLDER OR HOLDERS INCLUDED IN THIS NOTICE BE LIABLE FOR ANY CLAIM, OR
ANY SPECIAL INDIRECT OR CONSEQUENTIAL DAMAGES, OR ANY DAMAGES
WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT
OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THE DATA FILES OR
SOFTWARE.

Except as contained in this notice, the name of a copyright holder
shall not be used in advertising or otherwise to promote the sale, use
or other dealings in these Data Files or Software without prior
written authorization of the copyright holder.


-- 
David Kastrup

Re: No sources for lisp/international/uni-*.el?

[Richard Stallman]

    > Hm, for example lisp/international/uni-name.el doesn't look like "the
    > preferred form of the work for making modifications to it". The file
    > rather has the appearence of line noise. ;-)

Maybe that memory concerned some other files.  I said I might be
wrong.

Since these files are not under the GPL, we are not leading anyone to
violate the GPL.  So there is no legal issue, but there remains an
ethical issue: is it wrong for a release to contain generated files
made from sources that are distributed separately?

At the very least, it is good to avoid this.  So why not add these
sources to the distrbution?

-- 
Dr Richard Stallman
President, Free Software Foundation
51 Franklin St
Boston MA 02110
USA
www.fsf.org  www.gnu.org
Skype: No way! That's nonfree (freedom-denying) software.
  Use free telephony http://directory.fsf.org/category/tel/

Re: No sources for lisp/international/uni-*.el?

[David Kastrup]

Richard Stallman <rms@gnu.org> writes:

>     > Hm, for example lisp/international/uni-name.el doesn't look like "the
>     > preferred form of the work for making modifications to it". The file
>     > rather has the appearence of line noise. ;-)
>
> Maybe that memory concerned some other files.  I said I might be
> wrong.
>
> Since these files are not under the GPL, we are not leading anyone to
> violate the GPL.  So there is no legal issue,

   5.c) You must license the entire work, as a whole, under this
    License to anyone who comes into possession of a copy.  This
    License will therefore apply, along with any applicable section 7
    additional terms, to the whole of the work, and all its parts,
    regardless of how they are packaged.  This License gives no
    permission to license the work in any other way, but it does not
    invalidate such permission if you have separately received it.

  12. No Surrender of Others' Freedom.

  If conditions are imposed on you (whether by court order, agreement or
otherwise) that contradict the conditions of this License, they do not
excuse you from the conditions of this License.  If you cannot convey a
covered work so as to satisfy simultaneously your obligations under this
License and any other pertinent obligations, then as a consequence you may
not convey it at all.  For example, if you agree to terms that obligate you
to collect a royalty for further conveying from those to whom you convey
the Program, the only way you could satisfy both those terms and this
License would be to refrain entirely from conveying the Program.


-- 
David Kastrup

Re: No sources for lisp/international/uni-*.el?

[Andreas Röhler]

Am 09.08.2011 03:44, schrieb Richard Stallman:
>      >  Hm, for example lisp/international/uni-name.el doesn't look like "the
>      >  preferred form of the work for making modifications to it". The file
>      >  rather has the appearence of line noise. ;-)
>
> Maybe that memory concerned some other files.  I said I might be
> wrong.
>
> Since these files are not under the GPL, we are not leading anyone to
> violate the GPL.  So there is no legal issue, but there remains an
> ethical issue: is it wrong for a release to contain generated files
> made from sources that are distributed separately?

In case you answer "yes", you must include the gcc-sources as well.

Also the sources, the gcc is build from...

IMHO in short: copyleft is saying: you must include the universe.

Where again the wrong with copyleft is it's requiring, it's menace with 
the prosecutor and the courts.

Would the GPL say politely: please do, it's recommended, than it's fine 
for sure.

Real existing powers with it's police-systems, services and courts are 
not ethical because of the way they are build up nowadays.

Maybe once in a better world constructed from Emacs-Sources... :)

Andreas

>
> At the very least, it is good to avoid this.  So why not add these
> sources to the distrbution?
>

Re: No sources for lisp/international/uni-*.el?

[David Kastrup]

Andreas Röhler <andreas.roehler@online.de> writes:

> Am 09.08.2011 03:44, schrieb Richard Stallman:
>>      >  Hm, for example lisp/international/uni-name.el doesn't look like "the
>>      >  preferred form of the work for making modifications to it". The file
>>      >  rather has the appearence of line noise. ;-)
>>
>> Maybe that memory concerned some other files.  I said I might be
>> wrong.
>>
>> Since these files are not under the GPL, we are not leading anyone to
>> violate the GPL.  So there is no legal issue, but there remains an
>> ethical issue: is it wrong for a release to contain generated files
>> made from sources that are distributed separately?
>
> In case you answer "yes", you must include the gcc-sources as well.

[And it gets more absurd with every following sentence]

Could you take your confused nonsense elsewhere?  As usual, you have no
idea what you are actually talking about.  To quote from the GPL:

      The "Corresponding Source" for a work in object code form means all
    the source code needed to generate, install, and (for an executable
    work) run the object code and to modify the work, including scripts to
    control those activities.  However, it does not include the work's
    System Libraries, or general-purpose tools or generally available free
    programs which are used unmodified in performing those activities but
    which are not part of the work.

You have not read and understood the GPL, _and_ you have not read and
understood relevant commentary about it.  Yet you proceed not just
lecturing the Emacs developer list about it, but feel self-important
enough to include Richard Stallman in an additional copy of your
nonsensical blatherings.

Do you really think you can find anybody on this list who is even more
clueless than you are about what is written in the GPL?  So where is the
point in your tirades?  Not even on the XEmacs lists you pestered for
while, where you should be preaching to the choir, did anybody join in
your incoherent licensing rants.

So could you please stay on the sidelines when you have nothing useful
to contribute?  Thanks.

If you have problems with the English language, there are German
translations of the GPL to be found.  Try reading the one at
<URL:http://www.gnu.de/documents/gpl-3.0.de.html>.  Whether you prefer
reading the original or a translation, you really should try to
understand the main point of each clause before making statements about
the GPL, whether on this list or anywhere else.  There is also the FAQ
about the GPL at <URL:http://www.gnu.org/licenses/gpl-faq.html> which
might clarify a few things.

I apologize to the other list members for the tone in this mail.

-- 
David Kastrup

Re: No sources for lisp/international/uni-*.el?

[Andreas Röhler]

> I apologize to the other list members for the tone in this mail.
>

which pretty good illustrates my concern

Re: No sources for lisp/international/uni-*.el?

[Eli Zaretskii]

> Date: Tue, 09 Aug 2011 10:21:09 +0200
> From: Andreas Röhler <andreas.roehler@online.de>
> 
> 
> > I apologize to the other list members for the tone in this mail.
> >
> 
> which pretty good illustrates my concern

Tone aside, your repeated attacks on the GPL are really off-topic
here.  There are specialized forums for that kind of discussions.

Re: No sources for lisp/international/uni-*.el?

[Andreas Röhler]

Am 09.08.2011 10:45, schrieb Eli Zaretskii:
>> Date: Tue, 09 Aug 2011 10:21:09 +0200
>> From: Andreas Röhler<andreas.roehler@online.de>
>>
>>
>>> I apologize to the other list members for the tone in this mail.
>>>
>>
>> which pretty good illustrates my concern
>
> Tone aside, your repeated attacks on the GPL

Hi Eli,

even if my contributions are modest, it should be clear I'm a supporter 
of free software. BTW published several stuff under GPL. I'm far from 
attacking.

OTOH should I hide concern if envisaging danger, which might destroy the 
whole project still, if not faced?

The point is that some requirings of the GPL are de facto impossible to 
fulfil. Well, that impossibility only exists to a certain extend, but it 
grows with the project.

 From there seeing larger projects in constant danger being sued.

That's well known for long for people who cared for. Just we reached the 
point it might be brought to consideration to people pretty close to the 
GPL-authors. Sometimes we need real examples to understand...

Anyway, as said earlier, for me an accessible and working Emacs is the 
final proof of concept.

Thanks so far

Andreas



  are really off-topic
> here.  There are specialized forums for that kind of discussions.
>

Re: No sources for lisp/international/uni-*.el?

[Ulrich Mueller]

>>>>> On Tue, 09 Aug 2011, Andreas Röhler wrote:

> Am 09.08.2011 03:44, schrieb Richard Stallman:
>> >  Hm, for example lisp/international/uni-name.el doesn't look like
>> >  "the preferred form of the work for making modifications to it".
>> >  The file rather has the appearence of line noise. ;-)
>> 
>> Maybe that memory concerned some other files.  I said I might be
>> wrong.
>> 
>> Since these files are not under the GPL, we are not leading anyone
>> to violate the GPL. So there is no legal issue, but there remains
>> an ethical issue: is it wrong for a release to contain generated
>> files made from sources that are distributed separately?

> In case you answer "yes", you must include the gcc-sources as well.

The difference between gcc and unidata-gen.el is that the former is a
general-purpose tool that normally is installed on your system, while
the latter is not.

And (what a surprise ;-) the GPL has a clause that covers that:

# The "Corresponding Source" [...] does not include the work's System
# Libraries, or general-purpose tools or generally available free
# programs which are used unmodified in performing those activities
# but which are not part of the work.

For UnicodeData.txt one could even argue that it falls into this
category because it's freely available and used unmodified (it's also
part of the "miscfiles" package [1] that I believe is installed on
most GNU/Linux systems).

However, the same is not true for the corresponding build tools (like
unidata-gen.el or biditype.awk) which are specific to Emacs.

Ulrich

[1] <http://www.gnu.org/software/miscfiles/>

Re: No sources for lisp/international/uni-*.el?

[Stephen J. Turnbull]

Ulrich Mueller writes:

 > For UnicodeData.txt one could even argue that it falls into this
 > category because it's freely available and used unmodified (it's also
 > part of the "miscfiles" package [1] that I believe is installed on
 > most GNU/Linux systems).

Unfortunately, no, that doesn't work without specific guidance from
the relevant copyright owner(s).  Specifically, it *is* part of the
work.

Absent a clear declaration to the contrary from the FSF's lawyer, I
would say David's is the conservative position, and I'm surprised that
Richard doesn't see it that way.  I can't reverse engineer his
statement to get the logic that led to it, but I might be missing
something.

Re: No sources for lisp/international/uni-*.el?

[Eli Zaretskii]

> Date: Tue, 9 Aug 2011 11:59:50 +0200
> From: Ulrich Mueller <ulm@gentoo.org>
> Cc: rms@gnu.org, Emacs developers <emacs-devel@gnu.org>
> 
> However, the same is not true for the corresponding build tools (like
> unidata-gen.el or biditype.awk) which are specific to Emacs.

Don't worry about biditype.awk, it will be deleted soon, certainly
prior to release of Emacs 24.1.

Re: No sources for lisp/international/uni-*.el?

[Thien-Thi Nguyen]

() Andreas Röhler <andreas.roehler@online.de>
() Tue, 09 Aug 2011 11:41:56 +0200

   OTOH should I hide concern if envisaging danger,
   which might destroy the whole project still, if not faced?

The first danger is not understanding how you (mis)understand danger.
I think everyone is trying to help you overcome that danger (which,
due to its nature, you might not even recognize).

Re: No sources for lisp/international/uni-*.el?

[Eli Zaretskii]

> Date: Tue, 09 Aug 2011 11:41:56 +0200
> From: Andreas Röhler <andreas.roehler@online.de>
> CC: emacs-devel@gnu.org
> 
> Am 09.08.2011 10:45, schrieb Eli Zaretskii:
> >> Date: Tue, 09 Aug 2011 10:21:09 +0200
> >> From: Andreas Röhler<andreas.roehler@online.de>
> >>
> >>
> >>> I apologize to the other list members for the tone in this mail.
> >>>
> >>
> >> which pretty good illustrates my concern
> >
> > Tone aside, your repeated attacks on the GPL
> 
> Hi Eli,
> 
> even if my contributions are modest, it should be clear I'm a supporter 
> of free software. BTW published several stuff under GPL.

That's besides the point in this case.

> I'm far from attacking.

Then perhaps your wording needs to be revised, because any time
related issues come up, your messages are explicitly negative.

> OTOH should I hide concern if envisaging danger, which might destroy the 
> whole project still, if not faced?

This is not the place to express those concerns, that's all I'm
saying.  There are other forums for that.  Why you insist talking
about this here is beyond me.

Re: No sources for lisp/international/uni-*.el?

[Richard Stallman]

    > Since these files are not under the GPL, we are not leading anyone to
    > violate the GPL.  So there is no legal issue, but there remains an
    > ethical issue: is it wrong for a release to contain generated files
    > made from sources that are distributed separately?

    In case you answer "yes", you must include the gcc-sources as well.

That is mistaken -- nothing in Emacs is generated from source files in
GCC, so GCC is not part of the sorce code of Emacs.  Perhaps you have
overlooked the distinction between the source files and tools.

    Would the GPL say politely: please do, it's recommended, than it's fine 
    for sure.

If you want to criticize the GPL, please do it in
gnu-misc-discuss@gnu.org, not here.

-- 
Dr Richard Stallman
President, Free Software Foundation
51 Franklin St
Boston MA 02110
USA
www.fsf.org  www.gnu.org
Skype: No way! That's nonfree (freedom-denying) software.
  Use free telephony http://directory.fsf.org/category/tel/

Re: No sources for lisp/international/uni-*.el?

[Richard Stallman]

If we consider these files part of Emacs, then the GPL applies to
them.  But since they are made from a file with a lax license, they
themselves don't have to be under the GPL.  We can consider them
something distributed along with Emacs, in which case the GPL doesn't
have to apply to them.

That explains why I don't see a legal issue here, and why the users
are not violating the GPL.

Notwithstanding that, is there any reason not to add those source
files to Emacs?  Nobody responded to that.


-- 
Dr Richard Stallman
President, Free Software Foundation
51 Franklin St
Boston MA 02110
USA
www.fsf.org  www.gnu.org
Skype: No way! That's nonfree (freedom-denying) software.
  Use free telephony http://directory.fsf.org/category/tel/

Re: No sources for lisp/international/uni-*.el?

[Chong Yidong]

Richard Stallman <rms@gnu.org> writes:

> Notwithstanding that, is there any reason not to add those source
> files to Emacs?  Nobody responded to that.

I don't think there is a reason other than the fact that admin/ was the
best place to put UnicodeData.txt (which is a 1.4Mb file that is not
used directly by Emacs itself, and must not be changed apart from
updating to new Unicode releases, so it doesn't make sense to put in
etc/).

Maybe we should just add the admin/ directory to the source tarballs.
WDYT?

Re: No sources for lisp/international/uni-*.el?

[Eli Zaretskii]

> From: Chong Yidong <cyd@stupidchicken.com>
> Date: Wed, 10 Aug 2011 12:01:35 -0400
> Cc: David Kastrup <dak@gnu.org>, emacs-devel@gnu.org
> 
> Richard Stallman <rms@gnu.org> writes:
> 
> > Notwithstanding that, is there any reason not to add those source
> > files to Emacs?  Nobody responded to that.
> 
> I don't think there is a reason other than the fact that admin/ was the
> best place to put UnicodeData.txt (which is a 1.4Mb file that is not
> used directly by Emacs itself, and must not be changed apart from
> updating to new Unicode releases, so it doesn't make sense to put in
> etc/).
> 
> Maybe we should just add the admin/ directory to the source tarballs.
> WDYT?

Why wouldn't etc/unidata-gen.el be enough?

Re: No sources for lisp/international/uni-*.el?

[Chong Yidong]

Eli Zaretskii <eliz@gnu.org> writes:

>> I don't think there is a reason other than the fact that admin/ was the
>> best place to put UnicodeData.txt (which is a 1.4Mb file that is not
>> used directly by Emacs itself, and must not be changed apart from
>> updating to new Unicode releases, so it doesn't make sense to put in
>> etc/).
>>
>> Maybe we should just add the admin/ directory to the source tarballs.
>> WDYT?
>
> Why wouldn't etc/unidata-gen.el be enough?

unidata-gen.el is not useful to normal users.  It doesn't make sense to
add it to etc/ because that directory is installed by "make install".

Re: No sources for lisp/international/uni-*.el?

[Eli Zaretskii]

> From: Chong Yidong <cyd@stupidchicken.com>
> Cc: rms@gnu.org, dak@gnu.org, emacs-devel@gnu.org
> Date: Wed, 10 Aug 2011 12:46:19 -0400
> 
> > Why wouldn't etc/unidata-gen.el be enough?
> 
> unidata-gen.el is not useful to normal users.  It doesn't make sense to
> add it to etc/ because that directory is installed by "make install".

If that's the only problem, we could teach "make install" not to
install that file, or find some other place to hold it.  Adding all of
admin/ for the benefit of a single file sounds excessive to me.

Re: No sources for lisp/international/uni-*.el?

[Chong Yidong]

Eli Zaretskii <eliz@gnu.org> writes:

> If that's the only problem, we could teach "make install" not to
> install that file, or find some other place to hold it.  Adding all of
> admin/ for the benefit of a single file sounds excessive to me.

Then someone is going to ask about whether something else in admin/
ought to be included for ethical and/or GPL reasons, and the story will
go on.  Since admin/ is "only" 2.5 Mb, we might as well just distribute
it.

Re: No sources for lisp/international/uni-*.el?

[Eli Zaretskii]

> From: Chong Yidong <cyd@stupidchicken.com>
> Cc: rms@gnu.org, dak@gnu.org, emacs-devel@gnu.org
> Date: Wed, 10 Aug 2011 12:58:07 -0400
> 
> Since admin/ is "only" 2.5 Mb, we might as well just distribute it.

There are files there that violate 8+3 restrictions.

Re: No sources for lisp/international/uni-*.el?

[Chong Yidong]

Eli Zaretskii <eliz@gnu.org> writes:

>> Since admin/ is "only" 2.5 Mb, we might as well just distribute it.
>
> There are files there that violate 8+3 restrictions.

Which ones?

Re: No sources for lisp/international/uni-*.el?

[Eli Zaretskii]

> Date: Wed, 10 Aug 2011 20:19:59 +0300
> From: Eli Zaretskii <eliz@gnu.org>
> Cc: dak@gnu.org, rms@gnu.org, emacs-devel@gnu.org
> 
> > From: Chong Yidong <cyd@stupidchicken.com>
> > Cc: rms@gnu.org, dak@gnu.org, emacs-devel@gnu.org
> > Date: Wed, 10 Aug 2011 12:58:07 -0400
> > 
> > Since admin/ is "only" 2.5 Mb, we might as well just distribute it.
> 
> There are files there that violate 8+3 restrictions.

In any case, I see no reasons to distribute these:

  admin.el
  alloc-colors.c
  build-configs
  bzrmerge.el
  check-doc-strings
  CPP-DEFINES
  cus-test.el
  diff-tar-files
  emacs-pretesters
  FOR-RELEASE
  MAINTAINERS
  make-announcement
  make-changelog-diff
  make-emacs
  make-tarball.txt
  quick-install-emacs

  charsets/mapfiles/*

  notes/*

  nt/*

  unidata/UnicodeData.txt

Re: No sources for lisp/international/uni-*.el?

[Eli Zaretskii]

> From: Chong Yidong <cyd@stupidchicken.com>
> Cc: rms@gnu.org, dak@gnu.org, emacs-devel@gnu.org
> Date: Wed, 10 Aug 2011 13:36:13 -0400
> 
> Eli Zaretskii <eliz@gnu.org> writes:
> 
> >> Since admin/ is "only" 2.5 Mb, we might as well just distribute it.
> >
> > There are files there that violate 8+3 restrictions.
> 
> Which ones?

Here's the relevant part of output from doschk:

  The following files are not valid DOS file names:
  admin/charsets/mapfiles/bulgarian-mik.txt.gz - too many dots
  admin/charsets/mapfiles/cns2ucsdkw.txt.gz - too many dots
  admin/charsets/mapfiles/CP720.map.gz - too many dots
  admin/charsets/mapfiles/CP858.map.gz - too many dots
  admin/charsets/mapfiles/CP932.TXT.gz - too many dots
  admin/charsets/mapfiles/JISX213A.map.gz - too many dots
  admin/charsets/mapfiles/MULE-ethiopic.map.gz - too many dots
  admin/charsets/mapfiles/MULE-ipa.map.gz - too many dots
  admin/charsets/mapfiles/MULE-is13194.map.gz - too many dots
  admin/charsets/mapfiles/MULE-lviscii.map.gz - too many dots
  admin/charsets/mapfiles/MULE-sisheng.map.gz - too many dots
  admin/charsets/mapfiles/MULE-tibetan.map.gz - too many dots
  admin/charsets/mapfiles/MULE-uviscii.map.gz - too many dots
  admin/charsets/mapfiles/stdenc.txt.gz - too many dots
  admin/charsets/mapfiles/symbol.txt.gz - too many dots

Personally, I think that the files in charsets/mapfiles do not need to
be distributed at all, but that's me.

Re: No sources for lisp/international/uni-*.el?

[Chong Yidong]

Eli Zaretskii <eliz@gnu.org> writes:

> Here's the relevant part of output from doschk:
>
>   The following files are not valid DOS file names:
>   admin/charsets/mapfiles/bulgarian-mik.txt.gz - too many dots
>   admin/charsets/mapfiles/cns2ucsdkw.txt.gz - too many dots
>   admin/charsets/mapfiles/CP720.map.gz - too many dots
>   admin/charsets/mapfiles/CP858.map.gz - too many dots
>   admin/charsets/mapfiles/CP932.TXT.gz - too many dots
>   admin/charsets/mapfiles/JISX213A.map.gz - too many dots
>   admin/charsets/mapfiles/MULE-ethiopic.map.gz - too many dots
>   admin/charsets/mapfiles/MULE-ipa.map.gz - too many dots
>   admin/charsets/mapfiles/MULE-is13194.map.gz - too many dots
>   admin/charsets/mapfiles/MULE-lviscii.map.gz - too many dots
>   admin/charsets/mapfiles/MULE-sisheng.map.gz - too many dots
>   admin/charsets/mapfiles/MULE-tibetan.map.gz - too many dots
>   admin/charsets/mapfiles/MULE-uviscii.map.gz - too many dots
>   admin/charsets/mapfiles/stdenc.txt.gz - too many dots
>   admin/charsets/mapfiles/symbol.txt.gz - too many dots

Thanks.  Gunzipping these brings the total for admin/ to "only" 4.3 Mb,
which we can live with.

> Personally, I think that the files in charsets/mapfiles do not need to
> be distributed at all, but that's me.

If UnicodeData.txt is distributed (as OP requested), then those files
ought to be, too.  If we should distribute some of the files in admin/,
then I would rather distribute that entire directory, instead of picking
and choosing.

I'll defer the distribute-or-not question to RMS.

Re: No sources for lisp/international/uni-*.el?

[Eli Zaretskii]

> From: Chong Yidong <cyd@stupidchicken.com>
> Cc: rms@gnu.org, dak@gnu.org, emacs-devel@gnu.org
> Date: Wed, 10 Aug 2011 13:48:39 -0400
> 
> > Personally, I think that the files in charsets/mapfiles do not need to
> > be distributed at all, but that's me.
> 
> If UnicodeData.txt is distributed (as OP requested), then those files
> ought to be, too.

I don't think UnicodeData.txt needs to be distributed, and I didn't
understand that Richard requested that.  The file can be downloaded at
any time by anyone who needs it.  I understood that the problem was
with unidata-gen.el, not with the database.

> If we should distribute some of the files in admin/, then I would
> rather distribute that entire directory, instead of picking and
> choosing.

We can move the files we distribute to another directory.

Re: No sources for lisp/international/uni-*.el?

[Richard Stallman]

For the long term, one possible solution could be to use C code to read
UnicodeData.txt directly instead of generating other files from it.

It MIGHT even be faster, though I can't be sure of that.

-- 
Dr Richard Stallman
President, Free Software Foundation
51 Franklin St
Boston MA 02110
USA
www.fsf.org  www.gnu.org
Skype: No way! That's nonfree (freedom-denying) software.
  Use free telephony http://directory.fsf.org/category/tel/

Re: No sources for lisp/international/uni-*.el?

[Richard Stallman]

    I don't think there is a reason other than the fact that admin/ was the
    best place to put UnicodeData.txt (which is a 1.4Mb file that is not
    used directly by Emacs itself, and must not be changed apart from
    updating to new Unicode releases, so it doesn't make sense to put in
    etc/).

Aside, but it's important: why do you say "must not be changed"?  It
is free -- it could be changed.  We probably will never have a good
reason to change it, so we will probably never change it, but that is
not the same as "must not".

-- 
Dr Richard Stallman
President, Free Software Foundation
51 Franklin St
Boston MA 02110
USA
www.fsf.org  www.gnu.org
Skype: No way! That's nonfree (freedom-denying) software.
  Use free telephony http://directory.fsf.org/category/tel/

Re: No sources for lisp/international/uni-*.el?

[Richard Stallman]

    Maybe we should just add the admin/ directory to the source tarballs.
    WDYT?

That is an option, but adding just the admin/unidata directory would
be enough.

If consider those as part of Emacs, so that the GPL applies, then
the GPL requires inclusion of unidata-gen.el because it's a script
for compilation, and likewise requires UnicodeData.txt because
it's the source they are compiled from.

Here's an idea.  We could make a second tarball with admin/unidata, or
perhaps all of admin, and say that the full Emacs sources consist of
both together.  People who redistribute source ought to redistribute both,
but users won't have to download both.

-- 
Dr Richard Stallman
President, Free Software Foundation
51 Franklin St
Boston MA 02110
USA
www.fsf.org  www.gnu.org
Skype: No way! That's nonfree (freedom-denying) software.
  Use free telephony http://directory.fsf.org/category/tel/

Re: No sources for lisp/international/uni-*.el?

[Chong Yidong]

Richard Stallman <rms@gnu.org> writes:

>     I don't think there is a reason other than the fact that admin/ was the
>     best place to put UnicodeData.txt (which is a 1.4Mb file that is not
>     used directly by Emacs itself, and must not be changed apart from
>     updating to new Unicode releases, so it doesn't make sense to put in
>     etc/).
>
> Aside, but it's important: why do you say "must not be changed"?  It
> is free -- it could be changed.  We probably will never have a good
> reason to change it, so we will probably never change it, but that is
> not the same as "must not".

I used "must not" in the sense of "should not, unless you are insane".

Re: No sources for lisp/international/uni-*.el?

[Chong Yidong]

Richard Stallman <rms@gnu.org> writes:

> If consider those as part of Emacs, so that the GPL applies, then the
> GPL requires inclusion of unidata-gen.el because it's a script for
> compilation, and likewise requires UnicodeData.txt because it's the
> source they are compiled from.

From a technical point of view, AFAICT there is no reason to prefer
either state of affairs, i.e. between treating the Unicode data as
something distributed alongside Emacs vs part of Emacs.

> Here's an idea.  We could make a second tarball with admin/unidata, or
> perhaps all of admin, and say that the full Emacs sources consist of
> both together.  People who redistribute source ought to redistribute
> both, but users won't have to download both.

This would be an inconvenience to redistributors.  Simply including
admin/ in the source tarball is just an extra 4 Mb (unzipped), which
doesn't seem unreasonable.  So I prefer to go that route.

Re: No sources for lisp/international/uni-*.el?

[Richard Stallman]

    This would be an inconvenience to redistributors.  Simply including
    admin/ in the source tarball is just an extra 4 Mb (unzipped), which
    doesn't seem unreasonable.  So I prefer to go that route.

I have nothing against it.

-- 
Dr Richard Stallman
President, Free Software Foundation
51 Franklin St
Boston MA 02110
USA
www.fsf.org  www.gnu.org
Skype: No way! That's nonfree (freedom-denying) software.
  Use free telephony http://directory.fsf.org/category/tel/

Re: No sources for lisp/international/uni-*.el?

[Eli Zaretskii]

> Date: Fri, 12 Aug 2011 13:56:18 -0400
> From: Richard Stallman <rms@gnu.org>
> Cc: dak@gnu.org, emacs-devel@gnu.org
> 
>     This would be an inconvenience to redistributors.  Simply including
>     admin/ in the source tarball is just an extra 4 Mb (unzipped), which
>     doesn't seem unreasonable.  So I prefer to go that route.
> 
> I have nothing against it.

But do we really need to distribute the Unicode database, and every
other similar data file?  These days they are freely available from
their original creators, and in fact the connectivity to those sites
is better than to the GNU sites.  Why should we need to bloat our
tarballs by these data files?  They are not software.