As a musician, I agree with the recent comment that musicians won't use Lily in her current state. If I didn't have programming experience (although my expertise is 20 years out of date), I wouldn't be able to myself. However, unlike the recent poster, I do have a (I hope) useful suggestion - integrate Lily with the many WYSIWYG MIDI sequencers that are available, via mi2mu. Musicians understand and read score notation far better than numbers or Lily-like codes. (I learned to read music well before I could read English - that's not unusual.) Any keyboardist can input music using a MIDI keyboard far faster and more accurately than with a typewriter keyboard - in my case, more than 10 times as fast when input error correction is included. Hit the sequencer quantize function then legato and you have a workable input to mi2mu within seconds. Musicians will rely on audio output for proofreading, as I do. It is very fast to drag a note from one staff to another, kick the sequencer back a bar, then restart the audio output to check that the correction has been done right. To remove a note from a Lily voice and insert it in another once such an error is identified is extremely slow and indirect, as is the checking that the correction is correct. Many musical issues are dealt with very efficiently by standard score sequencers, especially by the ones that are designed for musicians as Cubase is - quantizing of note starts and durations, transposition, time scaling and voicing in particular. It would take a huge amount of work to make Lily as accomplished in these areas as sequencers already are - she won't have to do any of it if MIDI files are integrated into her. It would take much less work to improve mi2mu the little bit required so that all but the most minor corrections could be best done by returning to the sequencer, then through mi2mu to Lily again. Hope this comes across as positively as I want it to... John http://www.ncf.carleton.ca/~bf250/harpsichord.html
