[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Bug#353277: Please reject to rule on the ndiswrapper question



On Wed, 01 Mar 2006, Steve Langasek wrote:
> On Wed, Mar 01, 2006 at 01:03:56AM -0300, Henrique de Moraes Holschuh wrote:
> > Otherwise, the ctte could overrule just about everything in Debian.  Were
> > they not bound by the SC themselves, they could overrule even the SC itself
> > by determining that the files that determine in which suite a package go
> > should make all packages in the non-free suite go into the main suite.
> 
> I wonder why you think it's *not* the intention of the constitution that the
> technical committee be in a position to overrule "just about everything in
> Debian".  The exact phrasing of the constitution is:

*IMHO* the constitution doesn't feel like that was the intention, and it
doesn't read like that either.  So, please excuse me if I will not read into
the constitution overlord powers for any body (ctte included) unless it is
explicitly written in there.

Of course, I can be convinced that the constitution does give the ctte that
power, but so far, I am not.   Otherwise, why didn't we pose to the ctte a
request for how the GFDL issue should be handled?

>   4. Overrule a Developer (requires a 3:1 majority).
> 
>   The Technical Committee may ask a Developer to take a particular technical
>   course of action even if the Developer does not wish to; this requires a
>   3:1 majority. For example, the Committee may determine that a complaint
>   made by the submitter of a bug is justified and that the submitter's
>   proposed solution should be implemented.

Overruling a Developer is so far from "being able to overrule just about
everything in Debian" that it ain't funny.  The example given is also of a
Debian Developer in typical package maintenance function, and not of the
Secretary, ftp-masters, or other delegates (and de-facto delegates) of the
DPL when acting on their roles.

To me, it is obvious that the ctte can resolve a dispute with the
ftp-masters when the interpretation of the DFSG, SC, a GR or the
constitution is not the object of the dispute.

> Nowhere do I see anything that says the committee must limit itself to
> requiring a developer to take a particular technical course of action *that
> agrees with the developer's pre-existing political views on the issue*.  I

Who said anything about the developer's pre-existing views? of course the
ctte can't be bound by that, it wouldn't be able to do its job otherwise.

I *specifically* talked about project-wide policy, as in that set by GRs,
the SC, the DFSG, the constitution and whatever else we have that sets
policy (and I am *NOT* talking about the Debian policy document, btw).

> only appelate body defined in the constitution, I don't believe it was ever
> intended that their authority could be vetoed by claiming that a particular
> technical decision was made on religious grounds.

Constitution section 6.1.3 makes it clear that if the right person appeals,
the ctte is brought into play without any possibility of other parts
complaining (well, of their complaints affecting the appeal, anyway).

You can appeal to the ctte always, but it doesn't have powers to *enforce*
its decision unless its under 6.1.1, 6.1.2, 6.1.3 and 6.1.4 sections of the
constitution.  AFAIK the ftp-masters didn't appeal, so 6.1.3 is out (if they
did appeal, then the question is indeed in the ctte enforcement sphere
because of that).  6.1.2 is out as this is not a jurisdiction question.

6.1.4 could be worded much better. Does that apply to any registered member
of the Debian project (that includes everyone, DPL and Secretary, DPL
delegates and common Developers)?  Or does it apply to the Developers as
defined by 2, which clearly separates the DPL, Secretary, DPL delegates, the
CTTE, a Developer doing an unamed task of his and the body of Developers?

The way it is written, IMHO 6.1.4 is directed to a single Developer doing
one of the usual Developer things (i.e. packaging, bug-fixing, etc), as
opposed to delegates and other specialized tasks, or the entire body of
Developers.

As for 6.1.1, IMHO the examples and the emphasis on technical make it quite
clear that stuff like interpreting the constitution and DFSG is not in
6.1.1's scope.  I think this is uncontroversial, at least Ian Jackson seems
to hold a similar position about 6.1.1, in
http://lists.debian.org/debian-ctte/2004/06/msg00002.html (last paragraph).

> the constitution invests in the TC, and you kinda have to trust that we
> won't abuse it.

I do trust the ctte not to go around abusing its power. I would trust the
ctte to not take extreme decisions that would split the project even when
within its power as well (instead, I expect the ctte would call for an GR on
the matter).

But I still don't think the ctte has power to overlord about everything in
Debian, unless you mean that it can be *requested* to do so by a body that
has the power to address a given issue already, through 6.1.3.

> > > The question we have been asked to consider is, "which section should the
> > > ndiswrapper package list in its control file?"  This is a technical
> 
> > The answer to that question is: the one policy determines it to.  The ctte
> > can not say much more than that, packages are not placed into a *suite*
> > (main or contrib) because of any sort of technical concern.
> 
> Policy claims to be a technical document.

I did mean policy as policy, not the "Debian policy document".  Policy in
Debian is set by a lot of things, the foremost being the constitution and
the social contract.

The technical Debian policy document is clearly inside the ctte sphere.

If you can convince me that placement of a package in contrib or main is
purely a matter of interpretation of the Debian policy document (which is
technical), then yes, I will be convinced that the ctte can rule the
ndiswrapper placement, even if ftp-masters never asked it to.

-- 
  "One disk to rule them all, One disk to find them. One disk to bring
  them all and in the darkness grind them. In the Land of Redmond
  where the shadows lie." -- The Silicon Valley Tarot
  Henrique Holschuh



Reply to: