[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: MusicXML exporter (was Re: Lilypond lobbying?)

From: Kieren MacMillan
Subject: Re: MusicXML exporter (was Re: Lilypond lobbying?)
Date: Wed, 24 Aug 2011 20:00:09 -0400

Hi Carl,

> Do you want
> 1) XML that captures only the music (and could be imported into some other
> program which will make the layout decisions)?

No: this is trivial to obtain from #2 or #3, via XSLT.

> 2) XML that captures both the music and the layout (and could therefore be
> printed by some as-yet-unknown MusicXML printer)?

Yes: since Lilypond already generates this entire set of data, why not include 
it in the output?

> My sense is that item 1) is relatively inexpensive (as Jan has discussed),
> but that item 2) is relatively expensive (I think it's more than 100 expert
> hours, but that's just a wild stab).

Maybe true… but that's what we should be going for, IMO.

> For me, item 1) is what we ought to be aiming at, at least initially.

My question is this: In what format is the final, typeset music stream such 
that extracting the music information only would be massively easier than 
extracting the music and layout information?

> It seems strange to use Finale to print a layout defined by LilyPond.  If you
> want to use a LilyPond layout and tweak a few things graphically, you should
> be using the svg output, IMO, and editing the svg.

I think there are many things "in the cracks" that don't come through with just 
"the music", but would be critical for translation to another program — for 
example, the cross-staff information that started this thread is clearly layout 
related, and not just music-specific.

> I think that holding out for 2) will probably delay completion of 1).

Possibly… If we're talking about 10 hours for #1 (truly well done) and 500 
hours for #2, then of course we should do that. If we're talking about 100 
hours for #1, and 250 hours for #2 where the first 100 hours must be redone 
(assuming, for argument's sake, the two are radically different in execution), 
then I would say no.

> But having a well-defined enhancement request will at least allow a
> developer to make a decision as to what they wish to attempt.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]