Jari Williamsson wrote: "But why would Igor decide the rules, why couldn't the user define the rules?"
This goes to my deepest frustration with notation programs. Almost every time I have turned to developers at one or another of the software firms with a request for help to do something somewhat unconventional, the inevitable first response is either "Why would you ever want to do that?" or "The correct way to notate that is (fill in whatever the program does by default)?" There is an assumption that I neither understand the convention nor know best what I want to do, rather than a straightforward attempt to provide what I need. I recognize that many users of these programs — perhaps the vast majority — have less experience than I have and have less demanding requirements, and they honestly appreciate a program that assists at each step with achieving a conventional score and parts, be it making sure all the note values in a measure add up or that instrumental ranges don't exceed someone's definition of what's doable or not. But I do have the experience to recognize the elements of my notation which are conventional and those which are not; having to prove my chops each time I ask for something out of the ordinary — and I'm not a notational extremist at that — is frustrating. Particularly frustrating is when a feature request does not require significant re-engineering, but is rejected because of over-commitment to the conventions. For example, take non--powers-of-two denominators in time signatures. (Please do not begin a discussion here about what these are or whether one needs them, just believe me that there is a critical mass of composers who find them useful, if not necessary for their work, and a critical mass of players who have no great objection to playing them.) By going through a few loops — usually involving some combination of hidden real and visible graphic-only tuplets, or, alternatively, hidden tempo changes are graphic-only tuplets — it is possible both to notate these and get reasonably accurate playback, fitting them into a measure which is as close to the desired measure in length as the minimum Finale midi note value allows. In Finale, it is already possible to have a display-only time signature, with any number from 1-99 in the numerator (or some additive combination), but only powers-of-two are allowed in the denominator. I can understand having this limitation as a default for most users, and there are kludges of hiding time signatures are inserting a graphic signature but it really ought to be possible to stick any number in the bottom of the signature as an option. Developers I've spoken to have indicated that the code change is as minimal as I imagine it to be, so it's low-hanging fruit, but, as one developer put it to me: "the risk that an amateur user would accidentally enter 4/3 or 7/10 measures is greater to us than the benefit it would bring to the three users who want such weirdness." Oh well. One program with notational capacity, Harmony Assistant, has a nice feature of user-defined rules that would, in principle, go some distance to solving such problems, but is unfortunately underdeveloped in some notational aspects and cumbersome in input, but it does show that, in principle, such a path is possible. djw -- Dr. Daniel James Wolf composer [email protected] home.snafu.de/djwolf renewablemusic.blogspot.com MaterialPress.com _______________________________________________ Finale mailing list [email protected] http://lists.shsu.edu/mailman/listinfo/finale
