John Chambers wrote - >The difference is somewhat minimal, though. The original text had as >example "K:D =c", which implies a key signature of two sharps but >with a natural written before all the c's. The natural reaction to >this is "If all the c's are natural, why put a c sharp in the >signature and then cancel it everywhere?" This is indeed rather >silly. It's better to just use ^f as the key signature.
The difference seems real enough to me. If I specify an explicit key signature, I want it to appear at the beginning of the staff. Someone who wants global accidentals wants them scattered through the music. As far as I can see, global accidentals only make sense if you start from a key of C when they might make things clearer for learners by emphasising what a key signature means. Does anybody actually use them? I'd be happy to see them go. I think you suggested making it a run time option which seems reasonable. >OTOH, using >"K:Dmix=c" with ^f=c as the signature can be sensible, because that c >natural in the key signature instead of a sharp is an "advisory >accidental" that emphasizes the fact that the c is not sharp. I would have thought that this would confuse people unfamiliar with modes who will think that K:Dmix=c and K:Dmix are something different, especially since you also suggest things like K:Dphr^F where the ^F changes the key given by Dphr and implies that the mode is phrygian. K:_b_e^f tonic=D mode=freygish would be much clearer. >If we were designing abc from scratch, I'd agree. This is where I came in three (?) years ago, by using this as an example of how we were stuck with an unsatisfactory system when a little more thought and cooperation could have come up with a better way. >ABC is compact and cryptic, but easy to type. but is it easy to understand? There are more people trying to read it than type it. Bryan Creer To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html