On 9 Jun 2003 at 17:09, Christopher BJ Smith wrote: > At 3:30 PM -0400 6/09/03, David W. Fenton wrote: > >On 9 Jun 2003 at 12:31, Christopher BJ Smith wrote: > > > >> Is there any reason why chords HAVE to be attached > >> to items in the measure? Why can't they just be attached to beats? > > > >There's a perfect example of something that derives from the type of > >file format. > > > >In a relational database, records of different types are related > >hierarchically. A record that is subsidiary cannot exist without a > >parent. > > > >In Finale, the frame is the highest-level record that matters. > >Beneath that are the entries in that frame. Then the articulations > >and expressions are attached to those entries. > > > >So, without an entry, there is no parent for the chord. > > Kind of like staff expressions? But those can be attached to an item > OR to the measure, at a horizontal and vercial distance that you can > specify. Why not chords? . . .
Because staff expressions are not precisely aligned with the rhythm. It's one of the reasons I don't like them -- I never know what I'm going to get. > . . . In fact, I was suggesting ONLY > measure-attached chords, but I see now that having the ability to > attach them to items AND beats is WAY better! It's the only way that it would be useful, don't you think? > >Now, the file structure could have been designed differently, yes. > >But then the chords would be attached to the measure, and positioned > >somehow with a snap-to-grid functionality, where the grid would be > >rhythmic, at the appropriate level of granularity. Otherwise, you'd > >get chords occuring at random locations. > > Right. And let us specify quarter notes, half notes, or even eighth > notes or smaller to snap to. I don't yet see any problems. I do. There is currently no such structure in Finale for specifying exact placement. What would happen would be some kind of translation of the visual placement into a precise realization in playback. Expressions now work by applying, I believe, to the next item following them (i.e., a fff just after the beginning of a half note does not change the velocity until the next note is initiated). Maybe that would be satisfactory for chords? I think of what kinds of things I've done with figured bass and don't really think it would be good enough, as you could easily want multiple chord changes over a single bass note. So, Coda would need to add some new data storage structure (file format change!) in order to record the information that is being captured with the rhythmic grid. > >Would the ability to attach chords to places in the measure be worth > >navigating the complexities of the UI for a snap-to-grid > >functionality? I don't know. > > What would get more complicated in the UI? It is much more complex to > enter items in another layer and hide them, making sure that the > option is selected to allow things attached to hidden items to show, > then attach a chord to the hidden item. I can't imagine anything MORE > complex than that! Well, consider: Should the grid be for the whole measure? Or beat-by-beat? If the former, what about a French overture style measure in which you have a dotted half tied to a dotted 8th/16th, in the right hand, and a pedal note in the left hand. You could perhaps want a chord change on that last 16th note, so you'd need to have a 16th-note grid. OK, not bad. But what if you also needed a chord change at the half measure? Then you'd need to spread the measure out so that the 16ths were far enough apart that you could be sure to hit the right place on the grid. Now, one possible UI would be to do something like the spacing handles for manual spacing, where you doubleclick between existing handles to get additional ones. If you could then pin the chord to the new handle, that would be perfect. What do you think? It has the advantage of sharing something with a familiar UI that already exists, while adding a different kind of functionality. Nonetheless, that data has to be stored somewhere in a manner that guarantees that the playback will be exact. I'm not sure if the current file format could handle that. > > But that is really what it would > >require, while the same thing is enabled in a rhythmically > >unambiguous manner by, shock of all shocks, simply notating the > >rhtythm you need for your chord chanes and attaching your chords to > >that rhythm. > > > >To me, the Finale way is more musical, even if it does cause > >complications and more work. The reason the Finale way is more > >musical is because the things you are attaching really are rhythmic > >and not visual, so it's critical that there be no ambiguity about > >where the chords are placed. > > I agree about the importance of precise placement. I hope it can be > easily implemented in a grid-type interface. I think the spacing handles would be a great way to do it. Say you placed your chords in the measure, then did something to reveal the spacing handles. You doubleclick to get appropriate subdivisions, then drag the handles down to the appropriate chord. What do you think? -- David W. Fenton http://www.bway.net/~dfenton David Fenton Associates http://www.bway.net/~dfassoc _______________________________________________ Finale mailing list [EMAIL PROTECTED] http://mail.shsu.edu/mailman/listinfo/finale
