I've downloaded the image and attached it for you :)

Comparing to the Solesmes editions, the first neume looks like it is sitting a bit high. Also, the fourth and eighth neumes should use the same glyph on the bottom note as the first.

I don't know what you exactly mean with ‘replicate’, but when I’m comparing with an engraving from Solesmes (Nocturnale Solesmense, 2017, p. 75), there are some striking differences and details you probably would like to reconsider:

- the placement of the horizontal and vertical episemas is inconsistent; sometimes too close to a note head, sometimes too far away; vertical episemas sometimes go through the staff lines
- the shape of your note heads seems to me too wide; Solesmes has more vertically elongated note heads, slightly rectangular
- the lower note of your podatus should be a bit wider and completely square, with straight edges
- the ‘scribbles’ of your quilisma are not evenly spaced
- your bar lines are thicker than the staff lines
- the placement of a series of descending rhombus, as in a climacus, is not perfectly diagonal; the corners of a rhombus from Solesmes is more subtle
- Solesmes puts more space between the staff lines

In general:
- hyphens are placed directly after a syllable; they are a not centered between two syllables
- Latin words of two syllables don’t get an accent to indicate the stress; it’s always the first syllable

@smvanroode, thanks for the feedback.
1. Yes, I'm working on the episema. Difficult to code but I will fix it
2. My goal was not to exactly copy Solesmes, sorry, "replicate" was not the best word to use. I should have said "be able to reproduce all notation and formatting." I intentionally made the note heads a little wider. We'll see what other users say. If everyone objects, I will change them.
3. Ok, I got completely different feedback from another chant expert on the podatus... also will wait and see what the consensus is.
4. I don't understand regarding the quilisma, can you clarify?
5. Bar lines and staff lines can all be easily changed in Dorico
6. I can adjust the design of the rhombus easily, thanks
7. This is a setting in Dorico

So in summary, some of these things are settings in Dorico that each user can choose. And it is not a priority to me to create an output that is indistinguishable from Solesmes, unless users request it. Thanks again for your feedback.

Hmm... that's unfortunate. But as you say, it needs to be accepted by the community. I'll wait for the next round of feedback and an willing to change it, it will just be a ton of work.

I find Solesmes too narrow, but that's just me. I might also consider widening the diamonds. That's valid, for sure.

>You have to remember that we are all used to looking at Solesmes & GABC output, so something that is different just strikes the eye as weird, even if it's perfectly legitimate. So you're really fighting the force of inertia, I'd say.

To the point where even non-Solesmes chant was rendered with type similar to that of the Solesmes editions, even for non-musical elements, and because of Gregorio, that style of neume is pretty widely adopted even for non-Solesmes editions of pretty much all kinds of chant today.

I like Vollkorn... it's one of the few fonts I've seen that has genuine bold typefaces that actually stand out from the regular weight, and it also has æ with a stress mark above it, which is useful for latin chant. Very few fonts include this ligature with the accent mark above it.

I, for one, don't mind the chunkier look

As a recovering Finale user, Bravura looks "chunky" to me. So this will fit right in. I don't know how many people will do all-chant projects in Dorico. But for alternatim projects, this is badly needed, and if the look is compatible with Bravura, all the better I say.

Thank you, Dan.

Ok everybody, feedback please. How easy are the current options to use, like Gregorio or Illuminare? Are they painful and slow? How long does it take to engrave a moderately complex 4-system chant, for example?

Interesting, thanks. This seems very complex to me, at least a steep learning curve initially. And you don't have fine-grain control of the final output necessarily. But I do see that it's predictable and "safe" once you learn the interface.

I'm asking because MusChant needs to be as easy to learn and work with as possible, or else it's not compelling. I think it's much easier than this. But I'll wait to see what other users say.

Well, that's promising to hear. That is, I'm not glad GABC is frustrating, but I am glad to hear that MusChant sounds like a significant improvement in user experience and control. Hopefully it is that.

Gregorio is not good for critical editions, but I don't think that MusChant would be either.

Can you elaborate on this? The goal with MusChant is complete control over literally every mark on the page. What aspect of critical editions are you concerned would be lacking? Thanks.

I may not be understanding, but Dorico can be output as PDF, PNG, or SVG even. So displaying it online is quite easy. Granted, it can't be edited online.

I may not be understanding, but Dorico can be output as PDF, PNG, or SVG even. So displaying it online is quite easy. Granted, it can't be edited online.

You’re not understanding. It’s that gabc can be used, in addition to PDF output by LaTeX, used independently to be displayed and edited online.

But I also am not interested in learning to code laTex or LilyPond and cobbling things together with photoshop or affinity publisher

I wasn’t interested either. As I said I don’t always appreciate Gregorio being wedded to LaTeX, and I can accept that LaTeX is a time-suck. But gabc? That is a lot harder to accept, especially as you don’t have to input it in the way it was originally intended to be transcribed.

But it is quite possible to have acceptable results inserting the graphics the other way around, i.e. do the chant stuff with Gregorio and add graphics of the modern scores. There is now very good LilyPond integration thanks to Fr Samuel.

I actually had the most problems once I tried Gregorio itself and then Scribus (which mishandled the font of the text somehow). So then I switched to only Gregorio and LaTeX, but I passed by Affinity… and I have to say, it was incredibly fiddly with graphics. I cannot imagine what would make a Gregorio score different from anything produced with notation software in that respect.

Anyway it really does seem like using Gregorio and learning to use some other LaTeX packages, as well as other PDF tools if necessary, would be helpful for some users here. I also don’t understand why adapting to the limits of Gregorio or these other tools (Illuminare/Source and Summit) isn’t mentioned. That is to say if your desktop publishing software is imposing requirements that you can’t fix, I’m not sure that dropping Gregorio or not using it and then trying to make the output from the web work with DTP software is any less tedious or easier than just using Gregorio and tweaking it in LaTeX, whether or not you finish the whole product in LaTeX.

I desperately desire to have new and old within the same document

That's my primary use-case as well. Choirs that sing chant read square notes; if you give them chant in round notes they get twitchy. And there's a lot of early polyphony that is alternatim (even verses in chant, say) or has a chant incipit. The chant verses are generally not in the manuscript (One finds them sometimes in 18th c, Italian Mss, on 5 lines with chunky Medicean notation that can be approximated well enough with square and rectangular noteheads, at least in Finale; I haven't done this in Dorico yet.). And as a matter of ease of use, the chant needs to be there. Since I'm doing editions for liturgical use, and I'm lazy (and 50 miles from most of the sources I'd need), I use Solesmes* chant rather than trying to find the exact historical dialect of chant appropriate to the music. So something like this would work well for me.

*I note that the Forum software autocorrects "Solesmes" to "Salesmen". There's a joke there somewhere.

I fail to see how designing yet another square note font brings us closer to the interoperability of chant engraving software.

If people confuse the layers of the chant engraving software stack, they will not go anywhere.

1. Description layer

For scores : the list of notes in a score and their attributes (rhythmic and otherwise). This description can be relative-height, absolute-pitch (like .ly) or absolute-height, relative-pitch (like .gabc). It can be segmented by measure (most fitting for metered
music) or by syllable (most fitting for free-rhythm music).

For printed texts (translations, psalms, readings, rubrics) : the actual text, appropriate metadata (like scripture references and liturgical function), and flags signaling the nature of the text (a rubric, a translation).

2. Structure layer

The ordered list of scores and texts to be included in some publication, with pointers to their descriptions (above).

3. Layout layer

The page size, the configuration of footers, headers and titles, pointers to the fonts used for the text and the music (and their font sizes), anything related to the production of indices and tables of contents.

4. Graphics layer

The actual fonts, images, and other resources used throughout the book that are assembled together according to the other three layers to produce a finished result.


GABC is a description language for chant that supports an extension towards adiastematic neumes (NABC) but does not support polyphony or length stems - yet. Chant described in GABC can be displayed by any software that supports it.

GABC, the description language, is not gregorio, the program that compiles it into a set of calls to a font, nor is it greciliae, the font used by LaTeX to build a PDF from the output of gregorio.

Sure, glyphs in greciliae have their own names and Dorico calls them by different names. This is a trivial issue - the issue of taking two or three hours to make a correspondence table.

Pieces of software that attempt to force every layer to their own standard - their own description language, their own structure language, their own layout language, their own fonts and graphics - are doomed to never be interoperable with the vast ecosystem of chant scores currently in place.

Gregobase has twenty thousand entries, Neumz has about half that, all in GABC, and Repertorium has trained optical music recognition software using GABC as training data, which means the contents of thousands of manuscripts may soon become available as GABC.

Is GABC ideal? No, for many reasons, one of the biggest, as evidenced by this thread, being that it does not play nicely with polyphony and/or modern notation.

But saying "GABC sucks, I'll make a font" ?!? A font is not an alternative to GABC. An alternative to GABC is a score description language with the ability to represent chant, polyphony and modern notation - and if the creator of this language has one teaspoon of wisdom, they will make it so existing GABC can be effortlessly and reliably converted into that more powerful description language.

I fail to see how designing yet another square note font brings us closer to the interoperability of chant engraving software.

That's not really the point of Dan's project as I understand it. He's found a way to in essence use neumes as a text layer in Dorico. His impetus was a project to reset The English Hymnal, which contains 4-line chant voices over a conventional organ accompaniment. I think he's seeing that there's more to this, and more possible uses for it, than he thought going in.

We've just discovered interoperability of notation software via SMuFL. And there are a bunch of neume bits in SMuFL, but nobody's SMuFL implementation really allows them to be used effectively. And spending a bunch of developer money on making that possible isn't happening, because the market's too small. I've set chant in Finale, and it was an unlovely PITA. I suspect that unaided Dorico would be the same, plus me not really knowing the program yet. I'm hoping Dan's font will be ready when I'm ready for it, which may be soon.

Xmarteo is right.

Yeah, the point is that I can do this extremely easily now. As James says, feel free to use whatever you want.

I think Xmarteo raised some valid points, yet I think there is no reason to rant. Dan does want he is interested in and what helps him in short term. Fine.

I think it would be good to give it a positive turn: What are the features GABC/gregorio is lacking? I've seen the development has lost pace, the last release of gregorio is from 2021. That's never a good sign for an Open Source project. Collecting new requirements could be a first step to revive it.

Integration with modern notation like in Dan's example would probably need an integration into lilypond or sth else.

PS: Repertorium -- need to look at that, thanks! Does anybody know about a good program for optical recognition for mensural music? I transcribed some falsobordone settings in the past, but it takes a lot of time and is so repetitive... seems like a perfect candidate for automation.

Xmarteo is also free as an eagle to ignore Dan’s font.

But it solves a real need for some others in the meantime.

I’m failing to see where anyone said otherwise. I just want these things done well.

yet I think there is no reason to rant.

I don’t consider that a rant, and I’d know what one is. But if we’re tone-policing like this, then the conversation isn’t fruitful. No one said that Dan can’t do what he wants, but this is a forum of people who might actually use this, and of people who happen to know a lot about chant rhythm, which necessarily includes notation, as we are active church musicians and not secular musicologists who claim to be indifferent to that. That also means getting on the same page about Gregorio and gabc — and defending them as one may see fit.

The feedback can be taken or left.

Collecting new requirements could be a first step to revive it.

it’d be better to ask the developers what needs to be done to achieve the current needs. But Fr Samuel has said a million times that a better way to handle lines is needed before doing a lot of things. Knowing precisely where the breaks are only occurs after a first pass when typesetting. This then allows for the typesetting of the space between lines and things like braces for psalmody that require this space to be defined.

Like e.g.: Easy and well built way to set multiple lines of text to one line of notation

That requires something which has long been requested (since at least Gregorio 4.1 I think) and not possible due to this problem. But it is not like the devs are silent about it. I was given this response in the last year or so!

change whether or not two syllable words get an accent mark, use æ instead of ae

I understand that ctrl+f can be a bit tedious… but in both cases I don’t see why the package needs to do this. However post it as a feature request.

Fork Musescore to allow for use of gabc or importing gtex files. I don’t think that Gregorio should concern itself much with other programs; the Scribus integration that someone devised is finicky enough, and using Lilypond and importing that to LaTeX may not be good or easy enough but it exists as an option. Otherwise, I am struggling to see why it is so hard to treat Gregorio scores as graphics — but this may well be related to certain DTP programs versus others.

For my part, I would like more clarity in the documentation about features that apply to only the next score versus all following scores until reverting to the default.

Otherwise, I am struggling to see why it is so hard to treat Gregorio scores as graphics

The issue is that if you're generating graphics for combination with modern notation as in the example Dan just shared, it's extremely onerous to do so. You have no way of dynamically making adjustments to either part to make them work together. At least one of the scores will be a static graphic that, if it needs any adjusting, will require going back to that program, guessing just how much to adjust spacing or whatever, exporting as a graphic, reimporting to another program, comparing, and then doing the whole process over again just because something doesn't line up correctly.

Dan's solution is a dynamic font that allows the two forms of notation to coexist in a dynamic and fluid way. That is a VERY major hurdle to jump. I've been in the habit of importing chant PDFs into worship aids and then tweaking them in affinity publisher. Selecting individual elements of neumes and grouping and then nudging them is a real time suck, let me tell you. And that's only to make fine graphical adjustments to chant-only scores, not combination scores as shown above. Perish the thought of having to use the workflow I just described.

Does anybody know about a good program for optical recognition for mensural music?

"Does anyone know about a good program for optical recognition of music?" FIFY, LOL! I sure don't.

Even dealing with published music, symbols are different enough between publishers that things could get tricky. I could see a situation where the software trained on various publishers, and you could set a default for a particular one. I think you would need some kind of AI to deal with translating triple time in particular, because the goal would not be recognizing notes, but auto-transcribing into modern notation. It's a thing to be wished for, certainly, because there's a ton of useful music that hasn't been edited yet, and if that could be automatic, so much the better.

That said, I find inputting your typical duple-time counterreformation church music print to be kind of relaxing. And one of the really nice things about Dorico in that regard is that I don't need to pay any attention to measures, but just enter what I see and let the software deal with barline crossing.

Over at scoring notes, there is now a really thorough blog post covering this typeface:

https://www.scoringnotes.com/meta/cantorum-a-plainchant-font-for-dorico/

I didn’t have a chance to weigh in on this thread before, but as someone who uses both Gregorio and Dorico every day, I will certainly adopt this. It’s not for everything; I really like Gregorio for chant functionality. But this looks like a really good option for doing bits of chant in Dorico. One example of a project where this would have been great is an edition I did of the Viadana one-voice Mass a while ago, which is in alternatim and needs some chant sections. I did those with Gregorio and importing images into Dorico.

Somebody mentioned an attempt doing the marriage within Indesign. I don’t know if they were referring to the work I did for the creation of hybrid notation, but that is a bit different. The method imports a GABC pdf and then manually superimposes the modern notation accomp below NEVER HAVING TO ADJUST the chant graphic. It also presumes that the chant melody is not played by the organ. Here was one of the latest examples of that experiment using my much preferred variant of the NOH.

The real benefit of this is condensing horizontal real estate, discarding the need for metered notation, and allowing an organist to easily sing, direct and play all at the same time. This also eliminates the need for turning pages which I don’t want to think about when I am praying.

One of my biggest pet peeves in trying to interpret Gregorian Chant with modern notation is that you lose the sense of the timing and flow of the arsis and thesis. It’s like using two brain spaces (schools of thought) at the same time. This method attempts to eliminate that conundrum. The chant really needs to emote a timeless and meterless aura. This is the only way I can think of doing that while it also maintains the Mora Vocis. In my view, setting the chant melody in modern notation stretches the spatial horizontal timing to be much larger than is preferred.

@dankreider
In your example above, I don’t see the need for Gregorian notation. It’s in English and it has a sense of hymn meter. (uses breves and semis) Can you explain why using Square Notes would be a benefit?