I’d like to continue to make the case for putting choral condensing improvements on the roadmap, as soon as possible.
My typical workflow uses four section players because I need to export complete MusicXML parts to another application (Synthesizer V as a plugin, in my case – others may use Cantai directly in Dorico) for rehearsal tracks and demos. Condensing is, therefore, the right underlying model: even in unison passages between Soprano and Alto where they may combine on a single staff, each track must retain a complete part.
The problem, of course, is that Dorico’s condensing still does not work well with lyrics, and it does not provide the automatic divisi/unison arrows and labels available with true divisi when condensing needs to change.
True divisi solves the engraving side beautifully, but not the export side. In MusicXML, the second track is blank during unison passages, so Synthesizer V does not receive two complete vocal lines. That leads to a lot of tedious copying and pasting.
This leaves choral writers choosing between one of several bad options:
correct underlying parts, but awkward engraving and workarounds when the divisions need to shift;
clean engraving and incomplete export data, resulting in copying and pasting later;
doing some sort of “home-grown” condensing/divisi where staves are manually hidden, risking divergence between the visible and hidden staves.
maintaining duplicate versions of the score (separate layouts within the same flow/file, or in separate flows/files), and risking divergent versions (also no better than the competition).
The workflow I’ve described, though, is becoming common for choral composers and arrangers creating rehearsal tracks, demos, and other accessibility materials. Dorico already has nearly all the necessary pieces; what is missing is some kind of fully choral-aware bridge between condensing, lyrics, divisi notation, and export.
Not to get too much into solutioning, since that’s up to Dorico’s product team and engineers, but one of the following would be a major win for the choral community:
Ideal solution: Improved lyric handling in condensing, with automatic choral divisi/unison indications.
Acceptable for now: An option for fully populating tracks in MIDI and XML exports for divisi staves that duplicate all exportable elements when staves become unisons.
While you are waiting for condensing to handle lyrics better, leave the lyrics off the Alto part when the singers are singing in unison. Score readers will still see the lyrics between the Soprano and Alto when staves diverge, as will singers using a piano/vocal score.
The problem with leaving any lyrics off, anywhere, is they don’t make it into the XML export.
I don’t disagree with you, though, at all – that’s another improvement I’d love to see, the option to have one set of lyrics be attached to multiple staves (namely, a staff above and a staff below them). At the moment, I’m keeping multiple lines of lyrics so they make it into the export – not ideal, but still in the realm of permissible for concert choral music (not so much for hymnals).
Sorta kinda. With normal condensing, when two staves merge down to a single staff visually, the lyrics still stay attached to the separate tracks on XML export.
I generally work in Galley view in open score, writing as I please, and then take care of condensing where possible in Engrave mode when I’m reasonably done with either the entire piece, or at least with different large sections of music. Condensing automatically solves so many issues well already, but there are some issues with lyrics where the only truly correct engraving solution right now involves deleting lyrics out of certain staves in Write mode.
Any solution that starts with “you have to have two versions of the file, one for XML and one for display” puts Dorico back on the same level as where other notation software has been stuck for decades, where I feel like they’re actually so close to having this one solved.
Synthesizer V generates sung vocals from MIDI data and lyrics. I use Dorico’s MusicXML export to bring in both notes and lyrics for rehearsal tracks and demos.
MIDI import was standard with Synth V version 1, but multisyllabic words and melismas required substantial cleanup. Synth V 2’s MusicXML import preserves syllable divisions and melismas fairly reliably, which saves a great deal of time.
Several choral composers and arrangers I know use the a similar workflow, whether with Dorico or Sibelius.
My understanding, based on what @eric.t.knechtges says above, is that the uncondensed score would work for XML, but lyrics in the same score would be problematic when condensed for printing.