Ooloi: An Organism Evolved

Half Measures

8/11/2025

Fractional numerators are now official. Ooloi’s time-signature system (ADR-0033) supports everything from common time to irrational and additive meters, and now fractional counts like “2½ / 4”. Halves only, for now – enough to cover Grainger and Chávez without venturing into Boulezian fantasy.

Full ADR: ADR-0033-Time-Signature-Architecture

8 Comments

Roland Gurt

8/11/2025 22:49:08

This is great and shows that this project really goes in depth with a lot of knowledge and experience!

How could one best acommodate cases like the "3x5/12" I mentioned in my comment to the previous post, or something like the "¢¢" (double cut c) seen in Schuberts G flat major impromptu D.899/3? Maybe it could be possible to substitute any glyph for displaying a time signature without changing its internal value? I feel like it would be a "unique selling point" for Ooloi to offer as much freedom as possible to the engraver who needs to "bend the rules" every now and then (as can be seen in many a great engraving) – in contrast to other scoring programs which try to "help" more but also inhibit or prevent more options.

Another thing that might be missing from the documentation: open time signatures, notated either as an "X", another symbol, or by simply not writing any time signature.

Thanks for your great work and best regards!

Peter Bengtson link

9/11/2025 09:20:05

Roland – thanks for this. You're absolutely right about escape hatches. The engraver should have full control, and the software shouldn't prevent things through arbitrary design choices.

The separation you're describing – semantic value vs visual presentation – is exactly how it should work. The Schubert ¢¢ example is straightforward: semantically 2/2, visually whatever glyph is needed. The architecture already has hooks for this kind of override, they just need extending for arbitrary glyph substitution.

Open time signatures (X, no signature, etc.) is just a flag on the measure – "this measure has duration but no displayed metre".

The Adès notation is exactly the kind of case I want to handle properly, but I need the foundation in place first. The question is whether it's consistent notation with clear semantics, or whether it's personal to Adès – that affects how to generalise it. The current design is deliberately extensible for cases like this. I'm looking forward to working through them once the basics are solid.

I really appreciate the engagement. Let's put a pin in this and return to it when the time comes. You can hold me to that.

Roland Gurt

11/11/2025 09:37:10

Thank you Peter! I myself know the "multiplicator"-x only from Adès, though Daniel Spreadbury might know more sources, as he included the x-glyph into the time signatures section of the SMuFL specification https://w3c.github.io/smufl/latest/tables/time-signatures.html
It seems to me that using the multiplicator would always just be a way to highlight a beat grouping, so one could just enter the "ordinary" time signature (4/2 in the Schubert example, 15/12 in the Adès example), and then change the visual appearance to 3x5/12 – no dire need for a direct entry method and "translation" by the software, since these time signatures are quite idiosyncratic choices, lying comfortably in the realm of a general "glyph substitution"-workflow IMO.

Regarding (fractional) metres: you may know this resource already, but many examples can be found in Gardner Read - Modern Rhythmic Notation (1978), p.17, p.84. He also has an impressive catalogue of composers using any particular metre, starting at p.177 (including usage of 3¼/4 and 4¾/4, though not by well-known composers).

Peter Bengtson link

11/11/2025 14:18:58

Ah, yes - beat grouping is of course key here, and the way it affects beaming. Ooloi's time signature handling controls this as well, as it should, so supporting grouping for x-based metres is actually quite straightforward.

Magnus Johansson

9/11/2025 11:11:23

Thanks, Peter, for ADR-0033. A few questions:

1. Shouldn't Ooloi also have a maxima beside the breve and the longa?
2. Will there in Ooloi be checkboxes in the Time signature dialog window for "Orff notation" and "No unit" like in Igor Engraver?

Peter Bengtson link

9/11/2025 11:43:19

1. It should, absolutely. The 1200s and 1300s deserve their fair share. However, the longa and maxima have certain ambiguities associated with them as the usual doubling of value also can be a tripling – meaning that the whole issue of mensural notation to which this properly belongs should be delegated to specialised plugins, or I'd be stuck in implementing unusual edge cases in the core engine, which would be a type of feature creep. And we all know what _that_ led to, 25 years ago. ;)

2. It's far too early for me to commit to specific UX design issues; I prefer to deal with that level of granularity when I get to it. Otherwise I would be bound to keep promises which I don't want to give prematurely. But I can speculate: there'll probably be a dropdown menu with three options: integer, note, or none.

Magnus Johansson

9/11/2025 16:30:07

Peter, I have started to study "Le rythme, la musique et l'éducation". In the text on page 209 Jaques-Dalcroze presents an extended system of prolongation dots:

a)
𝅗𝅥half note· = half note + quarter note
half note·· = half note + quarter note + eighth note
half note··· = half note + quarter note + eighth note + sixteenth note

b)
half note: = half note + eighth note
half note:· = half note + eighth note + sixteenth note
half note·: = half note + quarter note + sixteenth note

c)
Here he uses three vertical dots that prolongs the half note with a sixteenth note. They can be used together with the single dot placed before or after the other dots like under section b) here above.

Do you think Ooloi should cover also this notation variant?

Peter Bengtson link

9/11/2025 17:14:46

That’s a fascinating system – Dalcroze was trying to formalise proportional rhythm long before symbolic computing made it easy to express. But no, Ooloi won’t encode that directly. The aim is to model mainstream musical semantics precisely; once those are solid, any historical or experimental variants like this belong in plugins. Otherwise the core risks collapsing under special-case logic instead of remaining general and predictable.

When the plugin framework is in place, this kind of notation will be straightforward to implement on top of the existing duration model without burdening it.

OOLOI

OVERVIEW

DOCUMENTATION

NEWSLETTER

Half Measures

Leave a Reply.

Author

Search

Archives

Categories