Re: Sponsoring lilypond development Was Re: Score parts: instrument and duration

[Trevor Baca]

On 8/17/05, Pedro Kröger <address@hidden> wrote:
> Johannes Schindelin <address@hidden> writes:
> 
> > AFAIK SCORE was developed by a single person, in FORTRAN,
> 
> ha! that's something.

It's true: SCORE is the work of Leland Smith, now professor (emeritus)
of music at Stanford. SCORE was originally FORTRAN when Leland started
developing on it the late *70s* (!) and I think that the current 4.x
version is *still* FORTRAN (running, however, only under DOS).

IMO, some of the most sophisticated engravers working today are using
SCORE and I think the results show in the fact that something like 4
or 5 of the 6 Revere awards a year have recently gone to SCORE
engravers (although I'm quoting from memory here, so doublecheck
before passing that on).

I think that stealing (or courting?) the SCORE users is an excellent
idea; they're not in the slightest bit adverse to really learning the
most detailed internals of a program and would probably bring an
excellent eye to some of the very real problems we're tackling right
now in LilyPond development: the tie problem is only the most recent
example.

> 
> > Also, he was not very forthcoming with information about internals.
> 
> yeah, I didn't think so.
> 
> > If my memory serves me, there is a SCORE list.
> 
> yep, here is:
> 
> http://ace.acadiau.ca/score/s-list.htm

Just mail to address@hidden ; the list is run by Gordon Callon in
Canada. Quite unfortunately there are no list archives :-( but, quite
happily, the list is responsive, open and professional (much like our
own, really).

> > We could steal the users directly from there... But for this to
> > happen, we would have to reverse engineer the SCORE format.
> 
> hum, it's a binary format right? I've forgotten that.

No, no, not to worry: the SCORE native format .mus (not the same as
the Finale .mus) *is* binary ... but there's this magic cleartext
format ending in .pmx. From the SCORE UI you can turn *any* file into
.pmx and read, parse, tweak to your heart's content; many SCORE users
do it all the time and the documentation is complete.

Once you've parsed / tweaked your .pmx, you can read the modified file
back into SCORE directly. This .pmx format gives the ability to write
extremely sophisticated editing macros (or tweak algorithmically, if
you like).

> I just checked this webpage:
> 
> http://www.scoremus.com/score.html
> 
> They have it now for $200-360 (the old price was ridiculous)

No, the price is still ridiculous: it's $750 / license (you must be
quoting the price for one of the ancillary programs).

I'm by no means an expert SCORE user, but I have used the program on
and off for the last 2 years. It's an excellent program for many of
*exactly* the same reasons that Lily is an excellent program: you have
real control -- not fake, Sibelius-style control -- over anything you
want; you have a cleartext input format; you have outstanding
postscript output; you have an incredibly intelligent user community
of expert engravers, composers and instrumentalists to support you.

The primary downpoints to SCORE are all quite apparent:

1. SCORE runs *only* under DOS (though it works fine in emulation
under whatever Windows you want, and it works fine in Windows
emulation under OS X)

2. SCORE has unbelievably harsh memory- and filesize restrictions; if
you go beyond a certain number of postscript vectors per file, SCORE
blows an error; the practical upshot of this is that I never have more
than about 8 or 10 staves in a single file; if I'm working on an
orchestral SCORE, well, then, it's 3 or 4 *files* per single system
(muchless page) of music

3. SCORE is developed by a genius ... but that genius is 80 years old
(in fact there's been a minor birthday celebration going on for Leland
on the SCORE list) and the rate of release is absolutely glacial; I'm
not criticizing Leland ... as far as I'm concerned the work he put
into SCORE and the state of refinement the program reached justifies
absolutely anything he wants to do, developmentwise or other. However,
the fact remains that the development cycle has, imo, essentially
ended and, more severely, that the code is utterly closed; no one has
access to the sources and so there's a grave concern about what
happens to the tool in the eventual future :-(

* * *

So, the obvious point here is that LilyPond absolutely excels at all
three of SCORE's major weaknesses:

1. LilyPond runs on absolutely everything under the sun

2. LilyPond has absolutely no memory or size limitations at all,
AFAICT; I render Lily scores with several thousands glyphs per page on
tabloid paper all the time and nothing ever breaks (well ... so long
as you turn off point-and-click!); there's absolutely *no*
circumstance under which Lily users have to separate out musical
material into separate files to accomodate an extramusical technical
requirement

3. Lily is also developed by a genius ... but a genius pursuing the
exact *opposite* political approach to code access and the development
cycle: Lily's not going anywhere, the sources are all available
forever, new releases show up constantly, and the only barrier to
contribution is the learning curve

So it should be clear by now where I stand: I think it's an obvious
transition for SCORE users to start moving over the Lily. They're
reluctant to give up the tool, of course, because of its
sophistication, but I think that literally every 2 or 3 weeks the
reasons to jump to Lily increase, even for the community of
professionals currently working in SCORE.

* * *

I considered crossposting this response to the score list (again,
address@hidden, if anyone's interested) but thought it might be more
appropriate to let discussion run private to Lily for a while longer.

Another reflection: I feel certain that some type of constructive,
even semi-formal dialogue is possible with the SCORE community. The
wealth of accumulated knowledge there is almost unbelievable: if
you're looking for exact answers to questions of tie placement, slur
placement, hairpin positioning on the horizontal, accidental
tesselation before complex chords, the SCORE people are a good place
to talk. I don't have the time to moderate that discussion right now
myself, but I have to imagine that there should be some way to welcome
that knowledge into Lily and help our own growth here.

Trevor.