Re: bounties

[Joseph Wakeling]

On 06/20/2010 06:10 PM, David Kastrup wrote:
> People want a _solution_ to their problem, not new problems they never
> thought about and which are not actually in their personal problem
> space.

That's true, but it only shows that Lilypond isn't yet capable of
operating as a general-purpose best solution.  That's only a problem if
that's what Lilypond wants to _be_, or more precisely, what Lilypond
tries to sell itself as.  ('Wants to be' is fine as long as you have a
plan to get there and don't sell yourself as such prematurely...)

It's a bit like GNU/Linux a few years ago, and to an extent even now --
it wasn't possible to market it as a general-purpose operating system
suitable for all, because learning to use it involved an expenditure of
effort that only made sense if you had a deliberate motivation.  That
might be ideological/philosophical, it might be the opportunity for
hacking and customization, it might be that it provides better for your
particular technical needs, but whatever it was, you _needed_ that
self-motivation.

Lilypond's situation is almost exactly parallel, and like GNU/Linux, it
doesn't mean that you can't sell it -- it means you have to target your
sales pitch at the people in whom you can create that self-motivation.
Or rather, you have to target cases where the problem of learning to use
Lilypond is small compared to the benefit of having it.

The first such pitch can obviously be at those of us who already _have_
the motivation and are involved -- as Valentin is doing -- to create a
Lilypond 'subscription' model.  I'd go further than that and create
classes of sponsorship, the most prominent of which come with very overt
display (logos on the webpage and in documentation, etc.).  'Lilypond,
sponsored by ...' etc. etc.  Even if no one (for now) takes up the
higher classes of sponsorship, I think there are lots of us who would
happily offer regular monthly donations to Lilypond if there was a
well-defined funding (and spending) structure.

(I would suggest splitting any such initial funds between development
work and outreach -- spend some of the money on getting new features
implemented and some of it on pursuing a wider range of funding sources.
 Someone mentioned the hassle of writing grants and tracking
deliverables and project bureaucracy, but once you've _got_ a grant you
can and should dedicate part of that money towards such admin work.)

Above and beyond that, look at special or specialist needs that Lilypond
can fulfil and even more, at special or specialist needs where _further
development_ can fulfil them -- that gives an obvious reason to people
to fund Lilypond development.

> You can't stop at selling Lilypond to somebody without having an
> answer to "how do I start this thing"?  How many people are complaining
> that they double-click on the Lilypond icon and nothing happens?

Hmmm ... so why not start by having the icon open a 'Quickstart' help
page that explains how to use Lilypond on the platform in question?
Trivial to implement, no?

One other thing: yesterday a developer colleague of mine showed me the
following page,
http://tryhaskell.org/

... which he'd been inspired to implement by the equivalent 'Try Ruby'
page.  (There are now apparently a bunch of 'Try X' sites out there.)

Lilypond isn't exactly equivalent, since AFAICS it can't meaningfully be
run in any kind of 'scripting mode', but there is surely scope for some
kind of online interactive Lilypond environment (most likely a primitive
online editor + output display + tutorial).

I'll try and keep this in mind as a potential pet project (I don't have
the skill, but together he and I probably do).  There is no way we can
put time towards it in the immediate future, since we have strong
immediate priorities for the projects of our employer, but maybe a few
months down the line something can be done ...

Best wishes,

    -- Joe