Luk Claes | 1 May 18:55 2009
Picon

Overriding vs Amending vs Position statement

Hi

There seem to be some disagreements about the terms in the subject. As 
far as I'm concerned it's pretty clear though and would not need any 
vote to clarify:

Overriding is only used in combination with decisions. You cannot 
override a document or its interpretation/meaning. You can only override 
a DD's (or delegate's) decision.

Amending is changing something written like a foundation document or a 
proposal. As far as I can see you can only amend explicitly so there 
should be no confusion.

A position statement is a decided on proposal that clarifies the 
position of the Debian project, but does not explicitly amend a 
foundation document.

Supersession of a Foundation Document is replacing a Foundation Document 
with another version: introduce a new one, ammend one or remove one.

According to the Debian Constitution there is only a 3:1 majority needed 
to ammend the Constitution or supersede a Foundation Document.

So I don't really see what we should vote on unless someone disagrees 
with above interpretations?

Cheers

Luk
(Continue reading)

Don Armstrong | 1 May 20:56 2009
Picon

Re: Overriding vs Amending vs Position statement

On Fri, 01 May 2009, Luk Claes wrote:
> A position statement is a decided on proposal that clarifies the
> position of the Debian project, but does not explicitly amend a
> foundation document.

[...]

> So I don't really see what we should vote on unless someone
> disagrees with above interpretations?

The only question resides with the effect of passing such position
statements. Without modifying foundation documents or the
constitution, they are effectively non-binding advisory statements
when operating within areas that are the remit of foundation documents
or the constitution.

Developers can ignore (or follow) such statements as they wish.

Furthermore, the statements must be non-technical.

Don Armstrong

--

-- 
Filing a bug is probably not going to get it fixed any faster.
 -- Anthony Towns

http://www.donarmstrong.com              http://rzlab.ucr.edu

Matthew Johnson | 1 May 21:08 2009
Picon

Re: Overriding vs Amending vs Position statement

On Fri May 01 11:56, Don Armstrong wrote:
> > So I don't really see what we should vote on unless someone
> > disagrees with above interpretations?
> 
> The only question resides with the effect of passing such position
> statements. Without modifying foundation documents or the
> constitution, they are effectively non-binding advisory statements
> when operating within areas that are the remit of foundation documents
> or the constitution.

Indeed and there is the case of temporary exceptions. Does saying "we
will release with non-free stuff" involve modifying a foundation
document? I would say yes. Does saying "we will release Lenny with
non-free stuff" involve modifying a foundation document? There seems to
be less agreement on this. I think it does, but the previous discussion
showed that some people disagree.

Anyway, I'm going to try and push that discussion a bit more in a
couple of hours, so probably best to discuss in that context

--

-- 
Matthew Johnson
Luk Claes | 1 May 22:19 2009
Picon

Re: Overriding vs Amending vs Position statement

Matthew Johnson wrote:
> On Fri May 01 11:56, Don Armstrong wrote:
>>> So I don't really see what we should vote on unless someone
>>> disagrees with above interpretations?
>> The only question resides with the effect of passing such position
>> statements. Without modifying foundation documents or the
>> constitution, they are effectively non-binding advisory statements
>> when operating within areas that are the remit of foundation documents
>> or the constitution.
> 
> Indeed and there is the case of temporary exceptions. Does saying "we
> will release with non-free stuff" involve modifying a foundation
> document? I would say yes. Does saying "we will release Lenny with
> non-free stuff" involve modifying a foundation document? There seems to
> be less agreement on this. I think it does, but the previous discussion
> showed that some people disagree.

This always sounds very awkward to me. So if we would just not fix bugs 
about non-free stuff everything is ok, but if we want to release it has 
either to be fixed very quickly or get a vote that modifies a foundation 
document? Sorry, but I did not and will not agree with that.

We will not release with random non-free stuff, nor will we release with 
easily fixable non-free stuff, nor will we release with non-free stuff 
where it's clear that upstream does not care in fixing it. We will 
release with non-free stuff that does not get fixed in time where 
upsteam is working on it though.

I don't see why this would need any vote, though if you really think 
it's useful to have a vote on this, so be it.
(Continue reading)

Manoj Srivastava | 2 May 00:13 2009
X-Face
Face
Picon

Re: Overriding vs Amending vs Position statement

On Fri, May 01 2009, Don Armstrong wrote:

> On Fri, 01 May 2009, Luk Claes wrote:
>> A position statement is a decided on proposal that clarifies the
>> position of the Debian project, but does not explicitly amend a
>> foundation document.
>
> [...]
>
>> So I don't really see what we should vote on unless someone
>> disagrees with above interpretations?
>
> The only question resides with the effect of passing such position
> statements. Without modifying foundation documents or the
> constitution, they are effectively non-binding advisory statements
> when operating within areas that are the remit of foundation documents
> or the constitution.

> Developers can ignore (or follow) such statements as they wish.

        If the statements are in contradiction of the foundation
 document (which is the case in a couple of prior situations), then are
 you saying that anything in the foundation documents can ve worked
 around by putting out a position statement, and have the developers
 proceed to ignore the foundation document on that basis?

        That also begs the question: do we _have_ to follow the
 foundation documents? Or can one just issue a statement "I do not agree
 with the foundation doc" and just ignore it at will?

(Continue reading)

Luk Claes | 2 May 00:52 2009
Picon

Re: Overriding vs Amending vs Position statement

Manoj Srivastava wrote:
> On Fri, May 01 2009, Don Armstrong wrote:
> 
>> On Fri, 01 May 2009, Luk Claes wrote:
>>> A position statement is a decided on proposal that clarifies the
>>> position of the Debian project, but does not explicitly amend a
>>> foundation document.
>> [...]
>>
>>> So I don't really see what we should vote on unless someone
>>> disagrees with above interpretations?
>> The only question resides with the effect of passing such position
>> statements. Without modifying foundation documents or the
>> constitution, they are effectively non-binding advisory statements
>> when operating within areas that are the remit of foundation documents
>> or the constitution.
> 
>> Developers can ignore (or follow) such statements as they wish.
> 
>         If the statements are in contradiction of the foundation
>  document (which is the case in a couple of prior situations), then are
>  you saying that anything in the foundation documents can ve worked
>  around by putting out a position statement, and have the developers
>  proceed to ignore the foundation document on that basis?

Of course not. If a position statement contradicts a foundation document 
it's time to update the foundation document accordingly or drop the 
position statement again.

>         That also begs the question: do we _have_ to follow the
(Continue reading)

Matthew Johnson | 2 May 00:54 2009
Picon

Re: Overriding vs Amending vs Position statement

On Sat May 02 00:52, Luk Claes wrote:
> It would be a clear indication that the foundation document should get an 
> update or that the postition statement should get dropped again.

I think Manoj's point is that if voting some option X (a position
statement in conflict with an FD) means that we have to vote to change
the FD or drop X, then why wasn't X a vote to change the FD in the first
place? Surely we don't need a vote just to then have another vote...

Matt
--

-- 
Matthew Johnson
Steve Langasek | 2 May 01:16 2009
Picon

Re: Overriding vs Amending vs Position statement

On Fri, May 01, 2009 at 11:54:15PM +0100, Matthew Johnson wrote:
> On Sat May 02 00:52, Luk Claes wrote:
> > It would be a clear indication that the foundation document should get an 
> > update or that the postition statement should get dropped again.

> I think Manoj's point is that if voting some option X (a position
> statement in conflict with an FD) means that we have to vote to change
> the FD or drop X, then why wasn't X a vote to change the FD in the first
> place? Surely we don't need a vote just to then have another vote...

No one has the authority to declare, a priori, for the entire project, that
a given position statement is in conflict with a FD.

--

-- 
Steve Langasek                   Give me a lever long enough and a Free OS
Debian Developer                   to set it on, and I can move the world.
Ubuntu Developer                                    http://www.debian.org/
slangasek <at> ubuntu.com                                     vorlon <at> debian.org

Manoj Srivastava | 2 May 01:50 2009
X-Face
Face
Picon

Re: Overriding vs Amending vs Position statement

On Fri, May 01 2009, Steve Langasek wrote:

> On Fri, May 01, 2009 at 11:54:15PM +0100, Matthew Johnson wrote:
>> On Sat May 02 00:52, Luk Claes wrote:
>> > It would be a clear indication that the foundation document should get an 
>> > update or that the postition statement should get dropped again.
>
>> I think Manoj's point is that if voting some option X (a position
>> statement in conflict with an FD) means that we have to vote to change
>> the FD or drop X, then why wasn't X a vote to change the FD in the first
>> place? Surely we don't need a vote just to then have another vote...
>
> No one has the authority to declare, a priori, for the entire project, that
> a given position statement is in conflict with a FD.

        Does anyone have authority, a posteriori, to declare that any
 given position statement is in contradiction of a foundation document?

        Or is it only deliverable by a GR?

        This will be interesting. So, in order to determine whether a
 foundation document is being modified, we first ask the  project, via a
 GR, whether it is indeed a contradiction. _THEN_ we hold a vote, with
 or without the 3:1 majority, based o the previous vote, to see if it
 passes or not.

        I think Joey Hess is right.

        manoj
--

-- 
(Continue reading)

Matthew Johnson | 2 May 09:29 2009
Picon

Re: Overriding vs Amending vs Position statement

On Fri May 01 16:16, Steve Langasek wrote:
> On Fri, May 01, 2009 at 11:54:15PM +0100, Matthew Johnson wrote:
> > On Sat May 02 00:52, Luk Claes wrote:
> > > It would be a clear indication that the foundation document should get an 
> > > update or that the postition statement should get dropped again.
> 
> > I think Manoj's point is that if voting some option X (a position
> > statement in conflict with an FD) means that we have to vote to change
> > the FD or drop X, then why wasn't X a vote to change the FD in the first
> > place? Surely we don't need a vote just to then have another vote...
> 
> No one has the authority to declare, a priori, for the entire project, that
> a given position statement is in conflict with a FD.

I think that is somewhat of an orthogonal issue. I don't think anyone
would disagree that the vote: 

   We agree to ship the nvidia binary drivers in main

conflicts with one of the foundation documents. At the moment, however,
we could run that vote and since it doesn't explicitly modify one of the
foundation documents, it would only require a simple majority. Now, if
people think that a simple majority should be able to decide this, then
fine, but drop the 3:1 requirement from the constitution. 

If the project thinks this _should_ require 3:1 then I would like that
enshrined in the constitution so that Kurt doesn't have to resign over
it as well, next time this comes up. If that is the case, then of course
we also need to decide who makes that decision. you say "for the entire
project"---surely a position statement conflicts with a FD or it
(Continue reading)

Ben Finney | 2 May 10:48 2009
Picon

Re: Overriding vs Amending vs Position statement

Matthew Johnson <mjj29 <at> debian.org> writes:

> On Fri May 01 16:16, Steve Langasek wrote:
> > No one has the authority to declare, a priori, for the entire
> > project, that a given position statement is in conflict with a FD.

This seems to advocate the possibility that a statement could be in
conflict with the foundation documents “for some people”.

Are you saying the statement “this proposal conflicts with the
foundation documents” can be true for some people simultaneously with
being false for other people?

[…]

> If the project thinks [proposals which conflict the foundation
> documents, but don't say so explicitly] _should_ require 3:1 then I
> would like that enshrined in the constitution so that Kurt doesn't
> have to resign over it as well, next time this comes up. If that is
> the case, then of course we also need to decide who makes that
> decision. you say "for the entire project"---surely a position
> statement conflicts with a FD or it doesn't, whole project or no.

Absolutely agreed with this. We may not agree on *whether* a given
proposal conflicts with the foundation documents, but (unless we want to
have the ludicrous notion that the conflict both exists and does not
exist) someone needs to decide which is the case in order to determine
whether a supermajority requirement applies.

--

-- 
(Continue reading)

Russ Allbery | 2 May 16:15 2009
Picon

Re: Overriding vs Amending vs Position statement

Ben Finney <ben+debian <at> benfinney.id.au> writes:
> Matthew Johnson <mjj29 <at> debian.org> writes:
>> On Fri May 01 16:16, Steve Langasek wrote:

>>> No one has the authority to declare, a priori, for the entire
>>> project, that a given position statement is in conflict with a FD.

> This seems to advocate the possibility that a statement could be in
> conflict with the foundation documents “for some people”.
>
> Are you saying the statement “this proposal conflicts with the
> foundation documents” can be true for some people simultaneously with
> being false for other people?

Of course it can be!  That would only not be true if we had unanimity
over the meaning of the foundation documents, which we clearly do not,
or if we had a body in Debian with the power to declare the canonical
meaning of the foundation documents for all developers, which similarly
we do not.

--

-- 
Russ Allbery (rra <at> debian.org)               <http://www.eyrie.org/~eagle/>

Ben Finney | 3 May 02:21 2009
Picon

Re: Overriding vs Amending vs Position statement

Russ Allbery <rra <at> debian.org> writes:

> Ben Finney <ben+debian <at> benfinney.id.au> writes:
> > Are you saying the statement “this proposal conflicts with the
> > foundation documents” can be true for some people simultaneously
> > with being false for other people?
> 
> Of course it can be!  That would only not be true if we had unanimity
> over the meaning of the foundation documents, which we clearly do not,

So, in effect, you advocate the position that “the foundation documents”
refers to a different set of documents depending on who is being asked?

> or if we had a body in Debian with the power to declare the canonical
> meaning of the foundation documents for all developers, which similarly
> we do not.

To the extent that we need to take different action depending on whether
a proposal conflicts with the foundation documents, is it not true that
we need a body with the power to *make decisions* about the truth of
statements like “this proposal conflicts with the foundation documents”?

The only way I can see that power being unnecessary is if nothing hinges
on whether a proposal conflicts with foundation documents. If, on the
other hand, anything *does* hinge on that determination, someone needs
to *make* that determination in cases where actions depend on it.

--

-- 
 \                             “Holy tintinnabulation, Batman!” —Robin |
  `\                                                                   |
(Continue reading)

Russ Allbery | 3 May 04:31 2009
Picon

Re: Overriding vs Amending vs Position statement

Ben Finney <ben+debian <at> benfinney.id.au> writes:
> Russ Allbery <rra <at> debian.org> writes:
>> Ben Finney <ben+debian <at> benfinney.id.au> writes:

>>> Are you saying the statement “this proposal conflicts with the
>>> foundation documents” can be true for some people simultaneously
>>> with being false for other people?

>> Of course it can be!  That would only not be true if we had unanimity
>> over the meaning of the foundation documents, which we clearly do not,

> So, in effect, you advocate the position that “the foundation documents”
> refers to a different set of documents depending on who is being asked?

No.  That's an absurd interpretation of what I said.

>> or if we had a body in Debian with the power to declare the canonical
>> meaning of the foundation documents for all developers, which
>> similarly we do not.

> To the extent that we need to take different action depending on
> whether a proposal conflicts with the foundation documents, is it not
> true that we need a body with the power to *make decisions* about the
> truth of statements like “this proposal conflicts with the foundation
> documents”?

Apparently not, since the project has survived for over a decade without
one.  It would reduce the number of these sorts of discussions we have
to waste time with on mailing lists, so there would be some benefits,
but it's obviously not something we *need*.
(Continue reading)

Ben Finney | 3 May 04:48 2009
Picon

Re: Overriding vs Amending vs Position statement

Russ Allbery <rra <at> debian.org> writes:

> Ben Finney <ben+debian <at> benfinney.id.au> writes:
> > Russ Allbery <rra <at> debian.org> writes:
> >> Ben Finney <ben+debian <at> benfinney.id.au> writes:
> 
> >>> Are you saying the statement “this proposal conflicts with the
> >>> foundation documents” can be true for some people simultaneously
> >>> with being false for other people?
> 
> >> Of course it can be!  That would only not be true if we had unanimity
> >> over the meaning of the foundation documents, which we clearly do not,
> 
> > So, in effect, you advocate the position that “the foundation
> > documents”refers to a different set of documents depending on who
> > is being asked?
> 
> No.  That's an absurd interpretation of what I said.

Yet I can't disambiguate it from this:

> > The only way I can see that power being unnecessary is if nothing
> > hinges on whether a proposal conflicts with foundation documents.
> > If, on the other hand, anything *does* hinge on that determination,
> > someone needs to *make* that determination in cases where actions
> > depend on it.
> 
> And who makes that decision has already been explained at *ridiculous*
> length on this mailing list, so I'll assume you already know how that
> works.
(Continue reading)

Russ Allbery | 3 May 05:03 2009
Picon

Re: Overriding vs Amending vs Position statement

Ben Finney <ben+debian <at> benfinney.id.au> writes:
> Russ Allbery <rra <at> debian.org> writes:
>> Ben Finney <ben+debian <at> benfinney.id.au> writes:
>>> Russ Allbery <rra <at> debian.org> writes:

>>> So, in effect, you advocate the position that “the foundation
>>> documents”refers to a different set of documents depending on who
>>> is being asked?

>> No.  That's an absurd interpretation of what I said.

> Yet I can't disambiguate it from this:

> I presume this is referring to the practice of leaving the
> determination to each individual person acting. Which, in effect, is
> allowing that the foundation documents have a different meaning for
> each person and none of them are wrong.

Yup.  If there's no project 3:1 majority about what the foundation
documents mean in a specific case, that is indeed the case.

This is not the same thing as using completely different documents.

This happens in law all the time.  That's one of the big reasons why
nations have a court system: resolving disputes between people with
different competing legal interpretations of the law.  It's not
equivalent to everyone having a completely different legal code.

Maybe an analogy would help.  I'm pointing out that people have a bunch
of different philosophical beliefs about the nature of the world, and
(Continue reading)

Russ Allbery | 2 May 16:10 2009
Picon

Re: Overriding vs Amending vs Position statement

Matthew Johnson <mjj29 <at> debian.org> writes:

> I think that is somewhat of an orthogonal issue. I don't think anyone
> would disagree that the vote:
>
>    We agree to ship the nvidia binary drivers in main
>
> conflicts with one of the foundation documents. At the moment, however,
> we could run that vote and since it doesn't explicitly modify one of the
> foundation documents, it would only require a simple majority. Now, if
> people think that a simple majority should be able to decide this, then
> fine, but drop the 3:1 requirement from the constitution. 

I have to say, I'm finding it rather frustrating that people keep
repeating this argument without apparently acknowledging that it's been
addressed.  If you don't agree with the response to this argument,
that's fine, but not acknowledging the rebuttal is getting on my nerves.
(It's possible, though, that you've just not seen the previous
discussion for whatever reason.)

To recap, the counter-argument is that such a *non-binding* position
statement is obviously nonsensical and hence people aren't going to
follow it even if it passes, which it won't because it's non-sensical.
In other words, you're making a reductio ad absurdum argument in a place
where there are other controls.  If a majority of people in Debian voted
for a non-binding position statement that so obviously, clearly, and
directly contradicts a foundation document, we have considerably more
problems than the question of a supermajority.

Furthermore, I think there's general consensus even among those of us
(Continue reading)

Clint Adams | 2 May 16:43 2009
Picon

Re: Overriding vs Amending vs Position statement

On Sat, May 02, 2009 at 07:10:07AM -0700, Russ Allbery wrote:
>     The Debian project believes that shipping NVidia drivers in main is
>     consistent with the current DFSG and Social Contract.
> 
> If you think there's any serious danger of that passing with a majority,
> I would contend that you're essentially arguing there's such a serious
> disagreement in Debian over this issue that we do not even share the
> same language, terms, and basis for discussion.  I don't see that
> pessimism supported by any of the previous votes or by the general
> discussion here.

Would your statement still apply if you replace "NVidia drivers" with
"non-free kernel firmware"?  Since I view those as equivalent, and
the latter seems far more dangerous, I do think the pessimism is
warranted.

Russ Allbery | 2 May 22:16 2009
Picon

Re: Overriding vs Amending vs Position statement

Clint Adams <schizo <at> debian.org> writes:
> On Sat, May 02, 2009 at 07:10:07AM -0700, Russ Allbery wrote:

>>     The Debian project believes that shipping NVidia drivers in main is
>>     consistent with the current DFSG and Social Contract.

>> If you think there's any serious danger of that passing with a
>> majority, I would contend that you're essentially arguing there's
>> such a serious disagreement in Debian over this issue that we do not
>> even share the same language, terms, and basis for discussion.  I
>> don't see that pessimism supported by any of the previous votes or by
>> the general discussion here.

> Would your statement still apply if you replace "NVidia drivers" with
> "non-free kernel firmware"?

Well, part of the dispute is over the definition of non-free.  If one
defines non-free as non-compliant with the DFSG, then such a statement
would be internally contradictory.  However, semantics aside, for the
definition of non-free I'm pretty sure you're using (no source
available), I think that's basically the GR we passed for the release of
lenny.  So no.  :)

> Since I view those as equivalent, and the latter seems far more
> dangerous, I do think the pessimism is warranted.

However, for two releases in a row, the project has wanted to release
with non-source kernel firmware.

I think this gets to the heart of the current conflict: the people who
(Continue reading)

Manoj Srivastava | 3 May 00:48 2009
X-Face
Face
Picon

Re: Overriding vs Amending vs Position statement

On Sat, May 02 2009, Russ Allbery wrote:

> Clint Adams <schizo <at> debian.org> writes:
>> On Sat, May 02, 2009 at 07:10:07AM -0700, Russ Allbery wrote:
>
>>>     The Debian project believes that shipping NVidia drivers in main is
>>>     consistent with the current DFSG and Social Contract.
>
>>> If you think there's any serious danger of that passing with a
>>> majority, I would contend that you're essentially arguing there's
>>> such a serious disagreement in Debian over this issue that we do not
>>> even share the same language, terms, and basis for discussion.  I
>>> don't see that pessimism supported by any of the previous votes or by
>>> the general discussion here.
>
>> Would your statement still apply if you replace "NVidia drivers" with
>> "non-free kernel firmware"?
>
> Well, part of the dispute is over the definition of non-free.

        I thought the whole poing of including the DFSG in the SC was
 that there would be no ambiguity about what the Debian project means
 when it says something is non-free. Let me see how it is phrased:

     We provide the guidelines that we use to determine if a work is
     "free" in the document entitled "The Debian Free Software
     Guidelines".  

> If one defines non-free as non-compliant with the DFSG, then such a
> statement would be internally contradictory.  However, semantics
(Continue reading)

Matthew Johnson | 2 May 18:09 2009
Picon

Re: Overriding vs Amending vs Position statement

On Sat May 02 07:10, Russ Allbery wrote:
> To recap, the counter-argument is that such a *non-binding* position
> statement is obviously nonsensical and hence people aren't going to
> follow it even if it passes, which it won't because it's non-sensical.
> In other words, you're making a reductio ad absurdum argument in a place
> where there are other controls.  If a majority of people in Debian voted
> for a non-binding position statement that so obviously, clearly, and
> directly contradicts a foundation document, we have considerably more
> problems than the question of a supermajority.

Sorry if it looks like I was ignoring that Russ, I wasn't (and I
presented, I hope, your view in one of the options on my proposed
ballot).

I was trying to demonstrate that there are things which are in conflict
with a foundation document and that something needs to be done about
vote options which are such conflicts but don't explicitly amend that
document.

The solution could be as you say that they are therefore non-binding
(although I don't see that is is a _useful_ solution, see Don and
 Manoj's posts to this thread, it's at least _consistent). It could be
that they need 3:1 to pass whether they are explicit modifications to
the document or not (which is what I always thought was the case), or
it could be that they are binding even without a super majority (and
it's this view I was trying to address here.

All of these are consistent views held by several of the contributors to
the various threads, and all of them consider that to be what the status
quo is. I wish to clarify it one way or another
(Continue reading)

Russ Allbery | 3 May 04:36 2009
Picon

Re: Overriding vs Amending vs Position statement

Matthew Johnson <mjj29 <at> debian.org> writes:

> I was trying to demonstrate that there are things which are in
> conflict with a foundation document and that something needs to be
> done about vote options which are such conflicts but don't explicitly
> amend that document.

Oh, okay, sorry.  Yes, I do agree with that.

> The solution could be as you say that they are therefore non-binding
> (although I don't see that is is a _useful_ solution, see Don and
> Manoj's posts to this thread, it's at least _consistent).

I've posted in the past why I think it's useful, but I'm assuming people
have seen that and just don't agree with it.  It's certainly not as
useful as making a binding 3:1 decision, but there are cases where we
*can't* make a 3:1 decision because no option has a 3:1 majority.

> It could be that they need 3:1 to pass whether they are explicit
> modifications to the document or not (which is what I always thought
> was the case), or it could be that they are binding even without a
> super majority (and it's this view I was trying to address here.
>
> All of these are consistent views held by several of the contributors
> to the various threads, and all of them consider that to be what the
> status quo is. I wish to clarify it one way or another

I think the first option (that they need 3:1 to pass whether they're
explicit or not) just begs the question and therefore won't solve any of
our problems unless we also identify a specific body who decides what
(Continue reading)

Matthew Johnson | 3 May 09:38 2009
Picon

Re: Overriding vs Amending vs Position statement

On Sat May 02 19:36, Russ Allbery wrote:

> I think the first option (that they need 3:1 to pass whether they're
> explicit or not) just begs the question and therefore won't solve any of
> our problems unless we also identify a specific body who decides what
> does and does not modify foundation documents, which in practice is
> identical to defining a body that states what the foundation documents
> mean for all developers.

In all cases this is required. In your case, it's needed to determine
whether an option is binding or not. After all, if something does not
conflict with a foundation document at all, passing it by simple
majority is binding, whereas if it does conflict then it is not.

To put it another way, who decides whether to demand the author chose
between non-binding and 3:1, if they think it's not conflicting.

Matt

--

-- 
Matthew Johnson
Russ Allbery | 3 May 10:14 2009
Picon

Re: Overriding vs Amending vs Position statement

Matthew Johnson <mjj29 <at> debian.org> writes:
> On Sat May 02 19:36, Russ Allbery wrote:

>> I think the first option (that they need 3:1 to pass whether they're
>> explicit or not) just begs the question and therefore won't solve any
>> of our problems unless we also identify a specific body who decides
>> what does and does not modify foundation documents, which in practice
>> is identical to defining a body that states what the foundation
>> documents mean for all developers.

> In all cases this is required. In your case, it's needed to determine
> whether an option is binding or not. After all, if something does not
> conflict with a foundation document at all, passing it by simple
> majority is binding, whereas if it does conflict then it is not.

> To put it another way, who decides whether to demand the author chose
> between non-binding and 3:1, if they think it's not conflicting.

Hm.  Section 4.1 lays out what GRs are for.  Most of the classes of
binding GRs look rather distinct from each other to me.  Binding GRs
are:

* Appoint or recall the Project Leader
* Make or override a DPL or delegate decision
* Make or override a tech-ctte decision (2:1)
* Modify foundation documents (3:1)
* Make decisions about property held in trust by Debian
* Appoint a secretary

I'd say that everything else is non-binding, specifically including
(Continue reading)

Matthew Johnson | 3 May 10:55 2009
Picon

Re: Overriding vs Amending vs Position statement

On Sun May 03 01:14, Russ Allbery wrote:
> Hm.  Section 4.1 lays out what GRs are for.  Most of the classes of
> binding GRs look rather distinct from each other to me.  Binding GRs
> are:
> 
> * Appoint or recall the Project Leader
> * Make or override a DPL or delegate decision
> * Make or override a tech-ctte decision (2:1)
> * Modify foundation documents (3:1)
> * Make decisions about property held in trust by Debian
> * Appoint a secretary
> 
> I'd say that everything else is non-binding, specifically including
> anything passed under point 5 that doesn't modify a foundation document.

Really? I don't see anything which says they are non-binding, but I do
see 2.1.1: "Nothing in this constitution imposes an obligation on anyone
to do work for the Project. A person who does not want to do a task
which has been delegated or assigned to them does not need to do it.
However, they must not actively work against these rules and decisions
properly made under them."

This would suggest that any decision made under the constitution (eg, by
way of GR) is as binding as it is possible to be (you can always refuse
to do the work)

Matt

--

-- 
Matthew Johnson
(Continue reading)

Russ Allbery | 3 May 15:44 2009
Picon

Re: Overriding vs Amending vs Position statement

Matthew Johnson <mjj29 <at> debian.org> writes:
> On Sun May 03 01:14, Russ Allbery wrote:

>> Hm.  Section 4.1 lays out what GRs are for.  Most of the classes of
>> binding GRs look rather distinct from each other to me.  Binding GRs
>> are:

>> * Appoint or recall the Project Leader
>> * Make or override a DPL or delegate decision
>> * Make or override a tech-ctte decision (2:1)
>> * Modify foundation documents (3:1)
>> * Make decisions about property held in trust by Debian
>> * Appoint a secretary

>> I'd say that everything else is non-binding, specifically including
>> anything passed under point 5 that doesn't modify a foundation document.

> Really? I don't see anything which says they are non-binding, but I do
> see 2.1.1: "Nothing in this constitution imposes an obligation on
> anyone to do work for the Project. A person who does not want to do a
> task which has been delegated or assigned to them does not need to do
> it.  However, they must not actively work against these rules and
> decisions properly made under them."

> This would suggest that any decision made under the constitution (eg, by
> way of GR) is as binding as it is possible to be (you can always refuse
> to do the work)

Well, "position statements about issues of the day" are obviously
non-binding, so I guess we're disagreeing over "nontechnical policy
(Continue reading)

Matthew Johnson | 4 May 01:47 2009
Picon

Re: Overriding vs Amending vs Position statement

On Sun May 03 06:44, Russ Allbery wrote:
> > Really? I don't see anything which says they are non-binding, but I do
> > see 2.1.1: "Nothing in this constitution imposes an obligation on
> > anyone to do work for the Project. A person who does not want to do a
> > task which has been delegated or assigned to them does not need to do
> > it.  However, they must not actively work against these rules and
> > decisions properly made under them."
> 
> > This would suggest that any decision made under the constitution (eg, by
> > way of GR) is as binding as it is possible to be (you can always refuse
> > to do the work)
> 
> Well, "position statements about issues of the day" are obviously
> non-binding, so I guess we're disagreeing over "nontechnical policy
> documents and statements" other than those.  The constitution offers the
> examples of "documents describing the goals of the project, its
> relationship with other free software entities, and nontechnical
> policies such as the free software licence terms that Debian software
> must meet."  The first two sound pretty non-binding to me, and the
> latter is the DFSG, for which foundation document rules apply.
> 
> Am I missing something?

Well, where would you say that the following GRs would fit:

http://www.debian.org/vote/2006/vote_001 ("GFDL w/o invariant sections is free", 1:1)
http://www.debian.org/vote/2007/vote_002 ("DDs can do binary only uploads", 1:1)
http://www.debian.org/vote/2007/vote_003 ("Endorse Debian Maintainers", 1:1)

to pick some examples. These aren't  in your list of "things which are
(Continue reading)

Russ Allbery | 4 May 02:08 2009
Picon

Re: Overriding vs Amending vs Position statement

Matthew Johnson <mjj29 <at> debian.org> writes:

> Well, where would you say that the following GRs would fit:
>
> http://www.debian.org/vote/2006/vote_001 ("GFDL w/o invariant sections is free", 1:1)

Non-binding position statement.  It doesn't really need to be binding
since the people who were doing the work didn't think it contradicted
the DFSG.  That's the best use of project policy statements, I think:
they're highly persuasive to the people doing the work.

> http://www.debian.org/vote/2007/vote_002 ("DDs can do binary only uploads", 1:1)

Delegate override.  That was pretty clearly a delegate override at the
time.  It explicitly reversed a decision made by a delegate.

> http://www.debian.org/vote/2007/vote_003 ("Endorse Debian Maintainers", 1:1)

Hm.  Not sure about this one.

> These aren't in your list of "things which are binding GRs", but I
> think they should be something we can vote on and they should be
> binding. Possibly this means the constitution is deficient in this
> area.

For the last, yeah, I'm not sure what to make of that.  It does feel
like a binding 1:1 majority decision.  It feels kind of odd that a
decision of that magnitude can be made with a 1:1 majority vote when
things that have less significant impact on the project require 3:1
majorities.
(Continue reading)

Ben Finney | 4 May 02:52 2009
Picon

Re: Overriding vs Amending vs Position statement

Russ Allbery <rra <at> debian.org> writes:

> Matthew Johnson <mjj29 <at> debian.org> writes:
> 
> > Well, where would you say that the following GRs would fit:
> >
> > http://www.debian.org/vote/2006/vote_001 ("GFDL w/o invariant sections is free", 1:1)
> 
> Non-binding position statement. It doesn't really need to be binding
> since the people who were doing the work didn't think it contradicted
> the DFSG. That's the best use of project policy statements, I think:
> they're highly persuasive to the people doing the work.

And if someone doing the work is not persuaded by this?

What if one of the many who do *not* find that GR to be persuasive is in
the position to reject a package containing FDL-licensed work, and does
so on the basis that their interpretation of the DFSG and FDL mean that
the package is not free?

By your arguments earlier in this thread, it seems this person's
interpretation, though contradictory with the GR, is equally valid. The
GR is, you say, non-binding. So what is the point of going through the
GR process if it doesn't bind such a person to the decision?

--

-- 
 \         “I think a good gift for the President would be a chocolate |
  `\   revolver. And since he's so busy, you'd probably have to run up |
_o__)              to him real quick and hand it to him.” —Jack Handey |
Ben Finney
(Continue reading)

Russ Allbery | 4 May 03:26 2009
Picon

Re: Overriding vs Amending vs Position statement

Ben Finney <ben+debian <at> benfinney.id.au> writes:

> And if someone doing the work is not persuaded by this?

> What if one of the many who do *not* find that GR to be persuasive is
> in the position to reject a package containing FDL-licensed work, and
> does so on the basis that their interpretation of the DFSG and FDL
> mean that the package is not free?

Then you talk to that person and point out that their opinion is not
shared with the rest of the project, with the GR as strong evidence.  If
that isn't sufficient, that's what the delegate override is for, but in
practice it would never come to that, since if the GR passed, a delegate
override is going to pass too.  Plus, it's never going to come up in
anywhere near that clear-cut of a fashion.

> By your arguments earlier in this thread, it seems this person's
> interpretation, though contradictory with the GR, is equally
> valid. The GR is, you say, non-binding. So what is the point of going
> through the GR process if it doesn't bind such a person to the
> decision?

Because people treat them seriously and follow them voluntarily even if
they don't personally agree.  It feels to me like you're insisting on
adding mechanisms to force poeple to do things into the process that
simply aren't necessary historically.  I would really rather focus on
solving the problems that we actually have, rather than theoretical
problems that assume fellow DDs are going to do obviously stupid things.

--

-- 
(Continue reading)

Ben Finney | 4 May 03:50 2009
Picon

Re: Overriding vs Amending vs Position statement

Russ Allbery <rra <at> debian.org> writes:

> Ben Finney <ben+debian <at> benfinney.id.au> writes:
> > By your arguments earlier in this thread, it seems this person's
> > interpretation, though contradictory with the GR, is equally valid.
> > The GR is, you say, non-binding. So what is the point of going
> > through the GR process if it doesn't bind such a person to the
> > decision?
> 
> Because people treat them seriously and follow them voluntarily even
> if they don't personally agree.

And if they're not convinced? Either the GR is binding or it's not. You
say it's not; but if that's the case, when a person acts in
contradiction to such a GR, what basis does anyone else have for telling
them to stop?

If, on the other hand, the person's actions are prevented on the basis
of the GR, what sense is there in saying that the GR is non-binding?

> It feels to me like you're insisting on adding mechanisms to force
> poeple to do things into the process that simply aren't necessary
> historically.

No. I'm saying that there *are* such mechanisms, as pointed out earlier.
If a GR informs positive action but it's okay to interpret it as
“non-binding”, then we don't have a good basis for preventing actions
in contradiction to the GR. If, on the other hand, we say that GR *is*
binding, then actions that contradict it are harmful and can be stopped
on that basis.
(Continue reading)

Matthew Johnson | 4 May 11:22 2009
Picon

Re: Overriding vs Amending vs Position statement

On Mon May 04 11:50, Ben Finney wrote:
> No. I'm saying that there *are* such mechanisms, as pointed out earlier.
> If a GR informs positive action but it's okay to interpret it as
> “non-binding”, then we don't have a good basis for preventing actions
> in contradiction to the GR. If, on the other hand, we say that GR *is*
> binding, then actions that contradict it are harmful and can be stopped
> on that basis.

Indeed, and I think the constitution  is pretty clear that they _are_
binding, as I posted upthread:

Constitution  2.1.1: 

   "Nothing in this constitution imposes an obligation on anyone to do
   work for the Project. A person who does not want to do a task which
   has been delegated or assigned to them does not need to do it.
   However, they must not actively work against these rules and
   decisions properly made under them."

Any GR is clearly "a decision properly made under them" and therefore
"they must not actively work against [it]". That sounds pretty binding
to me.

> One of the prevalent themes in these discussions is that it isn't even
> close to universal in the project what is “obviously stupid” and what
> isn't. That's why we have decision-making systems.

Absolutely

Matt
(Continue reading)

Luk Claes | 10 May 18:37 2009
Picon

Re: Overriding vs Amending vs Position statement

Matthew Johnson wrote:
> On Sat May 02 00:52, Luk Claes wrote:
>> It would be a clear indication that the foundation document should get an 
>> update or that the postition statement should get dropped again.
> 
> I think Manoj's point is that if voting some option X (a position
> statement in conflict with an FD) means that we have to vote to change
> the FD or drop X, then why wasn't X a vote to change the FD in the first
> place? Surely we don't need a vote just to then have another vote...

Well, I think it's wrong to force the vote to be about changing the FD
when the proposer and the seconders agreed that it should not.

Note that you don't know if there would be another vote needed as one
cannot know beforehand if the vote would pass or not.

Note also that it might be easier to get a position statement than to
find appropriate wording and support to change the FD.

Cheers

Luk

Manoj Srivastava | 2 May 01:48 2009
X-Face
Face
Picon

Re: Overriding vs Amending vs Position statement

On Fri, May 01 2009, Luk Claes wrote:

> Manoj Srivastava wrote:
>> On Fri, May 01 2009, Don Armstrong wrote:
>>
>>> On Fri, 01 May 2009, Luk Claes wrote:
>>>> A position statement is a decided on proposal that clarifies the
>>>> position of the Debian project, but does not explicitly amend a
>>>> foundation document.
>>> [...]
>>>
>>>> So I don't really see what we should vote on unless someone
>>>> disagrees with above interpretations?
>>> The only question resides with the effect of passing such position
>>> statements. Without modifying foundation documents or the
>>> constitution, they are effectively non-binding advisory statements
>>> when operating within areas that are the remit of foundation documents
>>> or the constitution.
>>
>>> Developers can ignore (or follow) such statements as they wish.
>>
>>         If the statements are in contradiction of the foundation
>>  document (which is the case in a couple of prior situations), then are
>>  you saying that anything in the foundation documents can ve worked
>>  around by putting out a position statement, and have the developers
>>  proceed to ignore the foundation document on that basis?
>
> Of course not. If a position statement contradicts a foundation
> document it's time to update the foundation document accordingly or
> drop the position statement again.
(Continue reading)

Don Armstrong | 2 May 00:52 2009
Picon

Re: Overriding vs Amending vs Position statement

On Fri, 01 May 2009, Manoj Srivastava wrote:
> If the statements are in contradiction of the foundation document
> (which is the case in a couple of prior situations), then are you
> saying that anything in the foundation documents can ve worked
> around by putting out a position statement, and have the developers
> proceed to ignore the foundation document on that basis?

No. I'm in fact saying that developers can ignore the position
statement on that basis.

> if that is not the case, what value does a position statement in
> contradiction of a foundation document mean?

Next to no value, as far as I'm concerned.

> How binding _are_ the foundation documents?

Only as binding as we as a group consider them to be.

Since the language they're written in is ambiguous, we can have
reasonable differences of opinion as to what the foundation documents
actually mean. A position statement about the foundation documents
only serves to state what a majority of the project thinks the
documents say; it doesn't change what the documents actually say.[1]

As such, people who think differently are free to ignore the position
statement in carrying out their duties (though they can of course be
overridden by GR.)

Don Armstrong
(Continue reading)

Manoj Srivastava | 2 May 01:54 2009
X-Face
Face
Picon

Re: Overriding vs Amending vs Position statement

On Fri, May 01 2009, Don Armstrong wrote:

> Only as binding as we as a group consider them to be.

        Hmm. Certainly puts the social contract in a new light, though.

> Since the language they're written in is ambiguous, we can have
> reasonable differences of opinion as to what the foundation documents
> actually mean. A position statement about the foundation documents
> only serves to state what a majority of the project thinks the
> documents say; it doesn't change what the documents actually say.[1]
>
> As such, people who think differently are free to ignore the position
> statement in carrying out their duties (though they can of course be
> overridden by GR.)

        I think I can live with that.

        Wait.

        Oh. So this is a way, via two simple majority GR's, for any
 majority to do an end run around the 3:1 constitutional requirements?
 nifty.

        manoj
--

-- 
Behind every great man, there is a woman -- urging him on. Harry Mudd,
"I, Mudd", stardate 4513.3
Manoj Srivastava <srivasta <at> debian.org> <http://www.debian.org/~srivasta/>  
1024D/BF24424C print 4966 F272 D093 B493 410B  924B 21BA DABB BF24 424C
(Continue reading)

Don Armstrong | 2 May 02:26 2009
Picon

Re: Overriding vs Amending vs Position statement

On Fri, 01 May 2009, Manoj Srivastava wrote:
> On Fri, May 01 2009, Don Armstrong wrote:
> > Only as binding as we as a group consider them to be.
> 
> Hmm. Certainly puts the social contract in a new light, though.

It really shouldn't; as a group we decide whether we're going to
uphold the social contract. There's no way to force the group to
uphold it. [Given the anguish with which we struggle on -project and
-vote to figure out what the SC says, it's seems clear that large
numbers of us feel that we should be upholding the SC.]

> > As such, people who think differently are free to ignore the
> > position statement in carrying out their duties (though they can
> > of course be overridden by GR.)
> 
> Oh. So this is a way, via two simple majority GR's, for any majority
> to do an end run around the 3:1 constitutional requirements? nifty.

Sure. If we as a project are headed towards self-destruction, there's
really no way for the constitution to stop us. We always have to fall
back on the continued desire of developers to work together to create
the most technically excellent, free operating system possible.

Don Armstrong

--

-- 
This message brought to you by weapons of mass destruction related
program activities, and the letter G.

(Continue reading)

Ben Finney | 2 May 07:33 2009
Picon

Re: Overriding vs Amending vs Position statement

Don Armstrong <don <at> debian.org> writes:

> On Fri, 01 May 2009, Manoj Srivastava wrote:
> > On Fri, May 01 2009, Don Armstrong wrote:
> > > Only as binding as we as a group consider them to be.
> > 
> > Hmm. Certainly puts the social contract in a new light, though.
> 
> It really shouldn't; as a group we decide whether we're going to
> uphold the social contract. There's no way to force the group to
> uphold it.

That doesn't mean we can't make the explicit expectation that everyone
in the group *will* uphold it, as a condition of being in the group.

I had thought that expectation was embodied in the requirement for all
new members to declare they will uphold it.

-- 
 \      “Software patents provide one more means of controlling access |
  `\      to information. They are the tool of choice for the internet |
_o__)                                     highwayman.” —Anthony Taylor |
Ben Finney

--

-- 
To UNSUBSCRIBE, email to debian-vote-REQUEST <at> lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster <at> lists.debian.org

Manoj Srivastava | 2 May 11:19 2009
X-Face
Face
Picon

Re: Overriding vs Amending vs Position statement

On Sat, May 02 2009, Ben Finney wrote:

> That doesn't mean we can't make the explicit expectation that everyone
> in the group *will* uphold it, as a condition of being in the group.
>
> I had thought that expectation was embodied in the requirement for all
> new members to declare they will uphold it.

        Such was my understanding as well.

        manoj
--

-- 
Cheit's Lament: If you help a friend in need, he is sure to remember
you-- the next time he's in need.
Manoj Srivastava <srivasta <at> debian.org> <http://www.debian.org/~srivasta/>  
1024D/BF24424C print 4966 F272 D093 B493 410B  924B 21BA DABB BF24 424C


Gmane