Re: how to make decisions?

[David Kastrup]

Graham Percival <address@hidden> writes:

> On Mon, Sep 03, 2012 at 01:24:22AM -0300, Han-Wen Nienhuys wrote:
>> On Sat, Sep 1, 2012 at 4:39 PM, Graham Percival
>> <address@hidden> wrote:
>> 
>> > The meta-target is "after spending 5 years very publicly
>> > telling people *not* to talk about changing the syntax because
>> > we would do so 'in a year or two', I think I should encourage
>> > such discussions.".  I mean, people trusted me when I said
>> > that there
> -snip- 
>> Whatever you have been saying in the past is irrelevant.
>
> Yes, I'm definitely getting that impression.

For what it is worth, I disagree with Han-Wen here.  There is sadly a
grain of truth to it, but it is not the whole story, and it most
certainly is not the moral of the story.

> As you may recall, I launched a proposal to discuss GLISS:
> http://lists.gnu.org/archive/html/lilypond-devel/2012-07/msg00639.html
> You replied with a single email:
> http://lists.gnu.org/archive/html/lilypond-devel/2012-07/msg00786.html
> Discussion continued: "there seems to be fairly broad support for
> _some_ form of standardization":
> http://lists.gnu.org/archive/html/lilypond-devel/2012-07/msg00874.html
> The final proposal was posted:
> http://lists.gnu.org/archive/html/lilypond-devel/2012-08/msg00191.html
> with no substantial disagreement.
>
> That proposal became:
> http://lilypond.org/~graham/gop/gop_4.html
>
>
> I don't know where to go from here.  I spend a lot of effort
> trying to organize such discussions, because I think that LilyPond
> is a community project.  I think that we should encourage people
> to participate, but telling people "ok, thanks for your work on
> XYZ, now get lost while the real developers talk about ABC" might
> discourage people from working.

This is really a hard problem.  What makes up the community and a sense
of community, is a feeling of empowerment and mutual respect.  LilyPond
gives power to musicians to express music.  And this music can be passed
on and shared, and one can discuss the ways to create it, and exchange
experience.  And the developers are responsible for giving the users
that sense of empowerment.  But the actual power the users can wield
over developers sensibly is rather undirected, in the way the actual
power art critics can wield over the work an artist creates.

If a customer says "I like it" or "I don't like it", this is sort of a
feedback.  If a customer says "you should use the brush X more often" or
"I think the perspective is off in this corner" or "let me tell you how
I think you should paint the sky", the potential for an actual
improvement is rather limited.  This does not imply a lack of respect
for the critic's or customer's opinion.  It is just that the problem
space has a lot of its own inner logic that is not always easy to
represent.

> The idea behind having an explicit policy proposal, following by at
> least two updates of the proposal, is to allow everybody to see just
> what is being considered and to have lots of time to make objections.
> If some people ignore the later proposals (especially the one marked
> "final") and then object a month later, the whole process becomes a
> total waste of time.
>
> Granted, some people might see this as a good thing.

I have the serious fear that this will either lead to a sham process
and/or result in future problems.  And I definitely admit that it would
be mostly a relief to me if the whole process turned out a total waste
of time, and it would not be honest of me to encourage decision
structures that I see in conflict with the manner in which I strive to
arrive at the best results I consider to be in public interest.

I don't have a good answer here, and I am not particularly happy with
suggesting that the work I end up doing will not likely be shaped much
by committee or community decisions but rather mostly by my own
conscience and programmer instincts.  Which, in turn, are shaped by the
perceived needs of the community.

If you take a look at how, for example, religious communities work by
and large, you'll find that most people are guided by their own
conscience, but still feel being a part of a community, even though
there are no democratic decision processes about what aspect of a deity
to believe in.

> That's what I want to find out now.  Is there any point to having
> policy discussions in the form of GOP ?

I think they have their place, but will probably not work for governing
everything.

> One of the neat things about Waltrop was seeing a range of
> developers and users.  Some people were very concerned about
> Baroque music; others preferred pushing the boundaries of paper
> notation (and even going beyond static notation into video!).
> Some people ate meat, some people drank alcohol, some people were
> devoutly religious, some people were atheists.  Some people
> organized crowd-sourcing of a book of vocal scores; some people
> write algorithmic music with "write-once, read-never" ly files.
> Some people wake up early, some sleep in.  Some people work on
> converters to/from ly and musicxml, some people worked on
> online/realtime score generation.
>
> The LilyPond community has some shared values, some opposing
> values, but an overall interest in the specific code base called
> "lilypond".  How can we work on that together, while respecting
> our individual differences?  How can we encourage and keep
> ourselves motivated to improve LilyPond?

Mutual trust and respect, communication of one's needs, working on the
same or related overall goals.  Policies have their place in that
overall scheme, and also make some things easier to get done with.  With
regard to _designing_ the language of LilyPond, I think we are better
off discussing and relating our goals, and seeing where this takes
development.

Yes, this requires trust, but policies can't replace that trust for all
that they are worth.  They are more a reminder for everybody where we
want to be heading, and how we think we may arrive there.

-- 
David Kastrup