GCC association with the FSF

[David Brown]

On 08/04/2021 19:22, Giacomo Tesio wrote:
> No, David, 
> 
> On April 8, 2021 3:00:57 PM UTC, David Brown <david@westcontrol.com> wrote:
> 
>>  (And yes, I mean FOSS here, not just free software.)
> 
> you are not talking about Free Software, but Open Source.
> 
> FOSS, as a term, has been very successful to spread confusion.
> 

You have snipped the context.  Let me repeat it:

"""
... no one can
be in doubt that [RMS's] attitudes and behaviour are not acceptable by
modern standards and are discouraging to developers and users in the
FOSS community.  (And yes, I mean FOSS here, not just free software.)
"""

For most people that have enough interest in software to be aware of the
concepts of free and/or open source software, lump them together.  That
applies to users and developers.  To the majority of gcc users, they do
not care whether the project refers to itself as "free software" or
"open source software".  They often care that it is easily available at
zero cost (though some pay for it - I and my company have, at times,
bought gcc packages), and they like the fact that all the source code is
available even if they don't look at the source themselves.

But whoever you blame for spreading confusion, or for artificially
creating distinctions that rarely matter (this viewpoint has its
supporters too), the fact remains that the mix-up is real.  In almost
all circumstances, to almost all people, it is all "FOSS".  And the GNU
project, along with Linux, LibreOffice (or still OpenOffice, in most
people's minds), Firefox, and a few other big projects are viewed
together as a group and the opposite of "big company" software such as
MS Windows and Office, Apple software, and Adobe Photoshop (to take some
well-known examples).  The attitudes of GNU leaders have an influence on
all of this, as do other public leader figures such as Linus Torvalds.
Their influence (for good or bad) extends well outside the direct
hierarchy of their official positions within their projects.

> 
>> his attitudes and behaviour are not acceptable by
>> modern standards and are discouraging to developers and users in the
>> FOSS community.
> 
> In fact, I'm actively looking for alternatives to GCC (and LLVM) because I cannot trust a 
> GCC anymore and I cannot review each and every change.
> 

That is your choice, obviously.  I don't agree with your points
expressed in this list so far, but you make your own decisions here.
Call me naïve, but I trust the maintainers of gcc to make good technical
decisions and make changes that improve the compiler suite.

I do think it is entirely possible that - for example - Facebook will
pay an employee to add features to gcc with the specific aim of
improving the efficiency of the code Facebook uses.  I think that would
be entirely reasonable, and I would be quite happy with it - either the
changes will coincidentally improve that is useful to me, or it will do
it no harm.  I think it is /implausible/ that any company would exert an
influence over gcc in order to make it worse for competitors or other
users.  This is an open source project (in addition to being free
software) - it is hard to make hidden changes when all changes are
reviewed and visible to many people.  I don't believe in conspiracy
theories - they require the cooperation of too many people who would
disagree and make a noise.

(Mistakes happen, and attacks from outside occasionally happen in open
source projects, but that's another matter.)

> I won't contribute my port and in general will suggest people to look for alternatives.
> 
> 
> But that's not a problem for you, because you do not actually care about real developers 
> and users, just about the US corporations you effectively mentioned and now control 
> several GNU projects:

No, I have no particular interest in any companies (other than loyalty
to my own company).  I am not an American, nor do I live in America - I
am Scottish and live in Norway.  Not that that matters here.

And yes, I care about the gcc developers and their ability and freedom
to work as they want on the project.  I care about potential new
developers too - and I do not want to see them reject the idea of
working for gcc (or any other project) because they perceive a foul
atmosphere of bullying, sexual harassment or misogyny.  Nor would I want
anyone to avoid contributing to gcc because of perceived bias for or
against any particular country, culture, religion, or any other aspect
of life that has no relevance for code development.

And yes, I care about users - I am one, having used gcc for some 25
years on perhaps a dozen different targets.

I don't think any corporations control any GNU projects (with which I am
familiar) in the sense of deciding what goes into them, who works on
them, what direction they should take, or anything of that sort.  But a
big development project takes resources - it costs a lot of money.  This
usually comes from corporations that have an interest in the project's
success - companies that are big users, or re-sell the software.  This
can take effect in different ways - for gcc, this is done by employing
people and letting them work on gcc.  (As far as I understand it, the
individuals usually make their own choices as to what they will work
on.)  Alternative arrangements include for companies to pay sponsorship
and that money is used to employee developers directly in the project -
that model is certainly used on some projects.  However it is done, if
there is a dramatic situation and a major resource supply stops
suddenly, it is very bad for the project.

> 
>> someone in the public relations
>> department at IBM, Google, Facebook, ARM, or other big supporters of
>> the project will get the impression ...
> 
> As you explained, GCC itself is completelly  controlled by few US corporations with 
> strong and long term ties with the US DoD.

I did not "explain" that at all.  You are free to belief this if you
want (you have already said as much), but please do not imagine for an
instant that I have agreed with or supported that view.

I also do not see in any sense how you can conclude that these companies
have "strong and long term ties" with the US DoD.  They are not weapons
suppliers or military suppliers.  The US DoD, just as every country's
DoD, buys whatever computers, chips, software and services it feels make
sense for its needs.  And these companies sell to whatever customers
want their goods and services, subject to any national or internation
trade restrictions.

I am at a loss to understand how such conspiracy theories are relevant
in the discussion about whether or not it would be best for the gcc
project to be independent of the FSF and GNU.

> 
> For sure, it's a big software. And a big threat to everybody outside the US.
> 

You do realise that Italy is a member of NATO - one of the founding
members?  Like Norway, you are part of a military alliance with the USA.
 If you view the US DoD as a threat, you must also view your own
country's DoD as a threat.

But again, you are connecting dots to form a picture that does not
exist, and assigning undue relevance to thin and indirect relationships.
 David Malcolm (to pick one gcc developer) works for Red Hat, which is
owned by IBM, which has supplied computers to the US DoD.  By no
realistic stretch of the imagination does that mean David's
contributions to gcc are controlled by the US DoD.

> 
> Thanks for coming out.
> 

I am truly sorry that you have got such a twisted impression and
misunderstanding of what I wrote, and apparently also what others here
have written.  I hope you can learn to trust other people more and take
their word at face value, instead of seeing conspiracies and hidden
meanings behind everything.

David


> 
> Giacomo
>