[Fsfe-uk] Re: the GNU/Linux FAQ, was: BBC's DRM Iplayer windows only

[Yavor Doganov]

В Thu, 17 Jan 2008 15:40:43 +0000, MJ Ray написа:

> Yavor Doganov <address@hidden> wrote:
>> The GNU/Linux FAQ does not attempt to answer all questions 
> 
> That is another reason why it is inadequate.  No conversation like:
> 
>   Newbie: Can you summarise why we should call it GNU/Linux? 
[...] 
> will convince many people.

Sure.  A free software activist is doomed if she relies on simply
pointing to URL's as answers.  The essays are supplementary materials,
and cannot replace a speech, interview, discussion or a QA session.
If it was so, RMS would just sit in his lair and only committing
improvements of the articles.

Anyway, to this question of Newbie I'd explain in my own words and
then point to why-gnu-linux -- IMHO this is the most important
article, which also answers the first question from your "missing
questions" triplet.

> But the point I was trying to make before was that it both answers
> many questions which are not very frequently asked *and* ignores
> some that I suspect are among the most frequently asked - because
> they are hard, perhaps?

I am not aware of any question on this topic that is hard to answer.
Some require more thought and investigation, for example whether
Minix3 or Syllable are variants of the GNU system. 

> I often look at company-produced FAQs [...]  I'd much rather people
> didn't have the same reaction to this FAQ.

Yeah, the feeling is familiar, but I never had that feeling for the
GNU/Linux FAQ.

> Not really, as "we suggest that you avoid acronyms that contain
> 'lug'" actually seems to mean "you must..." in most cases and things
> like that.

Read below "A note about GNU/Linux user groups".  It is quite clear
that we will reject a GNU/Linux group that uses the proper name, but
has articles that "help" users to install non-free software (or
advertisements).

>> > and "How have groups been persuaded to call it GNU/Linux?"
>>
>> There is no single answer to this question.
> 
> Then there are answers to it!!!  [...] Give a few examples to
> inspire if poss, rather than just ducking it.

This is a probably a good idea.
 
>> > Some of the questions there that actually are frequently-asked
>> > ones, such as the X11-Apache-Perl one, have weak answers and
>> > don't give references to support the claims.
>>
>> Why do you think it's a weak answer?  IMO it's very persuasive.
> 
> It doesn't give references to support the claims.  See Dave
> Crossland's (IIRC) message earlier in the thread which linked a
> study by David Wheeler(?).

What references?  A study that shows how important components are
libc, binutils, the GNU build system, etc?  Or that the GNU project
has been involved in writing and assembling together the essential
components since day 1?  A reference like David Wheeler's is nothing
more than a his personal opinion and I'm not sure if it would help.
Even Wikipedia is not a reliable resource that will convince
everybody.

> Like the FAQ, it also fails to use available references.

What exactly from what RMS says has to be backed up with proof?  These
are well known facts, even to those from the other "camp".

> In the absence of an "OK, you're right" message, I think it's
> dangerous to infer much from that.

OK, I'll assume you're right.

>> This is because of the way this article has evolved.  You are wrong
>> that this is an artificial compilation;
> 
> OK, so can someone give the numbers then?

Not me.  You claimed that this FAQ was just made up for "marketing"
reasons, which is not true.  I don't think that it's worth spending my
time in digging just to figure out which question was asked when and how.
Maybe, if a sufficiently energetic person with archaeological
interests steps up, this is doable.  Many things are in the memories
of RMS and other people, though.

> That means they are "asked questions" but not necessarily "frequently
> asked questions".  See the difference?  *frequently*

Yes, I see the difference.  I am not aware of any frequently asked
question that is omitted (deliberately or not).  Perhaps some of the
questions are not frequently asked, but RMS has decided to include
them because they are important.

> most of them say things like "(always):
> New answer." with no suggestion where they came from.

Some people write verbose logs, others don't.  RMS usually follows the
GNU standards so this log entry describes the change but not the
reasons for it (the reasons for adding a new answer are obvious
anyway).  What would be the benefit if the log was:

        (always): New answer to a question by Misausuka Madoto asked
        at Osaka during my speech on Oct 17, 2005.

> There are also many cross-references to rt.gnu.org which I think is
> private

Yes, and rightly so.  Many people do not want their personality
disclosed and the correspondence need not be public.

> The GNU project could make it easier by providing better resources and
> more support to persuaders,

Yes, I agree.  Basically, like other free software projects, the
people in the GNU project do the best they can, and often they fail.  If 
you think there are hordes of people involved in maintaining the articles 
or doing tedious infrastructure work, you are wrong.

> but denying the weaknesses of materials like the GNU/Linux FAQ,

I didn't deny anything of that kind.  There is most definitely room
for improvement, likewise for all articles.

> suggests that ain't gonna happen any time soon.

You realize that a report that a specific article is vague, poorly
structured and doesn't serve its purpose is not very helpful when the
report itself is vague?  Please give concrete examples and
suggestions, like:

1) Questions "Foo" and "Bar" are frequently asked, but missing.
2) The answers should be in {chronological,thematical,importance}
   order, because ...
3) Answer "Baz" is poor, because...  (A beter one would be ...)

> The GNU/Linxu FAQ is just a personal opinion of the leader of the GNU
> project and not a practical tool?

I did not say this.  This article is a statement of the GNU project
and you and I cannot make arbitrary changes.  What we consider a
"bugfix" may be "introducing a new bug".

> I think that summarises many of the problems with both the FAQ and
> the GNU marketing effort.

There is no "GNU marketing effort", this is an educational campaign.

> If we couldn't defend this view from "Why we should call the OS Linux"
> and if it really is easy to change that FAQ into that

"Hyberbole" and "metaphor" are common methods in literature and casual
speech, so any attempt to read them literally leads to
misunderstanding, at best.  Rephrasing myself for those minds used to
mechanical parsing:

"If the license of the article allowed arbitrary modifications, it
would be possible, in theory and in practice, that someone makes a
change which does not correspond to the views of the GNU project.
Republishing such modified version would be confusing and
counter-productive, because one of the main purposes of these articles
is to raise awareness and *avoid* confusion."

The society does not suffer from these distribution terms, because
none of the basic freedoms is threatened and anyone can propose
improvement.  Also, anyone can quote and rebut statements/views
expressed in gnu.org essays without the obvious disadvantage of
misrepresenting them.

> It's a shame the GNU project is less confident than me, isn't it?

I wonder how nearly 24 years of constant, stubborn, consistent and
non-retreating way of managing the battle can possibly lead to such a
conclusion.