12 o'clock puncta inclinata

[eroux]

See this thread, it would be helpful for some project to have 12 o'clock puncta inclinata. The difficult part is to know when to use them I guess. i'm thinking something like:

nb_d = the number of pair of consecutive notes where the first is above the second
nb_a = the number of pair of consecutive notes where the first is below the second
if (nb_d == nb_a)
    use 12 o'clock version
if (nb_d > nb_a)
   use 1 o'clock version
else
   use 11 o'clock version

does that sound reasonable?

[henryso]

I tried drawing this figure, but since I cannot base it off another figure, I can't get the "weight" to feel right compared to the other glyphs.

[eroux]

Oh, interesting... Well, maybe we can start with a first draft and once we can experiment on it we can ask a few people on the list what they think... does that sound feasible?

[eroux]

I just realized that in the original caeciliae the puncta inclinata are 12 o'clock so maybe we should just take these? They've been rotated at some point after the import

[rpspringuel]

That's definitely where I'd start.

[henryso]

Do you remember how Caeciliae was scaled to form Greciliae? If I simply copy-paste, the sizes don't match.

[eroux]

Hmm not really... maybe it's a question of different em size?

[henryso]

Maybe if the tilted glyph was made by rotation, I can make the un-tilted glyph by rotation. I'll give this a shot.

[henryso]

On the gregorio font, the punctum inclinatum is curved towards its leaning. Should I just draw a straight up diamond in gregorio for this "12 o'clock" shape? Or maybe curve it downwards?

[rpspringuel]

Should I just draw a straight up diamond in gregorio for this "12 o'clock" shape?

That sounds like a good place to start. We can always modify it based on feedback later.

[henryso]

It seems we only support the tailed version "auctus" and the small-note version "deminutus" on the descending shape. Should we support these figures on the "12 o'clock" shape? On the ascending shape?

[rpspringuel]

The original thread didn't call for that support so I'm inclined not to add it for now.

[henryso]

Ok, so I've been coding this and the problem now is when to actually use the new figure. Currently, the orientation of the punctum inclinatum depends on the note after a "run" of punctum inclinatum figures. I'm thinking that if a run is in unison, regardless of the figure that follows, then use the 12 o'clock figure. The edge case is if the run is of exactly one note. Should the single note be the 12 o'clock figure or one that points at the next note? Or perhaps something else?

[eroux]

When there is just one note I would tend to to look at the previous note and follow the direction (if higher, then tilted on the right), unless there is a space before and no space after, in which case the note following the punctum inclinatum is taken into account. In the strange case where it's surrounded by two spaces, I don't really know, maybe not tilted... would that be reasonable? 2018-08-01 20:52 GMT+02:00 Henry So <notifications@github.com>:

…

Ok, so I've been coding this and the problem now is when to actually use the new figure. Currently, the orientation of the punctum inclinatum depends on the note after a "run" of punctum inclinatum figures. I'm thinking that if a run is in unison, regardless of the figure that follows, then use the 12 o'clock figure. The edge case is if the run is of exactly one note. Should the single note be the 12 o'clock figure or one that points at the next note? Or perhaps something else? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#1425 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AADtxNChIzHkH1tA2j7027hjbMDGu4Arks5uMfjUgaJpZM4UhQUx> .

-- Elie

[henryso]

After some reflection, I think the rules may be more complex as a lone punctum inclinatum followed by a note would likely lead into the note after it just as (you say) a lone punctum inclinatum after a note would likely follow that note before it. So then, what if it's between two notes? Therefore, I'm tempted to make a lone punctum inclinatum use the "12 o'clock" shape, run it through some scores and then decide what to do next. If it proves too difficult to make look right, we can decide to never automatically use the 12 o'clock shape, and just leave it as an option selectable in gabc (i.e., G2).

[eroux]

Well, for the corner cases you describe we can always keep the current behavior, but I think it would be useful to implement the cases we describes, as they'll probably cover 95% of the scores...

[henryso]

The algorithm that I've finally settled on is NOT the one on this issue, as that algorithm broke too many existing behaviors. Instead, the "12 o'clock" unison figure will be used when there are more than one puncta inclinata at the unison. When there is only one punctum inclinatum, it will follow the note before and/or after it within the syllable, meaning ascending should the melody to either side be ascending, descending should the melody to either side be descending, or unison should the melody to either side stay at the unison.