Re: GNU project _does_ discriminate contributors by classes (was: A GNU “social contract”?)

[Alfred M. Szmidt]

   To the best of my knowledge, thatʼs completely untrue: major GNU
   subprojects do discriminate contributors by classes: if a
   contributor-to-be happens to be an employee, FSF does not trust his
   words about origin of his contribution, he has to bring a reference
   from his employer.

The reason for the "reference" (I think you mean disclaimer?) is a
legal one and protection for the FSF, it isn't about trust.

Re: GNU project _does_ discriminate contributors by classes

[Dmitry Alexandrov]

Florian Weimer  wrote:
> Widely misunderstood terms can create quite a mess, as can be seen here:
>
>   
>
> Basically, it turns out that several Bacula contributors signed away their 
> rights twice (“Due to historical reasons Beneficiaries have sometimes signed 
> a second FLA afterhaving signed the first one with either FSFE or Kern 
> Sibbald.”), which is of course not legally possible.

Interesting!  Thank you.

It worth clarifying, though, that the actual link where it can be seen is not 
the above but .


signature.asc
Description: PGP signature

Re: GNU project _does_ discriminate contributors by classes

[Dmitry Alexandrov]

Mike Gerwitz  wrote:
> On Tue, Oct 29, 2019 at 17:43:05 +0300, Dmitry Alexandrov wrote:
>> Jean Louis  wrote:
>>> GNU project does not discriminate by gender, or other classes, neither 
>>> verifies genders of contributors, or their classes, as everybody is welcome 
>>> to contribute
>>
>> To the best of my knowledge, thatʼs completely untrue: major GNU subprojects 
>> do discriminate contributors by classes: if a contributor-to-be happens to 
>> be an employee, FSF does not trust his words about origin of his 
>> contribution, he has to bring a reference from his employer.
>>
>> Speaking frankly, even if we put aside how time-consuming it may be, itʼs 
>> hard for me to imagine what can be more degrading and thus ‘alienating’ to 
>> someone, than a straightaway demand to prove that he is not a liar.
>
> This is a legal issue and has nothing to do with discrimination.

Why do you say that as if discrimination cannot be a legal issue?  (In fact, 
sexual discrimination is sometimes a legal issue, discrimination of minors is 
almost always is a legal issue, discrimination of foreign citizens is a purely 
legal issue.)

In any case, by all means it has to do:

| discrimination (countable and uncountable, plural discriminations)
|
| 3. (sometimes discrimination against) Distinct treatment of an individual or 
group to their disadvantage; treatment or consideration based on class or 
category rather than individual merit
— https://en.wiktionary.org/wiki/discrimination#English

> I had to have my employer sign one of those waivers.  The purpose is to put 
> my contributions on solid legal ground.

No need to convince me that it has a good purpose.  I never supposed, that it 
is groundless, only that:

 a) being magnified in public mind, it contribute more to reputation of GNU as 
an unwelcoming place than vague “not feel at ease” stuff; and

 b) claim that “GNU welcomes contributions from all and everyone” while this 
issue is unresolved might be perceived as hypocrisy and make things even worse.


signature.asc
Description: PGP signature

Re: GNU project _does_ discriminate contributors by classes

[Dmitry Alexandrov]

Jason Self  wrote:
> My direct firsthand experience directly conflicts with what you allege to be 
> the case.

Nice to hear this!

However, it shows how inadequate the situation is: one have to collect 
firsthand experience rather than read clear and concise summary on the topic, 
published officially.

And what is published officially sometimes only makes things worse.  E. g., the 
maintainerʼs handbook, you linked above, seems not been updated to reflect, 
that FSF abandoned the requirement of snail mail exchange worldwide:

| Contributors residing outside the USA, Germany or India must mail the signed 
form to the FSF via postal mail.
— https://www.gnu.org/prep/maintain/html_node/Copyright-Papers.html

I also never found any official reference, that a contributor is entitled to 
get back from FSF an all-permissive licence on what he assigned — that 
radically changes the perception of the deal, but again it is backed only with 
anecdotal evidence.

Et cetera, et cetera.  Even the fact, that not every GNU package is owned by 
FSF, is not the common knowledge.

Some argue, that GNU gained an image of unwelcoming place because of some 
jokes.  What I observe in free and ‘open source’ software communities, though, 
is that GNU gained an image an unwelcoming place due to its bureaucratic 
practices — whether they are real or perceived.


> P.S.; there's no need to address the message to me directly - I am on the 
> list.

Excuse me, but such requests always abash me.

Even if we put aside, that (a) the practice of addressing the general public 
while actually taking to a specific person is harmful for readability and 
searchability of MLs in general, and (b) itʼs vital for unreliable premoderated 
lists (and as a bonus, it was exactly what enabled you to reply even before my 
message passed the censor); what exactly are you asking me to do with that 
piece of information?  To keep it in mind? :-\

If one has troubles in configuring his mail server / useragent and, despite all 
of the above, want to shift the burden of satisfying his preferences to _every_ 
his correspondent, there are formal ways to do that, the headers: a simple and 
static ‘Mail-Copies-To: never’, and no so simple ‘Mail-Followup-To’, which is 
no point to explain here, either his MUA supports it for the case or does not.

By the way, you might notice now, that all my mail have ‘Mail-Copies-To: 
always’ set, which supposed to instruct MUAs _not_ to tamper with the list of 
recipients in the way that excludes me from it.  In rarely helps, though: 
those, who use good MUAs, that respect it, rarely come up with an idea to drop 
an actual correspondent from ‘To’.


signature.asc
Description: PGP signature

Re: GNU project _does_ discriminate contributors by classes

[Mike Gerwitz]

On Tue, Oct 29, 2019 at 17:43:05 +0300, Dmitry Alexandrov wrote:
> Jean Louis  wrote:
>> GNU project does not discriminate by gender, or other classes, neither
>> verifies genders of contributors, or their classes, as everybody is
>> welcome to contribute
>
> To the best of my knowledge, thatʼs completely untrue: major GNU subprojects
> do discriminate contributors by classes: if a contributor-to-be happens to
> be an employee, FSF does not trust his words about origin of his
> contribution, he has to bring a reference from his employer.
>
> Speaking frankly, even if we put aside how time-consuming it may be, itʼs
> hard for me to imagine what can be more degrading and thus ‘alienating’ to
> someone, than a straightaway demand to prove that he is not a liar.

This is a legal issue and has nothing to do with discrimination.

I had to have my employer sign one of those waivers.  The purpose is to
put my contributions on solid legal ground.  _I_ can claim that I have
the legal right to make those contributions, but what happens when the
FSF goes to enforce copyright and my employer goes to argue that they
have no authority to do so because I never had the right to assign
copyright to the FSF for my contributions to begin with?

But since my employer has signed a waiver, there's no question.  And I'm
thankful for that, because it also removes any concern from my
mind.  (In fact, in my case, the FSF's legal team and my employer had to
negotiate terms under which my contributions are assignable.)

As Jason Self mentioned, this isn't always necessary.  In my case it was
necessary because some of the code I was assigning was written during
work hours.

-- 
Mike Gerwitz
Free Software Hacker+Activist | GNU Maintainer & Volunteer
GPG: D6E9 B930 028A 6C38 F43B  2388 FEF6 3574 5E6F 6D05
https://mikegerwitz.com


signature.asc
Description: PGP signature

Re: GNU project _does_ discriminate contributors by classes

[Florian Weimer]

* Dmitry Alexandrov:

> Jason Self  wrote:
>> This is related to the copyright assignment mentioned earlier. As an
>> example, some have employment contracts with wording to the effect
>> that "anything you ever do anywhere ever always belongs to us no
>> matter what." People with such employment contracts aren't able to
>> assign the copyright for their work because it was never theirs to
>> begin with and the employer needs to do so. The information at [0]
>> mentions this that "we ***may*** also need an employer’s
>> disclaimer..." because...
>
> ...FSF does not believe in GNU contributorsʼ honesty?

You could also say that these procedures are inplace to protect the
contributors.  Widely misunderstood terms can create quite a mess, as
can be seen here:

  

Basically, it turns out that several Bacula contributors signed away
their rights twice (“Due to historical reasons Beneficiaries have
sometimes signed a second FLA afterhaving signed the first one with
either FSFE or Kern Sibbald.”), which is of course not legally
possible.

Collecting additional paperwork may help to prevent that situation
from arising.

Re: GNU project _does_ discriminate contributors by classes

[Dmitry Alexandrov]

DJ Delorie  wrote:
>> if a contributor-to-be happens to be an employee, FSF does not trust his 
>> words about origin of his contribution,
>
> This seems reasonable to me in the USA.
> Given how complex employment contracts are, it's reasonable to ask for a 
> legal disclaimer from employers
> It's not about trusting the people involved...

...rather about having doubts whether they are intelligent enough to understand 
their own employment contract.  Much a relief! :-D

But thatʼs actually not the point, I never suggested that it is not reasonable. 
 I am not able to judge.  On the contrary, as I already said [1]:

| And that discrimination may be a well-justified choice — justified by the 
goal of GNU project: to develop a free operating system.  After all, nothing in 
it implies that the development should be welcoming for everyone.  Many, 
including probably the most widespread free program in the world — SQLite, do 
not welcome contributors at all.

What made me remind about it now, was not (only) the practice by itself, but 
(also) how it looked in context of the topic — a proposed ‘GNU social contract’:

| However, now @l...@gnu.org is proposing fundamental change: to make ‘welcomes 
contributions’ a principle in itself along with ‘respects freedom’.  And it 
also may turn out to be a very good decision: bazaar development indeed proved 
to be quite effective.  Besides direct interests, a positive populist effect is 
expected too: many of those who would be proud to become GNU contributors, but 
are stopped by current discriminative policies, will be happy about that change.
|
| But are you all really calling for changes, or these ‘all and everyone’ are 
mere buzzwords to please angry SJWs out there?  If the latter, they sound 
hypocritical as hell, sorry.

[1] 

> Also consider that some of us might be using the USA legal definition of 
> "class" here wrt discrimination:
>
>   https://definitions.uslegal.com/p/protected-class/
>
> Defining your own classes outside of those might lead to misunderstandings.

That’s really enlightening, thank you!  I was aware of the trend to overload 
definition of class:

| (sociology, countable) A social grouping, based on job, wealth, etc.
— https://en.wiktionary.org/wiki/class#English

with arbitrary meanings, but till today I did not realize, that it drown in 
them completely, so that the original one might be passed for ‘my own’.


signature.asc
Description: PGP signature

Re: GNU project _does_ discriminate contributors by classes

[Jason Self]

On Wed, 2019-10-30 at 15:04 +0300, Dmitry Alexandrov wrote:
> Jason Self  wrote:
> > 
> > On Tue, 2019-10-29 at 17:43 +0300, Dmitry Alexandrov wrote:
> > > 
> > > To the best of my knowledge, thatʼs completely untrue: major GNU
> > > subprojects do discriminate contributors by classes: if a
> > > contributor-to-be happens to be an employee, FSF does not trust
> > > his words about origin of his contribution, he has to bring a
> > > reference from his employer.
> > > 
> > > Speaking frankly, even if we put aside how time-consuming it may
> > > be, itʼs hard for me to imagine what can be more degrading and
> > > thus ‘alienating’ to someone, than a straightaway demand to
> > > prove that he is not a liar.
> > I believe this to be a mischaracterization of the situation.
> > 
> > This is related to the copyright assignment mentioned earlier. As
> > an example, some have employment contracts with wording to the
> > effect that "anything you ever do anywhere ever always belongs to
> > us no matter what." People with such employment contracts aren't
> > able to assign the copyright for their work because it was never
> > theirs to begin with and the employer needs to do so. The
> > information at [0] mentions this that "we ***may*** also need an
> > employer’s disclaimer..." because...
> ...FSF does not believe in GNU contributorsʼ honesty?

Um, no - that's not the case as I explained. Your original message had
that if any contributor-to-be happens to be an employee they would
need to get something from their employer. I'm quoting the "may" piece
exactly because it says "may" and not "must" because not all
situations will be the same. It's merely a case of getting the
copyright assignment from the actual holder of the copyright, whether
that's the individual or their employer based on whatever the
situation is.

> Could you elaborate, if thatʼs not a secret?  Did you not have to
> get one, because some paper already was in your possession and you
> send a copy of it?

They took my word for it and I never turned in anything. I did not
need to take any additional steps to "prove" anything. My direct
firsthand experience directly conflicts with what you allege to be the
case.

P.S.; there's no need to address the message to me directly - I am on
the list.

signature.asc
Description: This is a digitally signed message part

Re: GNU project _does_ discriminate contributors by classes

[Dmitry Alexandrov]

Jason Self  wrote:
> On Tue, 2019-10-29 at 17:43 +0300, Dmitry Alexandrov wrote:
>> To the best of my knowledge, thatʼs completely untrue: major GNU subprojects 
>> do discriminate contributors by classes: if a contributor-to-be happens to 
>> be an employee, FSF does not trust his words about origin of his 
>> contribution, he has to bring a reference from his employer.
>>
>> Speaking frankly, even if we put aside how time-consuming it may be, itʼs 
>> hard for me to imagine what can be more degrading and thus ‘alienating’ to 
>> someone, than a straightaway demand to prove that he is not a liar.
>
> I believe this to be a mischaracterization of the situation.
>
> This is related to the copyright assignment mentioned earlier. As an example, 
> some have employment contracts with wording to the effect that "anything you 
> ever do anywhere ever always belongs to us no matter what." People with such 
> employment contracts aren't able to assign the copyright for their work 
> because it was never theirs to begin with and the employer needs to do so. 
> The information at [0] mentions this that "we ***may*** also need an 
> employer’s disclaimer..." because...

...FSF does not believe in GNU contributorsʼ honesty?

> it is not required for all situations of all people with employers (with your 
> "if a contributor-to-be happens to be an employee" statement.) I did not have 
> to get one, for example, even though I am employed because I am not in such a 
> situation.

Could you elaborate, if thatʼs not a secret?  Did you not have to get one, 
because some paper already was in your possession and you send a copy of it?


signature.asc
Description: PGP signature

Re: GNU project _does_ discriminate contributors by classes

[Ruben Safir]

On 10/29/19 11:49 AM, DJ Delorie wrote:
>> if a contributor-to-be happens to be an employee, FSF does not trust
>> his words about origin of his contribution,
> This seems reasonable to me in the USA.  Many companies have a clause in
> their contracts that say that the company owns anything the employee
> creates during their tenure, *even off hours*.  Given how complex
> employment contracts are, it's reasonable to ask for a legal disclaimer
> from employers, much like we ask for assignments from contributors.
> It's not about trusting the people involved, it's about protecting
> against people *not* involved who may have bad intentions, who may take
> advantage of honest mistakes.


I've actually taken a magic marker out and crossed those clasues out of
the contract... take or leave it.

Nobody owns whats on my computer in the house.

It can get complex though when you are using software you created under
the GPL for work, which happens often.  There really needs to be a clear
understanding about what is yours and what is theirs.



-- 
So many immigrant groups have swept through our town
that Brooklyn, like Atlantis, reaches mythological
proportions in the mind of the world - RI Safir 1998
http://www.mrbrklyn.com
DRM is THEFT - We are the STAKEHOLDERS - RI Safir 2002

http://www.nylxs.com - Leadership Development in Free Software
http://www.brooklyn-living.com

Being so tracked is for FARM ANIMALS and extermination camps,
but incompatible with living as a free human being. -RI Safir 2013

Re: GNU project _does_ discriminate contributors by classes (was: A GNU “social contract”?)

[Jason Self]

On Tue, 2019-10-29 at 17:43 +0300, Dmitry Alexandrov wrote:
> if a contributor-to-be happens to be an employee

I believe this to be a mischaracterization of the situation.

This is related to the copyright assignment mentioned earlier. As an
example, some have employment contracts with wording to the effect
that "anything you ever do anywhere ever always belongs to us no
matter what." People with such employment contracts aren't able to
assign the copyright for their work because it was never theirs to
begin with and the employer needs to do so. The information at [0]
mentions this that "we ***may*** also need an employer’s
disclaimer..." because it is not required for all situations of all
people with employers (with your "if a contributor-to-be happens to be
an employee" statement.) I did not have to get one, for example, even
though I am employed because I am not in such a situation.

[0] https://www.gnu.org/prep/maintain/html_node/Copyright-Papers.html

signature.asc
Description: This is a digitally signed message part

Re: GNU project _does_ discriminate contributors by classes (was: A GNU “social contract”?)

[DJ Delorie]

> if a contributor-to-be happens to be an employee, FSF does not trust
> his words about origin of his contribution,

This seems reasonable to me in the USA.  Many companies have a clause in
their contracts that say that the company owns anything the employee
creates during their tenure, *even off hours*.  Given how complex
employment contracts are, it's reasonable to ask for a legal disclaimer
from employers, much like we ask for assignments from contributors.
It's not about trusting the people involved, it's about protecting
against people *not* involved who may have bad intentions, who may take
advantage of honest mistakes.

The FSF has always been careful about legal clarity of their ownership
of GNU contributions; employer disclaimers is just another one of these.

Also consider that some of us might be using the USA legal definition of
"class" here wrt discrimination:

  https://definitions.uslegal.com/p/protected-class/

Defining your own classes outside of those might lead to
misunderstandings.