Patchwork manual: Remove warning in the documentation of the abort function

login
register
mail settings
Submitter Florian Weimer
Date Oct. 7, 2019, 5:32 p.m.
Message ID <87a7ach367.fsf@oldenburg2.str.redhat.com>
Download mbox | patch
Permalink /patch/34850/
State New
Headers show

Comments

Florian Weimer - Oct. 7, 2019, 5:32 p.m.
The warning is confusing to those who do not understand the context,
and the warning is easy to misunderstand:

A reader needs to know that it was written by someone who is generally
skeptical of government influence and control, otherwise it reads as
an affirmation of the U.S. government's role as the ultimate editor of
the manual.  This is precisely the opposite of what the warning
intends to convey.  (Reportedly, it criticizes that several
U.S. administrations have tried to restrict the medical advice that
U.S.-funded health care workers can provide abroad, considering that
censorship.)

The warning is also misleading on a technical level.  A reader who
makes the connection to pregnancy termination will get the wrong
impression that calling the abort function will terminate subprocesses
of the current process, but this is not what generally happens.

Finally, for both GNU and the FSF, it is inappropriate to use female
reproductive health as mere joke material, since these organizations
do not concern themselves with such issues otherwise, and the warning
is purportedly about something else entirely.

This reinstates commit 340d9652b9d0e1d4136588f18b726662d195777c
("manual/startup.texi (Aborting a Program): Remove inappropriate
joke."), effectively reverting the revert in commit
ffa81c22a3ac0fb75ad9bf2b1c3cdbf9eafa0bc9 ("Revert:").

2019-10-07  Florian Weimer  <fweimer@redhat.com>

	* manual/startup.texi (Aborting a Program): Remove warning.
Paul Eggert - Oct. 7, 2019, 7:28 p.m.
Thanks, this change looks good to me, both for the reasons you mentioned 
and for other reasons discussed when the topic came up before.

For those who'd like a review of this controversial topic, here is a 
list of some of the relevant messages; each message starts a thread that 
can be quite lengthy. I've read all the messages in these threads, and 
taken collectively it does appear that although there are strong 
opinions on both sides, removal would be in the best interests of the 
GNU project.

https://sourceware.org/ml/libc-alpha/2018-04/msg00600.html
https://sourceware.org/ml/libc-alpha/2018-05/msg00001.html
https://sourceware.org/ml/libc-alpha/2018-05/msg00212.html
https://sourceware.org/ml/libc-alpha/2018-05/msg00291.html
https://sourceware.org/ml/libc-alpha/2018-05/msg00453.html
Joseph Myers - Oct. 7, 2019, 8:44 p.m.
I support this change, as removing something that has been confusing in 
practice to readers of the manual and does not in any way benefit those 
users in understanding the GNU C Library functionality.  While humor is 
explicitly accepted in GNU, it is also explicitly not required, and in 
this particular instance it has proven actively unhelpful.  Similarly, 
while "It is ok to refer once in a rare while to spatially or temporally 
localized reference points or facts, if it is directly pertinent or as an 
aside.", in this particular case it impedes understanding of the technical 
content for readers without the relevant cultural reference points to 
understand that it is intended as humorous content rather than information 
about the abort function.  Finally, the original reasons for inclusion of 
this joke have been thoroughly explored in previous discussions, and so 
the current glibc maintainers have all the information needed about those 
reasons to take into account in a decision to override a 1992 decision of 
a previous maintainer to include it at that time.
Carlos O'Donell - Oct. 7, 2019, 9 p.m.
On 10/7/19 1:32 PM, Florian Weimer wrote:
> The warning is confusing to those who do not understand the context,
> and the warning is easy to misunderstand:

I agree. I made the same comments myself in the original post about the issue,
and backed that up with a stackoverflow and reddit post showing confusion over
the statement in the manual.

> A reader needs to know that it was written by someone who is generally
> skeptical of government influence and control, otherwise it reads as
> an affirmation of the U.S. government's role as the ultimate editor of
> the manual.  This is precisely the opposite of what the warning
> intends to convey.  (Reportedly, it criticizes that several
> U.S. administrations have tried to restrict the medical advice that
> U.S.-funded health care workers can provide abroad, considering that
> censorship.)

Ageed.

> The warning is also misleading on a technical level.  A reader who
> makes the connection to pregnancy termination will get the wrong
> impression that calling the abort function will terminate subprocesses
> of the current process, but this is not what generally happens.

Agreed.

> Finally, for both GNU and the FSF, it is inappropriate to use female
> reproductive health as mere joke material, since these organizations
> do not concern themselves with such issues otherwise, and the warning
> is purportedly about something else entirely.

Agreed.

> This reinstates commit 340d9652b9d0e1d4136588f18b726662d195777c
> ("manual/startup.texi (Aborting a Program): Remove inappropriate
> joke."), effectively reverting the revert in commit
> ffa81c22a3ac0fb75ad9bf2b1c3cdbf9eafa0bc9 ("Revert:").

This looks good to me.

Reviewed-by: Carlos O'Donell <carlos@redhat.com>

> 2019-10-07  Florian Weimer  <fweimer@redhat.com>
> 
> 	* manual/startup.texi (Aborting a Program): Remove warning.
> 
> diff --git a/manual/startup.texi b/manual/startup.texi
> index 7395d32dd0..21c48cd037 100644
> --- a/manual/startup.texi
> +++ b/manual/startup.texi
> @@ -1005,14 +1005,6 @@ This function actually terminates the process by raising a
>  intercept this signal; see @ref{Signal Handling}.
>  @end deftypefun
>  
> -@c Put in by rms.  Don't remove.
> -@cartouche
> -@strong{Future Change Warning:} Proposed Federal censorship regulations
> -may prohibit us from giving you information about the possibility of
> -calling this function.  We would be required to say that this is not an
> -acceptable way of terminating a program.
> -@end cartouche
> -

OK. Removes the entire cartouche.

>  @node Termination Internals
>  @subsection Termination Internals
>  
>
Alexandre Oliva - Oct. 8, 2019, midnight
On Oct  7, 2019, Florian Weimer <fweimer@redhat.com> wrote:

> The warning is confusing to those who do not understand the context,
> and the warning is easy to misunderstand:

I don't think it's appropriate at all to bring this issue back at this
time.  It's too much of an shamelessly opportunistic move to take
advantage of the mob lynching underway.  Please hold this discussion for
another, more sensible time.
Yann Droneaud - Oct. 8, 2019, 8:12 a.m.
Le lundi 07 octobre 2019 à 19:32 +0200, Florian Weimer a écrit :
> The warning is confusing to those who do not understand the context,
> and the warning is easy to misunderstand:
> 
> A reader needs to know that it was written by someone who is
> generally
> skeptical of government influence and control, otherwise it reads as
> an affirmation of the U.S. government's role as the ultimate editor
> of
> the manual.  This is precisely the opposite of what the warning
> intends to convey.  (Reportedly, it criticizes that several
> U.S. administrations have tried to restrict the medical advice that
> U.S.-funded health care workers can provide abroad, considering that
> censorship.)
> 
> The warning is also misleading on a technical level.  A reader who
> makes the connection to pregnancy termination will get the wrong
> impression that calling the abort function will terminate
> subprocesses
> of the current process, but this is not what generally happens.
> 
> Finally, for both GNU and the FSF, it is inappropriate to use female
> reproductive health as mere joke material, since these organizations
> do not concern themselves with such issues otherwise, and the warning
> is purportedly about something else entirely.
> 
> This reinstates commit 340d9652b9d0e1d4136588f18b726662d195777c
> ("manual/startup.texi (Aborting a Program): Remove inappropriate
> joke."), effectively reverting the revert in commit
> ffa81c22a3ac0fb75ad9bf2b1c3cdbf9eafa0bc9 ("Revert:").
> 
> 2019-10-07  Florian Weimer  <fweimer@redhat.com>
> 
> 	* manual/startup.texi (Aborting a Program): Remove warning.
> 
> diff --git a/manual/startup.texi b/manual/startup.texi
> index 7395d32dd0..21c48cd037 100644
> --- a/manual/startup.texi
> +++ b/manual/startup.texi
> @@ -1005,14 +1005,6 @@ This function actually terminates the process
> by raising a
>  intercept this signal; see @ref{Signal Handling}.
>  @end deftypefun
>  
> -@c Put in by rms.  Don't remove.
> -@cartouche
> -@strong{Future Change Warning:} Proposed Federal censorship
> regulations
> -may prohibit us from giving you information about the possibility of
> -calling this function.  We would be required to say that this is not
> an
> -acceptable way of terminating a program.
> -@end cartouche
> -
>  @node Termination Internals
>  @subsection Termination Internals
>  

LGTM, +1

Thanks.
Adhemerval Zanella Netto - Oct. 8, 2019, 5:48 p.m.
On 07/10/2019 14:32, Florian Weimer wrote:
> The warning is confusing to those who do not understand the context,
> and the warning is easy to misunderstand:
> 
> A reader needs to know that it was written by someone who is generally
> skeptical of government influence and control, otherwise it reads as
> an affirmation of the U.S. government's role as the ultimate editor of
> the manual.  This is precisely the opposite of what the warning
> intends to convey.  (Reportedly, it criticizes that several
> U.S. administrations have tried to restrict the medical advice that
> U.S.-funded health care workers can provide abroad, considering that
> censorship.)
> 
> The warning is also misleading on a technical level.  A reader who
> makes the connection to pregnancy termination will get the wrong
> impression that calling the abort function will terminate subprocesses
> of the current process, but this is not what generally happens.
> 
> Finally, for both GNU and the FSF, it is inappropriate to use female
> reproductive health as mere joke material, since these organizations
> do not concern themselves with such issues otherwise, and the warning
> is purportedly about something else entirely.
> 
> This reinstates commit 340d9652b9d0e1d4136588f18b726662d195777c
> ("manual/startup.texi (Aborting a Program): Remove inappropriate
> joke."), effectively reverting the revert in commit
> ffa81c22a3ac0fb75ad9bf2b1c3cdbf9eafa0bc9 ("Revert:").
> 
> 2019-10-07  Florian Weimer  <fweimer@redhat.com>
> 
> 	* manual/startup.texi (Aborting a Program): Remove warning.

The subject has already being discussed to exhaustion and the previous
requested cooldown period has been granted. Let's moved on.

LGTM.

Reviewed-by: Adhemerval Zanella  <adhemerval.zanella@linaro.org>

> 
> diff --git a/manual/startup.texi b/manual/startup.texi
> index 7395d32dd0..21c48cd037 100644
> --- a/manual/startup.texi
> +++ b/manual/startup.texi
> @@ -1005,14 +1005,6 @@ This function actually terminates the process by raising a
>  intercept this signal; see @ref{Signal Handling}.
>  @end deftypefun
>  
> -@c Put in by rms.  Don't remove.
> -@cartouche
> -@strong{Future Change Warning:} Proposed Federal censorship regulations
> -may prohibit us from giving you information about the possibility of
> -calling this function.  We would be required to say that this is not an
> -acceptable way of terminating a program.
> -@end cartouche
> -
>  @node Termination Internals
>  @subsection Termination Internals
>  
>
Richard Stallman - Oct. 8, 2019, 8:42 p.m.
[[[ To any NSA and FBI agents reading my email: please consider    ]]]
[[[ whether defending the US Constitution against all enemies,     ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]

Before deciding this question, there are several questions, more basic
and general, that we need to tackle first.  The answers to them will
be the basis for this specific decision.

We have addressed three or four of them.  Perhaps only one remains.

With the current disturbance, this is not the time to discuss it.
But we should be able to take it up in a few months.  Then we could
decide this question.

Unless, that is, Republicans impose a decision on us before then.
At the rate things are going, it wouldn't surprise me.  ;-{.
Joseph Myers - Oct. 8, 2019, 9:10 p.m.
On Tue, 8 Oct 2019, Richard Stallman wrote:

> Before deciding this question, there are several questions, more basic
> and general, that we need to tackle first.  The answers to them will
> be the basis for this specific decision.

I think the conclusion is clear based on the answers so far; the further 
general discussions might help answer other questions, but they are not 
needed for this one.

This is not a tricky technical design question where the choice of an 
interface now will affect large amounts of code built on top of it and we 
need to take care to avoid choosing an interface that will be hard to 
maintain compatibly in future, or to avoid an interface that doesn't 
properly achieve its goals and will need a new interface added while 
keeping the old one around as well.  Nothing else depends on the details 
of which particular jokes are in the manual.  The glibc maintainers have 
plenty of information in the past discussions to come to a conclusion now 
on this issue.

For example, I think this case is clear enough that we don't need to 
restart the discussion of general principles around when it is or is not 
appropriate for current package maintainers to change decisions made by 
previous maintainers in order to reach a conclusion to apply to this case, 
even if that discussion might be helpful in other such less clear cases in 
future.
Florian Weimer - Oct. 8, 2019, 9:20 p.m.
* Richard Stallman:

> Before deciding this question, there are several questions, more basic
> and general, that we need to tackle first.  The answers to them will
> be the basis for this specific decision.
>
> We have addressed three or four of them.  Perhaps only one remains.
>
> With the current disturbance, this is not the time to discuss it.
> But we should be able to take it up in a few months.  Then we could
> decide this question.

There is always some excuse.  We have waited for more than a year.

You said repeatedly that you are still in charge in the GNU project, and
that GNU is not a democracy.  So this should really be simple.  Just say
you want the warning gone, and it will be gone.  Or you tell us you want
to keep it.

What is so difficult about this?

> Unless, that is, Republicans impose a decision on us before then.
> At the rate things are going, it wouldn't surprise me.  ;-{.

This has nothing to do with U.S. politics.  Not everything revolves
around the USA.  We ended up with this problem on our own.

Thanks,
Florian
Joseph Myers - Oct. 8, 2019, 9:40 p.m.
On Tue, 8 Oct 2019, Florian Weimer wrote:

> You said repeatedly that you are still in charge in the GNU project, and
> that GNU is not a democracy.  So this should really be simple.  Just say
> you want the warning gone, and it will be gone.  Or you tell us you want
> to keep it.

And if Richard wants it kept, I consider that just a previous maintainer's 
view to be considered by the maintainers.  This is not a difficult 
question about the exact boundaries of when it is appropriate for GNU to 
intervene in decisions made by individual package maintainers; it's a case 
for which it's extremely clear-cut that it's properly within the scope of 
package maintainers to make the final decision and would be improper for 
GNU to attempt to overrule such a decision.  The precise boundaries of 
appropriate intervention in individual packages may be worth considering 
elsewhere, but that's not relevant for this case.
Zack Weinberg - Oct. 8, 2019, 9:48 p.m.
On Mon, Oct 7, 2019 at 1:32 PM Florian Weimer <fweimer@redhat.com> wrote:
> This reinstates commit 340d9652b9d0e1d4136588f18b726662d195777c
> ("manual/startup.texi (Aborting a Program): Remove inappropriate
> joke."), effectively reverting the revert in commit
> ffa81c22a3ac0fb75ad9bf2b1c3cdbf9eafa0bc9 ("Revert:").
>
> 2019-10-07  Florian Weimer  <fweimer@redhat.com>
>
>         * manual/startup.texi (Aborting a Program): Remove warning.

As the person who committed 340d9652b9d0e1d4136588f18b726662d195777c
in the first place, I endorse its reinstatement.

I wish to say for the record that I think the overrule of the glibc
maintainers' original decision, exercised by RMS in the lengthy thread
last May (representative message:
<https://sourceware.org/ml/libc-alpha/2018-05/msg00001.html>), was an
illegitimate use of authority, REGARDLESS of whether he is to continue
as "Chief GNUisance" in the future, or whether the GNU Project should
have a person empowered to make such unilateral overruling decisions.
As the original author of the text in question he has a conflict of
interest and he should have either accepted the decision of the
maintainers or referred the issue to someone both sides could agree
would act as a neutral arbiter.

Because it was an illegitimate act, we should not consider ourselves
bound by it.

zw
Alexandre Oliva - Oct. 9, 2019, 12:03 p.m.
On Oct  8, 2019, Florian Weimer <fweimer@redhat.com> wrote:

> What is so difficult about this?

This patch has proven before to be an inflammatory topic.

The current state of affairs is also prone to inflammatory discussions.

It is just not possible to hold a sane discussion about this patch at
this time.

I therefore insist that the discussion about this patch be posponed to
some time in which moods are not so inflamed to begin with.
Siddhesh Poyarekar - Oct. 9, 2019, 2:03 p.m.
On 09/10/19 8:03 am, Alexandre Oliva wrote:
> On Oct  8, 2019, Florian Weimer <fweimer@redhat.com> wrote:
> 
>> What is so difficult about this?
> 
> This patch has proven before to be an inflammatory topic.
> 
> The current state of affairs is also prone to inflammatory discussions.
> 
> It is just not possible to hold a sane discussion about this patch at
> this time.
> 
> I therefore insist that the discussion about this patch be posponed to
> some time in which moods are not so inflamed to begin with.
> 

The two issues are unrelated.  You are only making things worse for
yourself and the community by objecting to removal of a joke (A JOKE!)
from a technical manual that multiple people on list and on social media
have flagged as offensive.

If you're so passionate about *jokes* on abortion rights then I'd
suggest posting a separate patch that proposes inclusion of a new joke
that does not offend people.  If you can't think of one then that's OK
too; while humour is nice to have in documentation, it's not absolutely
necessary.

Siddhesh
Samuel Thibault - Oct. 9, 2019, 3:10 p.m.
Hello,

+1 on this.

Samuel

Florian Weimer, le lun. 07 oct. 2019 19:32:32 +0200, a ecrit:
> The warning is confusing to those who do not understand the context,
> and the warning is easy to misunderstand:
> 
> A reader needs to know that it was written by someone who is generally
> skeptical of government influence and control, otherwise it reads as
> an affirmation of the U.S. government's role as the ultimate editor of
> the manual.  This is precisely the opposite of what the warning
> intends to convey.  (Reportedly, it criticizes that several
> U.S. administrations have tried to restrict the medical advice that
> U.S.-funded health care workers can provide abroad, considering that
> censorship.)
> 
> The warning is also misleading on a technical level.  A reader who
> makes the connection to pregnancy termination will get the wrong
> impression that calling the abort function will terminate subprocesses
> of the current process, but this is not what generally happens.
> 
> Finally, for both GNU and the FSF, it is inappropriate to use female
> reproductive health as mere joke material, since these organizations
> do not concern themselves with such issues otherwise, and the warning
> is purportedly about something else entirely.
> 
> This reinstates commit 340d9652b9d0e1d4136588f18b726662d195777c
> ("manual/startup.texi (Aborting a Program): Remove inappropriate
> joke."), effectively reverting the revert in commit
> ffa81c22a3ac0fb75ad9bf2b1c3cdbf9eafa0bc9 ("Revert:").
> 
> 2019-10-07  Florian Weimer  <fweimer@redhat.com>
> 
> 	* manual/startup.texi (Aborting a Program): Remove warning.
> 
> diff --git a/manual/startup.texi b/manual/startup.texi
> index 7395d32dd0..21c48cd037 100644
> --- a/manual/startup.texi
> +++ b/manual/startup.texi
> @@ -1005,14 +1005,6 @@ This function actually terminates the process by raising a
>  intercept this signal; see @ref{Signal Handling}.
>  @end deftypefun
>  
> -@c Put in by rms.  Don't remove.
> -@cartouche
> -@strong{Future Change Warning:} Proposed Federal censorship regulations
> -may prohibit us from giving you information about the possibility of
> -calling this function.  We would be required to say that this is not an
> -acceptable way of terminating a program.
> -@end cartouche
> -
>  @node Termination Internals
>  @subsection Termination Internals
>
Alexandre Oliva - Oct. 9, 2019, 5:17 p.m.
On Oct  9, 2019, Siddhesh Poyarekar <siddhesh@gotplt.org> wrote:

> The two issues are unrelated.

Even if they were, and that this coming up right now was just an
incredibly sorry coincidence, you really think the time to hold a
rational and sane debate is one in which some people are scared to be
beaten up for holding controversial opinions that are not in line with
that of the lynching mob, while others have already been beaten up for
speaking their minds?  Seriously?
Siddhesh Poyarekar - Oct. 9, 2019, 5:55 p.m.
On 09/10/19 1:17 pm, Alexandre Oliva wrote:
> On Oct  9, 2019, Siddhesh Poyarekar <siddhesh@gotplt.org> wrote:
> 
>> The two issues are unrelated.
> 
> Even if they were, and that this coming up right now was just an
> incredibly sorry coincidence, you really think the time to hold a
> rational and sane debate is one in which some people are scared to be
> beaten up for holding controversial opinions that are not in line with
> that of the lynching mob, while others have already been beaten up for
> speaking their minds?  Seriously?

My argument is that I have complete faith in your ability to stall
discussions forever and twist words to suit your purpose.

IMO the issue here is clear and has nothing to do with the intent of the
joke.  It has to do with feature creep of the FSF's fight against
censorship into the GNU projects manifesto of providing Software
Freedom.  Nothing in the GNU manifesto talks about censorship so this is
literally the only bit in the GNU manifesto and glibc combined that
talks about censorship.

The only barely valid argument I have seen in allowing such feature
creep from FSF to GNU was that RMS practically owns FSF and GNU and can
claim it's one and the same thing.  Using that to justify keeping the
text IMO is a sure road to obsolescence for the GNU project in the
coming years regardless of whether all of us continue development here.

Siddhesh
Alexandre Oliva - Oct. 9, 2019, 6:18 p.m.
On Oct  9, 2019, Siddhesh Poyarekar <siddhesh@gotplt.org> wrote:

> The only barely valid argument I have seen in allowing such feature
> creep from FSF to GNU

?!?  What do you imagine the FSF has to do with any of this?
Florian Weimer - Oct. 9, 2019, 6:24 p.m.
* Alexandre Oliva:

> On Oct  9, 2019, Siddhesh Poyarekar <siddhesh@gotplt.org> wrote:
>
>> The two issues are unrelated.
>
> Even if they were, and that this coming up right now was just an
> incredibly sorry coincidence, you really think the time to hold a
> rational and sane debate is one in which some people are scared to be
> beaten up for holding controversial opinions that are not in line with
> that of the lynching mob, while others have already been beaten up for
> speaking their minds?  Seriously?

I'm sorry, but you do not seem to understand the issue at all.

From my perspective, any chance of a rational and sane debate ceased
when you committed the revert in May 2018.  This was my worst day at Red
Hat, worse than when I was told that the team I was originally hired for
had been disbanded.  (Alexandre and I worked both on the Red Hat
Platform Tools team back then, albeit working on different things, which
is why this was also work-related matter for me.)  It took *a lot* of
effort from Carlos and others to prevent the glibc project from
disintegrating on the spot (and yes, I see it as a project on its own,
not merely a piece of code under GNU).

As I see things, here is what's going on:

The de-facto glibc maintainers want the warning removed.

The warning is offensive to many people inside and outside the GNU
community.

As a result, a GNU project decision not to remove the warning would
alienate many people.  Some of them would object to the warning itself,
some of them to the way that the glibc maintainer wishes were not
respected.

Due to how things stand today, inaction is still a decision, against
removal.

Thanks,
Florian
Siddhesh Poyarekar - Oct. 9, 2019, 6:28 p.m.
On 09/10/19 2:18 pm, Alexandre Oliva wrote:
> On Oct  9, 2019, Siddhesh Poyarekar <siddhesh@gotplt.org> wrote:
> 
>> The only barely valid argument I have seen in allowing such feature
>> creep from FSF to GNU
> 
> ?!?  What do you imagine the FSF has to do with any of this?
> 

The FSF actively speaks out against censorship and talking about
abortion rights and censorship regarding it is within its scope.  That
is not true for the GNU project because nothing in the GNU project,
especially its manifesto has any mention of censorship or fight against
censorship as its goal.  So within the context of the GNU project, this
abort joke is literally the only thing that mentions abortion rights and
censorship without any context whatsoever other than the fact that the
function happens to be named abort.

In other words, there is no ideological relationship of the abortion
rights joke with the GNU project unless you or RMS decide to add it.

Siddhesh
Alexandre Oliva - Oct. 10, 2019, 1:14 a.m.
Hello, Siddhesh,

Sorry this is a little terse; typed from the airport before boarding.

On Oct  9, 2019, Siddhesh Poyarekar <siddhesh@gotplt.org> wrote:

> That is not true for the GNU project because nothing in the GNU
> project, especially its manifesto has any mention of censorship or
> fight against censorship as its goal.

So let's see what Free Software is about, shall we?

You might recall freedom #1: the freedom to study the source code and
adapt the software so that it does what you wish.  That amounts to free
speech in code: you are free to take preexisting code and then express
whatever idea you see fit in it.

Freedom #0: the freedom to run the software for any purpose.  That
amounts to putting the ideas expressed in the software in action.
Furthermore, if the software is written so that running it conveys
ideas, it's actual free speech, expressed through the software.

Freedom #3: the freedom to improve the software, and to distribute your
improvements when you wish.  That amounts to conveying to others the
ideas you expressed in (vs through, already covered in #0) the software.

Do you see now that censorship, being the opposite of free speech, is
indeed something that the whole Free Software movement stands against?
Alexandre Oliva - Oct. 10, 2019, 1:20 a.m.
On Oct  9, 2019, Florian Weimer <fweimer@redhat.com> wrote:

> As a result, a GNU project decision not to remove the warning would
> alienate many people.

A decision to remove it also problematic.  It's no wonder several
complex conversations pertaining to this very topic have been started
and debated extensively in other GNU fora.

Now, I can't imagine whatever gave you the idea that now would be a good
time to add to the already boiling cauldron, so to speak, all of the
hurt feelings from the earlier discussion surrounding this patch.

It really isn't.  There is no possibility of a sane debate about it at
this moment.  Sorry, I really think it should wait a little longer.  I
acknowledge that there probably won't ever be a *good* time to discuss
this patch, because it does touch a number of inflammatory issues, but
that doesn't justify bringing it up at what is likely the worst possible
time.
Siddhesh Poyarekar - Oct. 10, 2019, 1:32 a.m.
On 09/10/19 9:14 pm, Alexandre Oliva wrote:
> So let's see what Free Software is about, shall we?
> 
> You might recall freedom #1: the freedom to study the source code and
> adapt the software so that it does what you wish.  That amounts to free
> speech in code: you are free to take preexisting code and then express
> whatever idea you see fit in it.
> 
> Freedom #0: the freedom to run the software for any purpose.  That
> amounts to putting the ideas expressed in the software in action.
> Furthermore, if the software is written so that running it conveys
> ideas, it's actual free speech, expressed through the software.
> 
> Freedom #3: the freedom to improve the software, and to distribute your
> improvements when you wish.  That amounts to conveying to others the
> ideas you expressed in (vs through, already covered in #0) the software.
> 
> Do you see now that censorship, being the opposite of free speech, is
> indeed something that the whole Free Software movement stands against?

Do you not see that the freedoms specifically talk about software?  How
is abortion rights or censorship about abortion related to software?

You're not making any sense at this point.

Siddhesh
Siddhesh Poyarekar - Oct. 10, 2019, 1:43 a.m.
On 09/10/19 9:20 pm, Alexandre Oliva wrote:
> On Oct  9, 2019, Florian Weimer <fweimer@redhat.com> wrote:
> 
>> As a result, a GNU project decision not to remove the warning would
>> alienate many people.
> 
> A decision to remove it also problematic.  It's no wonder several
> complex conversations pertaining to this very topic have been started
> and debated extensively in other GNU fora.
> 
> Now, I can't imagine whatever gave you the idea that now would be a good
> time to add to the already boiling cauldron, so to speak, all of the
> hurt feelings from the earlier discussion surrounding this patch.
> 
> It really isn't.  There is no possibility of a sane debate about it at
> this moment.  Sorry, I really think it should wait a little longer.  I
> acknowledge that there probably won't ever be a *good* time to discuss
> this patch, because it does touch a number of inflammatory issues, but
> that doesn't justify bringing it up at what is likely the worst possible
> time.

I strongly suggest we undo the reversion and then wait for the good time
to discuss bringing the text back; you've held the community hostage
long enough.

I don't think it makes any sense to wait any longer because there is
sustained support from a significant majority of the community to get
rid of the warning and my faith in your and RMS' ability to negotiate a
conclusion that I can personally agree with is running out.

Siddhesh
Richard Stallman - Oct. 10, 2019, 8:48 a.m.
[[[ To any NSA and FBI agents reading my email: please consider    ]]]
[[[ whether defending the US Constitution against all enemies,     ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]

  > The two issues are unrelated.  You are only making things worse for
  > yourself and the community

That is a matter of judgment, and my judgment is different.

In my judgment, it is vital to decide, first, several broader issues
that are bigger and more important.  Each of those issues affects the
right way to decide this question _and other questions_.

I have not been idle about this.  I've brought up several of those
issues, one by one, on this list, and we reached conclusions that are
now stated in the coding standards or maintainer's guide.  But not
quite all of them.  We should wait till the situation is clear and
calm, then take up the remaining issues, one by one, with a month's
pause in between.  Once they are resolved, we can wait another month,
then take up the question of the abort(2) joke and decide it the right
way.

That joke has been in the C Library manual for around 28 years, since
I thoroughly rewrote that manual, and the sky has not fallen.  Let's
not delay needlessly, but we should not be in a rush about the bigger
issues.
Carlos O'Donell - Oct. 10, 2019, 12:56 p.m.
On 10/8/19 5:40 PM, Joseph Myers wrote:
> On Tue, 8 Oct 2019, Florian Weimer wrote:
> 
>> You said repeatedly that you are still in charge in the GNU project, and
>> that GNU is not a democracy.  So this should really be simple.  Just say
>> you want the warning gone, and it will be gone.  Or you tell us you want
>> to keep it.
> 
> And if Richard wants it kept, I consider that just a previous maintainer's 
> view to be considered by the maintainers.  This is not a difficult 
> question about the exact boundaries of when it is appropriate for GNU to 
> intervene in decisions made by individual package maintainers; it's a case 
> for which it's extremely clear-cut that it's properly within the scope of 
> package maintainers to make the final decision and would be improper for 
> GNU to attempt to overrule such a decision.  The precise boundaries of 
> appropriate intervention in individual packages may be worth considering 
> elsewhere, but that's not relevant for this case.

I agree.
Siddhesh Poyarekar - Oct. 10, 2019, 1:13 p.m.
On 10/10/19 4:48 am, Richard Stallman wrote:
> [[[ To any NSA and FBI agents reading my email: please consider    ]]]
> [[[ whether defending the US Constitution against all enemies,     ]]]
> [[[ foreign or domestic, requires you to follow Snowden's example. ]]]
> 
>   > The two issues are unrelated.  You are only making things worse for
>   > yourself and the community
> 
> That is a matter of judgment, and my judgment is different.
> 
> In my judgment, it is vital to decide, first, several broader issues
> that are bigger and more important.  Each of those issues affects the
> right way to decide this question _and other questions_.
> 
> I have not been idle about this.  I've brought up several of those
> issues, one by one, on this list, and we reached conclusions that are
> now stated in the coding standards or maintainer's guide.  But not
> quite all of them.  We should wait till the situation is clear and
> calm, then take up the remaining issues, one by one, with a month's
> pause in between.  Once they are resolved, we can wait another month,
> then take up the question of the abort(2) joke and decide it the right
> way.

I do not agree.  I will always be grateful to you for the idea of
Software Freedom that is one of my core personal values but I have no
faith in your judgment or intention in this regard because you're making
it clearer that it is about assertion of your power and influence in the
project, something I am deeply uncomfortable with and strongly believe
is not in the interest of the glibc project.

Siddhesh
Alexandre Oliva - Oct. 10, 2019, 4:39 p.m.
On Oct  9, 2019, Siddhesh Poyarekar <siddhesh@gotplt.org> wrote:

>> Furthermore, if the software is written so that running it conveys
>> ideas, it's actual free speech, expressed through the software.

> Do you not see that the freedoms specifically talk about software?

I think you missed the bit I quoted above.

Consider writing a server program that includes in responses to requests
it receives a message that you are supportive of.  Then someone modifies
their copy so as to include a message you find abhorrent instead.

The freedom to do so is encompassed by those in the Free Software
Definition, is it not?
Siddhesh Poyarekar - Oct. 10, 2019, 5:03 p.m.
On 10/10/19 12:39 pm, Alexandre Oliva wrote:
> I think you missed the bit I quoted above.
> 
> Consider writing a server program that includes in responses to requests
> it receives a message that you are supportive of.  Then someone modifies
> their copy so as to include a message you find abhorrent instead.
> 
> The freedom to do so is encompassed by those in the Free Software
> Definition, is it not?

Are you kidding me?  THAT is your defense?

Siddhesh
Carlos O'Donell - Oct. 10, 2019, 8:09 p.m.
On 10/10/19 4:48 AM, Richard Stallman wrote:
> [[[ To any NSA and FBI agents reading my email: please consider    ]]]
> [[[ whether defending the US Constitution against all enemies,     ]]]
> [[[ foreign or domestic, requires you to follow Snowden's example. ]]]
> 
>   > The two issues are unrelated.  You are only making things worse for
>   > yourself and the community
> 
> That is a matter of judgment, and my judgment is different.
> 
> In my judgment, it is vital to decide, first, several broader issues
> that are bigger and more important.  Each of those issues affects the
> right way to decide this question _and other questions_.
> 
> I have not been idle about this.  I've brought up several of those
> issues, one by one, on this list, and we reached conclusions that are
> now stated in the coding standards or maintainer's guide.  But not
> quite all of them.  We should wait till the situation is clear and
> calm, then take up the remaining issues, one by one, with a month's
> pause in between.  Once they are resolved, we can wait another month,
> then take up the question of the abort(2) joke and decide it the right
> way.
> 
> That joke has been in the C Library manual for around 28 years, since
> I thoroughly rewrote that manual, and the sky has not fallen.  Let's
> not delay needlessly, but we should not be in a rush about the bigger
> issues.

The decision to remove the joke has been delayed for almost 1.5 years [1].
That we have no concrete conclusion today is again another indication to
me that the current governance structure of the GNU Project is not working.

As a GNU Maintainers for glibc I will no longer hold for further input
on the matter. I agree with Joseph, that your input shall be considered
as another previous maintainer's view on the topic.

We should have a discussion about when it is appropriate for GNU to 
intervene in decisions made by individual package maintainers. I look
forward to having that discussion (hopefully on a public mailing list).
I would hope such a discussion focuses on the GNU Maintainers Guide,
GNU's principles, and the GNU Coding Standard.
Carlos O'Donell - Oct. 10, 2019, 8:41 p.m.
On 10/7/19 1:32 PM, Florian Weimer wrote:
> The warning is confusing to those who do not understand the context,
> and the warning is easy to misunderstand:
> 
> A reader needs to know that it was written by someone who is generally
> skeptical of government influence and control, otherwise it reads as
> an affirmation of the U.S. government's role as the ultimate editor of
> the manual.  This is precisely the opposite of what the warning
> intends to convey.  (Reportedly, it criticizes that several
> U.S. administrations have tried to restrict the medical advice that
> U.S.-funded health care workers can provide abroad, considering that
> censorship.)
> 
> The warning is also misleading on a technical level.  A reader who
> makes the connection to pregnancy termination will get the wrong
> impression that calling the abort function will terminate subprocesses
> of the current process, but this is not what generally happens.
> 
> Finally, for both GNU and the FSF, it is inappropriate to use female
> reproductive health as mere joke material, since these organizations
> do not concern themselves with such issues otherwise, and the warning
> is purportedly about something else entirely.
> 
> This reinstates commit 340d9652b9d0e1d4136588f18b726662d195777c
> ("manual/startup.texi (Aborting a Program): Remove inappropriate
> joke."), effectively reverting the revert in commit
> ffa81c22a3ac0fb75ad9bf2b1c3cdbf9eafa0bc9 ("Revert:").
> 
> 2019-10-07  Florian Weimer  <fweimer@redhat.com>
> 
> 	* manual/startup.texi (Aborting a Program): Remove warning.
 
I have reviewed the full discussion thread.

I see 3 +1's from GNU Maintainers:

* Paul Eggert
* Joseph Myers
* Carlos O'Donell

The following developers have also given a +1:

* DJ Delorie
* Yann Droneaud
* Adhemerval Zanella
* Zack Weinberg
* Samuel Thibault

I note that Richard Stallman (GNU Project lead) has insisted that
there is more to discuss before this issue can be decided. Given
the current wait of ~1.5 years, and no progress on the issue,
I consider it a failing of the GNU Project to give timely input
on the matter. It seems clear to me that we can move forward on
this and discuss other issues separately.

I note that Alexandre Oliva (GNU Maintainer, FSF board member)
has asked we delay the decision. This was a sensible request,
request 1.5 years ago, but it is no longer a sensible request.
The status of the GNU Project and the FSF are issues to be
discussed in other venues, and they impact the glibc project only
in as much as some kind of governance change would impact glibc.

This looks like glibc project consensus to push the change.
Consensus need not imply unanimity.

Florian please feel free to push the change.
Alexandre Oliva - Oct. 11, 2019, 2:26 a.m.
On Oct 10, 2019, Siddhesh Poyarekar <siddhesh@gotplt.org> wrote:

> On 10/10/19 12:39 pm, Alexandre Oliva wrote:
>> I think you missed the bit I quoted above.
>> 
>> Consider writing a server program that includes in responses to requests
>> it receives a message that you are supportive of.  Then someone modifies
>> their copy so as to include a message you find abhorrent instead.
>> 
>> The freedom to do so is encompassed by those in the Free Software
>> Definition, is it not?

> Are you kidding me?  THAT is your defense?

?!?  I think you're just showing how impossible it is to hold a sane
conversation surrounding this topic at this time.

See, you claimed censorship was not in the GNU charter, and that the
stance against censorship was somehow coming from the FSF.

I explained that your assumption was mistaken, that free speech is
actually encompassed by the four software freedoms that the Free
Software Movement, GNU and FSF all promote and defend.

Instead of realizing that the statement only countered your mistaken
attribution of the anti-censorship position exclusively to the FSF, you
appear to mistake that as intended to support my objection to the patch.
It would indeed support an objection to the patch, but that's not the
context in which I presented the argument.

My objection to the patch is based on the understanding that
conversation about this topic at this time is not possible, because of
the already inflamed moods and the inflammatory topics surrounding the
patch.
Richard Stallman - Oct. 11, 2019, 2:53 a.m.
[[[ To any NSA and FBI agents reading my email: please consider    ]]]
[[[ whether defending the US Constitution against all enemies,     ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]

  > The only barely valid argument I have seen in allowing such feature
  > creep from FSF to GNU

That way of conceptualizing the situation surprises me.  It doesn't
fit with the events as they happened.

I suggest reading https://gnu.org/gnu/the-gnu-project.html
about how GNU and the FSF came to be.

			  was that RMS practically owns FSF and GNU and can
  > claim it's one and the same thing.

They are not the same thing.

One pertinent historical point for the C Library Manual is that I am
one of its main authors.  I rewrote most of its text to make it more
readable.  When I came to the abort system call, I added the joke.

I've appointed others to manage work on GNU libc, including this manual,
but I gave specific instructions not to delete this joke.

We can and will consider deleting it, after looking at the remaining
bigger issues.
Siddhesh Poyarekar - Oct. 11, 2019, 3:02 a.m.
On 10/10/19 10:26 pm, Alexandre Oliva wrote:
> ?!?  I think you're just showing how impossible it is to hold a sane
> conversation surrounding this topic at this time.
> 
> See, you claimed censorship was not in the GNU charter, and that the
> stance against censorship was somehow coming from the FSF.
> 
> I explained that your assumption was mistaken, that free speech is
> actually encompassed by the four software freedoms that the Free
> Software Movement, GNU and FSF all promote and defend.
> 
> Instead of realizing that the statement only countered your mistaken
> attribution of the anti-censorship position exclusively to the FSF, you
> appear to mistake that as intended to support my objection to the patch.
> It would indeed support an objection to the patch, but that's not the
> context in which I presented the argument.

No I did not interpret it as a defense of the patch, I understood (and
rejected) the idea that the GNU manifesto and software freedoms
*actually* talks about privacy.

> My objection to the patch is based on the understanding that
> conversation about this topic at this time is not possible, because of
> the already inflamed moods and the inflammatory topics surrounding the
> patch. 

It is no longer about the patch, it probably never was for me because I
don't have the American context of abortion rights.  For me it is about
your manipulative and frustrating arguments, overreach and your
wonderful ability to hold the community hostage and at the same time act
like you're the victim.

I am done engaging with you because there's only so much that a vacation
can heal.

Siddhesh
Siddhesh Poyarekar - Oct. 11, 2019, 3:29 a.m.
On 10/10/19 10:53 pm, Richard Stallman wrote:
> [[[ To any NSA and FBI agents reading my email: please consider    ]]]
> [[[ whether defending the US Constitution against all enemies,     ]]]
> [[[ foreign or domestic, requires you to follow Snowden's example. ]]]
> 
>   > The only barely valid argument I have seen in allowing such feature
>   > creep from FSF to GNU
> 
> That way of conceptualizing the situation surprises me.  It doesn't
> fit with the events as they happened.
> 
> I suggest reading https://gnu.org/gnu/the-gnu-project.html
> about how GNU and the FSF came to be.
>
> 			  was that RMS practically owns FSF and GNU and can
>   > claim it's one and the same thing.
> 
> They are not the same thing.

It isn't to me either, definitely not any more but thank you for making
it explicit.

> One pertinent historical point for the C Library Manual is that I am
> one of its main authors.  I rewrote most of its text to make it more
> readable.  When I came to the abort system call, I added the joke.
> 
> I've appointed others to manage work on GNU libc, including this manual,
> but I gave specific instructions not to delete this joke.
> 
> We can and will consider deleting it, after looking at the remaining
> bigger issues.

You continue to project yourself as a spiritual leader and the
contributor community as your faithful cabal that does your bidding.  I
am grateful to you for starting the GNU project, but I do not contribute
to GNU tools at your appointment and I never did.

Siddhesh
Sandra Loosemore - Oct. 11, 2019, 4:34 a.m.
On 10/10/19 2:41 PM, Carlos O'Donell wrote:
> I have reviewed the full discussion thread.
> 
> I see 3 +1's from GNU Maintainers:
> 
> * Paul Eggert
> * Joseph Myers
> * Carlos O'Donell
> 
> The following developers have also given a +1:
> 
> * DJ Delorie
> * Yann Droneaud
> * Adhemerval Zanella
> * Zack Weinberg
> * Samuel Thibault
> 
> I note that Richard Stallman (GNU Project lead) has insisted that
> there is more to discuss before this issue can be decided. Given
> the current wait of ~1.5 years, and no progress on the issue,
> I consider it a failing of the GNU Project to give timely input
> on the matter. It seems clear to me that we can move forward on
> this and discuss other issues separately.
> 
> I note that Alexandre Oliva (GNU Maintainer, FSF board member)
> has asked we delay the decision. This was a sensible request,
> request 1.5 years ago, but it is no longer a sensible request.
> The status of the GNU Project and the FSF are issues to be
> discussed in other venues, and they impact the glibc project only
> in as much as some kind of governance change would impact glibc.
> 
> This looks like glibc project consensus to push the change.
> Consensus need not imply unanimity.
> 
> Florian please feel free to push the change.

Forgive me for coming in late, but I've only just noticed this discussion.

My name is on the manual, and I support removing the "joke".  It's not 
funny, and it doesn't provide any useful technical information.

-Sandra
Samuel Thibault - Oct. 11, 2019, 5:59 a.m.
Alexandre Oliva, le jeu. 10 oct. 2019 23:26:12 -0300, a ecrit:
> On Oct 10, 2019, Siddhesh Poyarekar <siddhesh@gotplt.org> wrote:
> 
> > On 10/10/19 12:39 pm, Alexandre Oliva wrote:
> >> I think you missed the bit I quoted above.
> >> 
> >> Consider writing a server program that includes in responses to requests
> >> it receives a message that you are supportive of.  Then someone modifies
> >> their copy so as to include a message you find abhorrent instead.
> >> 
> >> The freedom to do so is encompassed by those in the Free Software
> >> Definition, is it not?
> 
[...]
> I explained that your assumption was mistaken, that free speech is
> actually encompassed by the four software freedoms that the Free
> Software Movement, GNU and FSF all promote and defend.

However with such reasoning, not only free speech is encompassed, but
basically anything in the world. You write some software to manage
selling bubblegum, somebody can modify it to manage selling anything,
including whatever you consider abhorrent. With such reasoning you can
claim that free software is about anything. Which is exactly what we
want in the end, indeed! But actually backing up anything from this is
meanless.

Samuel
Christian Brauner - Oct. 11, 2019, 5:48 p.m.
On Thu, Oct 10, 2019 at 04:41:58PM -0400, Carlos O'Donell wrote:
> On 10/7/19 1:32 PM, Florian Weimer wrote:
> > The warning is confusing to those who do not understand the context,
> > and the warning is easy to misunderstand:
> > 
> > A reader needs to know that it was written by someone who is generally
> > skeptical of government influence and control, otherwise it reads as
> > an affirmation of the U.S. government's role as the ultimate editor of
> > the manual.  This is precisely the opposite of what the warning
> > intends to convey.  (Reportedly, it criticizes that several
> > U.S. administrations have tried to restrict the medical advice that
> > U.S.-funded health care workers can provide abroad, considering that
> > censorship.)
> > 
> > The warning is also misleading on a technical level.  A reader who
> > makes the connection to pregnancy termination will get the wrong
> > impression that calling the abort function will terminate subprocesses
> > of the current process, but this is not what generally happens.
> > 
> > Finally, for both GNU and the FSF, it is inappropriate to use female
> > reproductive health as mere joke material, since these organizations
> > do not concern themselves with such issues otherwise, and the warning
> > is purportedly about something else entirely.
> > 
> > This reinstates commit 340d9652b9d0e1d4136588f18b726662d195777c
> > ("manual/startup.texi (Aborting a Program): Remove inappropriate
> > joke."), effectively reverting the revert in commit
> > ffa81c22a3ac0fb75ad9bf2b1c3cdbf9eafa0bc9 ("Revert:").
> > 
> > 2019-10-07  Florian Weimer  <fweimer@redhat.com>
> > 
> > 	* manual/startup.texi (Aborting a Program): Remove warning.
>  
> I have reviewed the full discussion thread.
> 
> I see 3 +1's from GNU Maintainers:
> 
> * Paul Eggert
> * Joseph Myers
> * Carlos O'Donell
> 
> The following developers have also given a +1:
> 
> * DJ Delorie
> * Yann Droneaud
> * Adhemerval Zanella
> * Zack Weinberg
> * Samuel Thibault

I'm in favor or removing the joke. I quote myself from May 2018 (cf. [1]):
"After having read the LVM article and following the thread here I feel
the need to voice my opinion: I'm in favor of removing this joke and
reverting the revert.
The comments on the LWN article and the wider discussion seem to
indicate that some question whether there is sufficient consensus among
maintainers. Even though I'm neither a senior developer nor steward, but
a simple maintainer I feel it's important to explicitly voice my opinion
to help build consensus."

[1]: https://sourceware.org/ml/libc-alpha/2018-05/msg00245.html

> 
> I note that Richard Stallman (GNU Project lead) has insisted that
> there is more to discuss before this issue can be decided. Given
> the current wait of ~1.5 years, and no progress on the issue,
> I consider it a failing of the GNU Project to give timely input
> on the matter. It seems clear to me that we can move forward on
> this and discuss other issues separately.
> 
> I note that Alexandre Oliva (GNU Maintainer, FSF board member)
> has asked we delay the decision. This was a sensible request,
> request 1.5 years ago, but it is no longer a sensible request.
> The status of the GNU Project and the FSF are issues to be
> discussed in other venues, and they impact the glibc project only
> in as much as some kind of governance change would impact glibc.
> 
> This looks like glibc project consensus to push the change.
> Consensus need not imply unanimity.
> 
> Florian please feel free to push the change.
> 
> -- 
> Cheers,
> Carlos.
Carlos O'Donell - Oct. 11, 2019, 6:20 p.m.
On 10/11/19 12:34 AM, Sandra Loosemore wrote:
> On 10/10/19 2:41 PM, Carlos O'Donell wrote:
>> I have reviewed the full discussion thread.
>>
>> I see 3 +1's from GNU Maintainers:
>>
>> * Paul Eggert
>> * Joseph Myers
>> * Carlos O'Donell
>>
>> The following developers have also given a +1:
>>
>> * DJ Delorie
>> * Yann Droneaud
>> * Adhemerval Zanella
>> * Zack Weinberg
>> * Samuel Thibault
>>
>> I note that Richard Stallman (GNU Project lead) has insisted that
>> there is more to discuss before this issue can be decided. Given
>> the current wait of ~1.5 years, and no progress on the issue,
>> I consider it a failing of the GNU Project to give timely input
>> on the matter. It seems clear to me that we can move forward on
>> this and discuss other issues separately.
>>
>> I note that Alexandre Oliva (GNU Maintainer, FSF board member)
>> has asked we delay the decision. This was a sensible request,
>> request 1.5 years ago, but it is no longer a sensible request.
>> The status of the GNU Project and the FSF are issues to be
>> discussed in other venues, and they impact the glibc project only
>> in as much as some kind of governance change would impact glibc.
>>
>> This looks like glibc project consensus to push the change.
>> Consensus need not imply unanimity.
>>
>> Florian please feel free to push the change.
> 
> Forgive me for coming in late, but I've only just noticed this
> discussion.
> 
> My name is on the manual, and I support removing the "joke".  It's
> not funny, and it doesn't provide any useful technical information.

Thank you for your review.
Alexandre Oliva - Oct. 12, 2019, 12:24 a.m.
On Oct 11, 2019, Samuel Thibault <samuel.thibault@gnu.org> wrote:

> With such reasoning you can claim that free software is about
> anything. Which is exactly what we want in the end, indeed!

Yes, exactly!  That's what freedom is about, and that's what the Free
Software Movement stands for.  If it was about imposing such constraints
as censorship, or prohibitions of certain activities, then it wouldn't
be the case that one could change it to express any idea whatsoever, or
to serve any function one could conceive of.
Samuel Thibault - Oct. 12, 2019, 1:24 a.m.
Alexandre Oliva, le ven. 11 oct. 2019 21:24:36 -0300, a ecrit:
> On Oct 11, 2019, Samuel Thibault <samuel.thibault@gnu.org> wrote:
> 
> > With such reasoning you can claim that free software is about
> > anything. Which is exactly what we want in the end, indeed!
> 
> Yes, exactly!

You dropped the rest.

> then it wouldn't be the case that one could change it to express any
> idea whatsoever, or to serve any function one could conceive of.

Which doesn't mean it backs free speech more than anything else. Again,
with your reasoning you could back *anything*, so it is a meaningless
reasoning.

Put in other words, "\forall A, X => A" means that X is meaningless
since anything can be deducted from it.

Samuel
Alexandre Oliva - Oct. 12, 2019, 4:54 p.m.
On Oct 11, 2019, Samuel Thibault <samuel.thibault@gnu.org> wrote:

> Alexandre Oliva, le ven. 11 oct. 2019 21:24:36 -0300, a ecrit:
>> then it wouldn't be the case that one could change it to express any
>> idea whatsoever, or to serve any function one could conceive of.

> Which doesn't mean it backs free speech more than anything else

No, no, the argument is precisely that *because* it appears to back
anything the way you pointed out, it can be concluded that it actually
stands for freedom, free speech specifically.

Conversely, if it constrained speech in any way, that would be
noticeable in that you could point out what it is that it excluded.

You'll notice that the four freedoms even enable software to be used to
build censorship systems.  That doesn't mean the four freedoms are for
censorship, it just shows that respecting others' freedom sometimes
involves respect for things that you may find very objectionable.  Free
Software does encompass that notion.  It is quite disturbing and
worrying to find out that some Free Software supporters do not realize
that it does, or even reject that it does :-/
Samuel Thibault - Oct. 12, 2019, 5:32 p.m.
Alexandre Oliva, le sam. 12 oct. 2019 13:54:17 -0300, a ecrit:
> On Oct 11, 2019, Samuel Thibault <samuel.thibault@gnu.org> wrote:
> > Alexandre Oliva, le ven. 11 oct. 2019 21:24:36 -0300, a ecrit:
> >> then it wouldn't be the case that one could change it to express any
> >> idea whatsoever, or to serve any function one could conceive of.
> 
> > Which doesn't mean it backs free speech more than anything else
> 
> No, no, the argument is precisely that *because* it appears to back
> anything the way you pointed out, it can be concluded that it actually
> stands for freedom, free speech specifically.

But my point is that it's not *specifically* about free speech.  Going
the way you suggested, anybody could argue anything and introduce any
kind of not-so-related things in the libc manual, and thus actually
*blurrying* the point of free software.

> You'll notice that the four freedoms even enable software to be used to
> build censorship systems.  That doesn't mean the four freedoms are for
> censorship,

But your reasoning *can* go that way. If you didn't mean your reasoning
to go that way, you'll have to understand that it is a meaningless
reasoning.

> It is quite disturbing and worrying to find out that some Free
> Software supporters do not realize that it does, or even reject that
> it does :-/

I am not rejecting that it does.  What I mean is that you can not claim
that free software is about freedom of speech in particular, because
that's just one of anything that free software allows.

Samuel
Richard Stallman - Oct. 14, 2019, 1:43 a.m.
[[[ To any NSA and FBI agents reading my email: please consider    ]]]
[[[ whether defending the US Constitution against all enemies,     ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]

									  I
  > am grateful to you for starting the GNU project, but I do not contribute
  > to GNU tools at your appointment and I never did.

You were appointed indirectly by the package maintainers that I
appointed.

Every GNU package has maintainers appointed by the GNU Project.  They
are in charge of the work on that package and responsible to the GNU
Project.  The maintainers of GNU libc are the people on the
steering committee.

The maintainers of a package can do the work alone -- that is the
usual case -- but with the maintainers of some packages delegate some
of the work to other developers that they appoint.  The package
maintainers can delegate responsibility but they cannot cede it.

Apparently an unfortunate miscommunication happened in your case.
Carlos O'Donell - Oct. 14, 2019, 1:30 p.m.
On 10/13/19 9:43 PM, Richard Stallman wrote:
> [[[ To any NSA and FBI agents reading my email: please consider    ]]]
> [[[ whether defending the US Constitution against all enemies,     ]]]
> [[[ foreign or domestic, requires you to follow Snowden's example. ]]]
> 
> 									  I
>   > am grateful to you for starting the GNU project, but I do not contribute
>   > to GNU tools at your appointment and I never did.
> 
> You were appointed indirectly by the package maintainers that I
> appointed.

He is still correct, you did not appoint him.

> Apparently an unfortunate miscommunication happened in your case.
 
I don't think there was any miscommunication. Would you care to expand
on exactly what was miscommunicated?
Richard Stallman - Oct. 18, 2019, 2:13 p.m.
[[[ To any NSA and FBI agents reading my email: please consider    ]]]
[[[ whether defending the US Constitution against all enemies,     ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]

  > >   > am grateful to you for starting the GNU project, but I do not contribute
  > >   > to GNU tools at your appointment and I never did.

  > > You were appointed indirectly by the package maintainers that I
  > > appointed.

  > He is still correct, you did not appoint him.

Did I say his statement was literally incorrect?  No.
My point was that it doesn't mean he is independent.

I did not appoint him directly.  I directly appoint only
package maintainers, who are then responsible directly to me.

Rather, the package maintainers of GNU libc recruited him to do
some of the work.  He is supposed to be responsible to them.

  > > Apparently an unfortunate miscommunication happened in your case.
 
  > I don't think there was any miscommunication. Would you care to expand
  > on exactly what was miscommunicated?

Apparently it was not explained to him that he was responsible to
the package maintainers who were responsible to me.
Siddhesh Poyarekar - Oct. 18, 2019, 2:30 p.m.
On 18/10/19 10:13 am, Richard Stallman wrote:
>   > He is still correct, you did not appoint him.
> 
> Did I say his statement was literally incorrect?  No.
> My point was that it doesn't mean he is independent.
> 
> I did not appoint him directly.  I directly appoint only
> package maintainers, who are then responsible directly to me.
> 
> Rather, the package maintainers of GNU libc recruited him to do
> some of the work.  He is supposed to be responsible to them.
> 
>   > > Apparently an unfortunate miscommunication happened in your case.
>  
>   > I don't think there was any miscommunication. Would you care to expand
>   > on exactly what was miscommunicated?
> 
> Apparently it was not explained to him that he was responsible to
> the package maintainers who were responsible to me.

You missed the tone and context of my statement; it has less to do with
my understanding of the GNU project hierarchy* and more to do with how I
see my contribution to the project.  My response was an attempt to bring
forward the fact that a number of us in the GNU project continue
contributing to the project not because of your involvement but despite it.

Siddhesh

* I've been contributing to GNU projects for a little less than a decade
so presuming that I don't understand is frankly a bit patronizing.

Patch

diff --git a/manual/startup.texi b/manual/startup.texi
index 7395d32dd0..21c48cd037 100644
--- a/manual/startup.texi
+++ b/manual/startup.texi
@@ -1005,14 +1005,6 @@  This function actually terminates the process by raising a
 intercept this signal; see @ref{Signal Handling}.
 @end deftypefun
 
-@c Put in by rms.  Don't remove.
-@cartouche
-@strong{Future Change Warning:} Proposed Federal censorship regulations
-may prohibit us from giving you information about the possibility of
-calling this function.  We would be required to say that this is not an
-acceptable way of terminating a program.
-@end cartouche
-
 @node Termination Internals
 @subsection Termination Internals