Mission statement and 1.1 license issues

[John Bradford <john@xxxxxxxxxxxx>]

Quote from Dr Andrew C Aitchison (A.C.Aitchison@xxxxxxxxxxxxxxx):
> > Maybe the real issue is in fact with the mission statement of the
> > XFree86 project, rather than it's license.
> >
> > The mission statement of the XFree86 project says nothing directly
> > about serving the existing user-base of the XFree86 project.
>
> I wonder how developers would feel if the XFree86 project mission
> statement was changed to one of "service"?

I didn't suggest changing it to that.  Adding it as a kind of
secondary consideration would hardly be a drastic move.

> I for one wouldn't have volunteered if that had been the mission.

Then maybe the project would have been better off without your
contributions anyway, because if what you contributed is only really
useful in the context of XFree86, then it can effectively be hijacked
and made useless to any users relying on it, by any action which makes
the whole project become useless to those users.  I am assuming here
that the average user isn't capable of forking the codebase, so
security vulnerabilities which are only fixed in versions with
different license agreements could make the project as a whole useless
to some users who have previously relied on it.

> > In practice, I think that this is an unwritten goal of many, if not
> > most, free, open source projects.  For example, note the generally
> > strong reluctance to break anything that is actively used by even a
> > small percentage of the user-base, in Linux kernel development.
> >
> > I have always assumed that the XFree86 project would adopt a similar
> > 'unwritten rule'.  Note that the mission statement does specify some
> > non-goals, and providing continuity for the existing user-base is not
> > one of those non-goals.
>
> We did drop support for a large number of cards (mostly based on S3
> chips) when we moved from 3.3.x to 4.0, so in the absence of a statement,
> our history might have been suggestive.

Fair point, although I assumed that it was mostly a technical
decision, rather than a political one.

> > For example, in my opinion, it would be a good service to the existing
> > user-base to go through and audit all the code, and either remove and
> > replace code which is under very restrictive licenses, or ask the
> > copyright holders to re-license it.
> >
> > Of course, this doesn't fit in with the stated goals in the mission
> > statement.
> 
> Maybe it is just not a very interesting task for current developers ?

Everybody's tastes differ.  I think it could be quite interesting for
anybody who is interested in the project, but perhaps isn't as skilled
in writing code as they need to be to contribute in other areas.

> Have you asked whether such work would be accepted it submitted ?

Not directly.

> > If the mission statement was not followed so aggressively, what it says
> > might not be as important, but - in my opinion unlike many, or even most
> > other free, open source projects - the XFree86 project seem to stick
> > VERY closely to their mission statement, even when it appears to be
> > detrimental to many of the existing user-base.
>
> For its size and complexity, XFree86 has a very small active
> developer pool.
> My impression is that it is the other way around:
> the mission statement was written to match the interests of those
> who do the work, and it only *appears* that developers and management
> stick rigidly to the mission statement ?

So why is there a general lack of interest in even discussing the
reasoning behind the license change?  Why has the mission statement
referred to in reference to the license change, if changing it would
be so easy, (because of the low number of developers)?

> > I urge users of XFree86 to read the mission statement, and decide
> > whether you want to rely on this project or not.
>
> I'm happy with that, but remember than Richard Stallman proposed that
> software development be free and software support should cost,
> since developers get their kicks from writing new interesting code,
> but only support programs if given a more tangible incentive.
> 
> Maybe the user has to chose between a free non-"service" project
> driven by the interested of the developers,
> and paying for some other, X server product, more directed at their needs.

You have completely mis-interpreted what I meant by 'serving the
existing user-base'.  Maybe it was a bad choice of words.

I am not asking the developers to put any special effort in to any
kind of 'support', _at all_.

However, if the XFree86 project is going to follow it's mission
statement to the letter, it seems to me, (and correct me if I'm
wrong), that a decision which would directly and significantly hurt
the existing userbase, but has even the slightest technical advantage,
and doesn't conflict with anything else, would be made without a
second thought.

That is _very_ unusual for a free, open source project.

All free, open source projects 'serve' their existing userbase to some
degree, simply by their continued existance, and it is only this
'implied' servitude that I was suggesting being spelt out in the
mission statement, (to which adherence seems to be unusually strict in
the XFree86 project).  An alternative could be to simply adopt it
anyway, without actually writing it down, but judging from the
discussions on this mailing list, and the frequency with which the
mission statement is referenced, I think this solution would be
unwelcome.

I don't believe that the above definition of 'serving the existing
user-base', is at all similar to 'software support' in the traditional
sense, but I have cc'ed Richard Stallman for his opinion on this.

John.
_______________________________________________
Forum mailing list
Forum@xxxxxxxxxxx
http://XFree86.Org/mailman/listinfo/forum