Braille

[Ralph Little]

Hi,I see that there is some renewed activity in Lilypond dev and that's always 
nice to see.I would like to drop something entirely different into the pot.
I have resurrected my Braille music output project, which was originated around 
12 years ago and has been picked up and dropped at various times, mainly due to 
family issues and emigration.
I've been beavering away it it for the past few months and I have a rough 
prototype, probably sort of a proof of concept really, to fully explore the 
issues. I have ended up with a set of C++ classes in a separate test bed, not 
integrated fully with Lilypond as yet and I can produce proper music Braille in 
bar-over-bar and single line format. The original  intention was to come up 
with an API that could reside in a separate library linked to Lilypond thereby 
providing a resource that others could use. I have now arrived at the point 
where I need to make some decisions.
Into my dev branch, dev/rlittle I have the beginnings of a framework based 
heavily on the midi implementation. Indeed it is composed of the the midi 
classes replicated and stripped down, into which I intended to seed calls to 
these classes to generate the Braille output.

I would like to elicit some opinion from the group as to where I should take 
the project from here.
I would be interested in views on any aspect of the project, including but not 
limited to whether or not it should be an integral part of Lilypond or perhaps 
entirely separate, should it involve less C++ and more Guile, is the midi 
framework a suitable vehicle?
I know that there have been suggestions that Braille could be generated from 
MusicXML and I know some people have started (and apparently abandoned) that 
effort and that is a valid option.
For me, I would like to see some integration with Lilypond. One the one hand, 
Braille requires curation probably through Lilypond macros to direct line 
breaking, preferences for certain encoding options and the like, as we do 
currently for traditional output. The formatting challenges are very different 
from traditional scoring but some editorial input is required in some areas for 
aesthetic and reading efficiency concerns. Secondly, I would eventually like to 
see Braille output as an option on Mutopia as midi is now.

RegardsRalph