On Fri, Feb 10, 2012 at 01:30:12PM +0000, Graham Percival wrote: > 2) set up a kickstarter to "perfect" that score, where "perfect" > means "work on lilypond such that good output is produced with > only semantic information". No tweaks, no workarounds, etc.
Naturally, once a particular score (+ lilypond functionality) has been "perfected", it would be folly not to add that score to some sort of regression tests. The first kickstarter should be padded to include time+effort required to add multi-page output to the regtests -- quite possibly adding a separate target for make largetests so that normal development with "make test" are not unduly slowed down by having those large scores. I know that lots of people will want lots of fancy stuff like lyrics, tablature, orchestral scores, contemporary notation, etc -- but I really, really, strongly suggest that you aim to have a turn-over of a month at most in the beginning, which means keeping the first few scores being fairly simple and small. It's always easy to have more complicated scores later on. - Graham _______________________________________________ lilypond-user mailing list lilypond-user@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-user