On Mon, 2016-03-14 at 12:11 +0100, Menu Jacques wrote: > Hello Richard, > > Thanks for your answer. > > If I understand well, backup and forward MusicXML elements change the current > 'position' in a voice along the way. > > Why is it difficult to handle that properly?
because it depends on what that means and it depends on people who write musicXML exporters having the same idea about what that means. A Backup element contains a "duration" element and it documents this immediately below as: "Duration is a positive number specified in division units. This is the intended duration vs. notated duration (for instance, swing eighths vs. even eighths, or differences in dotted notes in Baroque-era music). Differences in duration specific to an interpretation or performance should use the note element's attack and release attributes." which obviously has nothing to do with the "backup" description. I think the underlying problem is that musicXML has been created not for the benefit of free software but to help promote a specific commercial program. Characteristic of commercial products is that they should above all look very professional, and sound like the right thing. Unfortunately the free alternative (MEI I think it's called) has been designed to be written by hand rather than by machine - the creators explain that there are many different ways of describing the same thing in MEI, which is good for humans, but bad for machines. LilyPond is in that class, allowing you to structure your file to reflect your music rather than providing some rigid framework. Richard > > JM > > > Le 14 mars 2016 à 11:28, Richard Shann <rich...@rshann.plus.com> a écrit : > > > > On Mon, 2016-03-14 at 11:08 +0100, Menu Jacques wrote: > >> Hello, > >> > >>> Le 6 mars 2016 à 16:07, Richard Shann <rich...@rshann.plus.com> a écrit : > >>> > >>> It indeed should be able, but MusicXML isn't well adapted to the task: > >>> note durations are indicated in multiple ways and there is a timer that > >>> can move backwards to indicate notes that are to be sounded at the same > >>> time as other notes and the resulting spaghetti can be very difficult to > >>> interpret. The result can be nothing read at all, or (as in the case > >>> with Denemo's import at least) empty staffs to be deleted, redundant > >>> time signatures to be deleted, or worse, time signatures to be guessed > >>> and inserted. > >> > >> I’ve tried to find information about the MusicXML timer mentioned by > >> Richard, but no way. > >> > >> Can someone indicate where to find some? > > > > I think it's this > > > > http://www.musicxml.com/UserManuals/MusicXML/MusicXML.htm#CT-MusicXML-backup.htm > > > > Description > > > > The backup and forward elements are required to coordinate multiple > > voices in one part, including music on multiple staves. The backup type > > is generally used to move between voices and staves. Thus the backup > > element does not include voice or staff elements. Duration values should > > always be positive, and should not cross measure boundaries or > > mid-measure changes in the divisions value. > > > > although the description is typically vague and I'm not sure it > > corresponds much with what people do. The documentation seems to have > > changed a lot since I last saw it, it appears to be machine generated > > for the most part, making it look very comprehensive; unfortunately > > machines don't care whether it means anything. > > > > Richard > > > > > > > > _______________________________________________ > > lilypond-user mailing list > > lilypond-user@gnu.org > > https://lists.gnu.org/mailman/listinfo/lilypond-user > _______________________________________________ lilypond-user mailing list lilypond-user@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-user
