Re: short Musikmesse minutes

[Urs Liska]

Am 20.03.2014 00:18, schrieb Kieren MacMillan:
> Hi Urs,
>
> > > Time for a Lilypond Publishing House...
> > I'd say rather push LilyPond into the existing publishing houses.
>
> Why not both?

Nothing against it. But actually there _are_ already a number of 
"LilyPond Publishing Houses" - all of them of neglectable market impact.

> > And I think changes have never been better for that than now.
>
> True.
>
> In my opinion, here — in order of importance — are the things we need to make 
> established houses sit up and take notice:
>
> 1. Flawless MusicXML import and export.
> 2. Better “pixel-level” control of objects.
> 3. A finely-tuned stylesheet system.
> 4. Excellent, “turnkey" edition control features.
> 5. Several examples of fairly complex engravings, presented with “best 
> practice” coding.
>
> #4 is nearly in place, thanks to Jan-Peter.

Indeed. I hope it will be possible to make that generally usable (be it 
through inclusion in LilyPond itself or a easily usable place in one of 
the libraries.
Having this would be a valuable additional selling point.

> #3 would take, in my estimation, about 10 person-hours to prepare an exemplar 
> set of stylesheets (I’d be happy to do this myself), and 5-10 hours to create 
> an appropriate stylesheet-loading function.

I'm not so sure if this would really matter that much in terms of 
"market penetration".
Which isn't to say that I'd find that highly desirable myself.

> #5 could be put together in “no time”.

At the messe I had a compilation of samples (explicitly including 
non-publication quality "default" engravings) with me (including your 
Beethoven BTW), and this proved very useful.

> #2 would require some fundamental changes to Lilypond — likely this is the 
> largest hurdle.

Apart from pixel-level it will be a tremenduous step if we manage to get 
a certain kind of graphical approach to that, as is currently being 
worked on in Frescobaldi.
Graphically editing things while retaining strict control over the 
source code will be a killer feature.
So I'd add that as a separate item to your list.

> #1 is the next largest hurdle.

Yes, and a crucial one. But I think current development is very 
promising. For the first time someone is actually working on it. 
Although it is only a first step this is really a solid foundation. 
(Although only visible when using Frescobaldi from its Git repository.

> Thoughts?

One thing that isn't in your list - and which already is there - is 
everything around the power of version controlled workflows. This 
provides solutions to actual problems people have. At least this was my 
experience in Frankfurt. Everybody seems to know about the hassles one 
has with concurrent revisions of files when having to pass documents 
around. Meticulous project documentation, encapsulation in branches etc. 
were keywords editors could grasp immediately.
Question is how they will react when they see actual LilyPond code. I'm 
looking forward to that (there are two publishers I'll visit for closer 
demonstrations, with two more I have hopes to get to that point too).

Urs

> Kieren.
> _______________________________________________
> lilypond-user mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/lilypond-user