developers developers developers

[Graham Percival]

On Thu, Nov 12, 2009 at 09:11:40AM +0100, Jan Nieuwenhuizen wrote:
> So what are the actual problems?  Is LilyPond really too difficult?
> Do we rely too much on crufty input-language solutions?  How many
> ritardando-like hacks do we have for common problems?  Are they
> listed/categorised somewhere?

For the record, **I have never recommended that somebody use
lilypond**.  When meeting a technically-oriented composer,
especially one working on algorithmic music, I might suggest that
they should check it out.  But I think the original poster was
entirely justified in switching back to Finale.

That's why I cringe a bit whenever I hear people proudly
announcing that they advertized lilypond to meeting X or
conference Y.


The worst two problems, and their solutions:
- input syntax standardization, which I've been planning for over
  a year.
- approximately 300 *known* bugs that produce garbage output that
  nobody's working on.  (there's about 10-15 bugs that people are
  working on)

There's no point listing more problems unless those (especially
the second) are addressed.

> Maybe it's just that we are not tempting enough to be used.

The new website is better than the old one; I don't think we need
any more advocacy than that.

> I cannot imagine that OP would have stopped with a frustating
> message to sibelius-user after having paid $$$$ for the program?

Oh, people do that all the time.


> How about a frogs project to make our image sexier?

How about a project to finish the website?  Oh wait, Kieren is
trying to work on that.

How about a project to make it easier to build lilypond docs?  Oh
wait, I'm kind-of doing that with waf.  If we stay with waf.

How about a project to make it easier to build lilypond, period?
Chicken and egg problem: we don't have enough contributors on
windows+osx to make it worth the effort to make it easier, and we
__might__ not get some contributors on those platforms because
they can't build it[1].

[1] I doubt this, though.  It's just a convenient excuse.

How about a project to gather more developers?  Oh wait, that's
GOP, which I've been trying to organize for slightly less than a
year.  We currently don't have the infrastructure for an influx of
inexperienced developers.  The snag right now: Carl has made the
easy-git-gui tcl script, but nobody's testing it.



Our problems are numerous, known, and mostly involve a lot of
work.  If we have double the amount of developer-effort, then we
would _start_ to make a dent in the list (instead of piling up
jobs).  We really could use 5-10 times the amount of
developer-effort if we're serious about it.

Cheers,
- Graham