For the intrepid, especially those annoyed with the purported input-sluggishness of musescore et al, an interesting text-based alternative is LilyPond https://lilypond.org/
My dad wrote an opera using LilyPond in vim, though I believe these days he's actually doing more with supercollider, which skips sheetmusic and goes right to sounds: https://supercollider.github.io/
I love lilypond as a programmer because I can use git. I have a private git repo that has lead sheets for all my favorite jazz standards (with my own reharms), notation at various levels for my originals, arrangements of songs for my a cappella friends, and a cue for a full film score project I did. The output quality is amazing, but the best part is I don't have to worry about forward/backward compatibility. None of my old Sibelius scores are openable anymore without spending hundreds on an upgrade, but all my lilypond scores will always be available to me.
I use it for the same reason, but also it is worlds ahead of Musescore if you want to put together a songbook (particularly if you also have tex experience). The best advice Musescore has for making a songbook with many scores is to use a PDF editor on the output.
Also as a programmer, Lilypond is programmable, and nearly infinitely customizable (similar to tex). I love that I can write some style information in a "header" and include it in every score. In my current project I am writing a lot of SATB choral songs in book format. But I also want to do a melody only edition. Just tweaking a header file, I can hide all the other voices, no problem.
(I also use Musescore, which I prefer for composition or simple projects. But when I want to output a large finished product, I always go back to Lilypond.)
I used to be big on LilyPond, but in the end I concluded that my musical brain works much more effortlessly when there's an actual graphical score in front of it. (Maybe I'd feel differently if I had one of those keyboardist/composer/jazz brains and/or perfect pitch.) Plug in a MIDI keyboard, learn the shortcuts, and input is fast enough in MuseScore.
There's definitely room for something that bridges the advantages of the different approaches, but it's a difficult problem and it's a $0 billion market, so having passionate people make MuseScore better is probably the best path forward.
LilyPond is incredible for the one purpose it's built for: engraving notated music. But music engraving is the usually the very last step of a compositional or editorial process. As such, other methods, even pen and paper, may well be superior tools for the creative process of composition (although we should also remember the testimony of Bach's sons that the elder Bach always composed away from the keyboard, and regarded the inability to compose without an instrument at hand to be evidence of poverty of invention; of course, this does not rule out the possibility that Bach tested out out his draft compositions on a keyboard).
On the editorial side, LilyPond is an incredible tool for creating modern editions of earlier music, specifically music in mensural notation. It comes with common glyphs and has full ligature support, and the separation of content and presentation means that one can reproduce the look of early manuscripts and prints while generating a modern score from the same source file. This has been really great at reducing transcription errors by easing the mental burden of transcription, transposition and note reduction, which can all be handled automatically by LilyPond. It's not perfect, but miles ahead of any alternative.
It requires a tablet and pencil/stylus, and it’s a bit pricey. But I’ve found the handwritten interface, once I got used to it, to be fairly intuitive.
I agree with you that a graphical interface works better. All existing sheet music follow this same format. It seems strange that some are trying to go in a vastly different text based approach.
I’m a beginner pianist, and I’ve been writing down stuff in ABC notation[1] using Obsidian[2] and a plugin[3] for instant previews. I was already using Obsidian for all of my plain-text Markdown notes, so it was the obvious choice for me. Obsidian also has a nice WYSIWYG-preserving PDF export feature in case I need some printed sheets. :)
ABC is easier to learn but also not feature rich. ABC vs Lilypond is similar to asciiMath vs MathJax in this aspect. You'll soon find its limitation and needs more.
(But Lilypond is hard, harder than writing documents in LaTeX for example.)
I'd been trying to find a way to "write" the music in a way more similar to how I think about it.
from this pursuit, I've come to realize that "sheet music" is just the only technologically available way to store the music back in the day. a practice which had an entire publishing and printing industry around it; this is how composers in the 19th century made money
BUT THEN, sound recording technology made that be somewhat obsolete, and then synthesizers and computers came and made music writing notation be less and less capable of keeping up the pace of musical technology innovation.
all in all, I'm considering that what if I say that PCM is a writing and start 'handing down music' directly as 0s and 1s in at a 16 bit depth (two bytes per sample) at a 44100 samples per second?
then again that's cumbersome, so the super collider code (and everything needed to run it) IS now the 'music notation' which includes a description of the instrument as well as the song/the music. (traditional written sheet music notation does not really include a description of the instrument beyond a reference by name)
this also "turns" the music interpreter into a digital to analog converter. lol
A composition, to varying degrees depending on the composer, leaves various degrees of freedom for the performance. So the PCM encoding of a performance is entirely different to the composition ... conceptually.
However, a good musician may be able to listen to the PCM encoding (aka "digital audio") and from that frame their own understanding of the composition, thus leading to a new performance.
A good friend of mine used to be a core developer there. He started by nagging them about the horizontal alignment of notes in different instruments/voices and that blossomed to over a year of collaboration.
It's very much like with kerning - once you see a sheet with decent alignment, you can't unsee it.
When you say he wrote an opera using LilyPond in vim, do you mean to say that his setup for composing was a vim terminal? Or did he write a draft first on some other medium such as paper and then inputted that into vim?
If anyone is yet to see it, Tantacrul (on youtube) is now deeply involved with MuseScore (UX lead or similar?); he put out an awesome vid on the design of MS4 at the beginning of the year: https://youtu.be/Qct6LKbneKQ
His video on how Sibelius' UX is a pile of shit remains one of my favourite vids on yt: https://youtu.be/dKx1wnXClcI
Weirdly I checked his channel yesterday (for the first time since many months) if there happens to be a new video! Didn't know that there will be a new Musescore release.
Out of curiosity, what ever happened to all of the drama with MuseScore, locking downloads behind a subscription, the Audacity acquisition + telemetry debacle? It looks like Tenacity still exists in some form, though it does not appear like there is a ton of significant activity on it in the last month or so.
While I don't know about the OSS community's stance, I can say that most casual users have embraced the new version of MuseScore. They added a lot of important features with the latest version, and I think most people don't really care about the telemetry.
I think some of this might be avoidable by downloading Musescore separately from the larger package they've promoting now and not using some features? That's what I do, anyhow.
Note, this would not be the first open source music app to lock downloads behind paywall, e.g. Ardour is known for it: https://community.ardour.org/download
MuseScore desktop (the composition / notation app) has never been behind a paywall. This is a mixup with the mobile application which is a sheet music viewer that features copyrighted scores.
You can get it without charge from just about any and every Linux distribution.
You can get the source code without charge from ardour.org/download (or from our mirror repo on github).
You can get it (legally) without charge from anybody else who already has it.
The only thing you cannot do is to get a ready-to-run binary without charge from ardour.org itself. You can opt to pay as little as US$1 for that, however.
I believe it was the sheet music that was paywalled, not the software. That was a bit concerning since as far as I know a large portion of it was UGC and some of it even public domain, so it was weird to ask for a subscription payment for it. At this point though, it does look like they have made some amends, though I'm not sure if everyone's satisfied. Doesn't seem to matter too much in the grand scheme of things, overall, though it was an unfortunate situation to watch unfold even as an outsider.
Note: Linux distros still provide free downloads for Ardour, and Ardour's source code is free for anyone to build. Ardour's official binary builds are what is behind a paywall.
Also the "locking downloads behind a paywall" only applies to scores that users have uploaded to musical score social media sharing site musescore.com, but the desktop program MuseScore binaries are available for free on Win, Mac, Linux, BSDs via https://musescore.org/en/download. You can anyway post your scores anywhere for free. And there may be proprietary VSTs that you can pay to download via MuseHub, but MuseScore's baseline audio synthesis (as well as any free soundfonts you can freely download) are free.
MuseScore is one of those rare open source projects where it’s obvious that a serious amount of thought has been put into the user experience. I only dabble in engraving, but every time I’ve tried to do something more complex the UI has been ready and waiting to help me accomplish it.
Arguably WYSIWYG engraving software only exists to provide a workable user experience (vs editing lilypond or god forbid musicxml notation in a text editor).
Having worked a fair bit on it and as a regular user of it I'm only too aware of areas still in desperate need of a better UI/UX though. Still hope to go back to contributing more regularly at some point, though I didn't especially enjoy having to regularly deal with the foibles of a large complex C++ code base and digging into the internals of Qt to figure out weird UI bugs.
Musescore the company is a disaster. They took my money for a Pro subscription, without delivering the actual benefits. Their support organization is, as far as I can tell, nonexistent. After six months of trying, I have yet to find a way to get them to respond to my request for support.
Sounds like it's time to call them out (politely) on twitter with a tentacrul at-mentioned. That at least seems to get their attention. Unless you already did that in a non-polite fashion and got muted, of course.
Still using MuseScore 3. Even without the new sounds packs enabled, the latency typing notes on a (computer) keyboard is still too high. I would rather be stuck with the old version then deal with the unsettling 100-200ms lag for every input. It's quite an unfortunate trend that so many modern applications are neglecting basic UI responsiveness.
I've been heavily using MuseScore 4 lately and haven't noticed that type of latency on the Linux AppImage. My only gripe with 4 is the inability to open multiple scores in tabs. It opens them in separate windows instead. It makes closing scores in that situation difficult. If you have two open windows and quit one of them, both quit. If you instead just close one of the scores the window remains open with nothing in it. It seems to be a can't/won't fix problem as it has to do with their new soundpack system. It's probably the number one gripe users have with 4.
I think with 4.0 release, they were more focus on getting engraving more correct (among other big changes like the new sound system), and that resulted in a performance hit, but the 4.1 patch notes specifically say there are significant performance improvements, so it would be worth it to try it out again.
The audio driver still matters, though. If you're on Windows rather than a Mac, not using ASIO will always incur a truly incredible latency, even with real MIDI devices. Mac's Core Audio is quite a bit better in that respect, but pretty much everything ships with Asio4All for Windows users, and it's still an excellent idea to use it if you don't have an ASIO audio interface.
Actually, WASAPI can deliver quite low latencies (<10ms) and it can also operate in shared mode, so I would always prefer it over ASIO4all when using the built-in soundcard.
I found the app to be incredibly well-featured but ridiculously unintuitive. Nothing worked the way I wanted to and I spent half my time fighting it when trying to insert notes and such.
I have no background in authoring sheet music so my sense is that this is just me and my whole mental model is probably wrong.
Did anyone else experience this? Or did anyone else find it to be the opposite: super intuitive and smooth?
I hear if you create a 40 minute youtube video on how bad the UI is, with a bit of humour mixed in, but mostly valid criticism, they hire you as a product manager.
>I have no background in authoring sheet music so my sense is that this is just me and my whole mental model is probably wrong.
Probably, if you mean 4+ which is significantly intuitive with certain big changes. 3 and earlier was a lot more unituitive.
Note that all the industry-standard alternatives (except the relative newcomer Dorico) are much much less intuitive, and not because of the domain complexity of sheet music, but because of bad UI.
I've used it to make proper sheet music out of tabs people post on Ultimate Guitar. It was a slog to get started, but I suspect that is because of the wide range of symbols etc I needed to use. Once I had a sense for how to find things — after transcribing about a page of music — it was smooth sailing. I'd never written sheet music before using MuseScore 4.
The number of different things that different people want to do for different audiences with music that goes through a computer somewhere along the way is so great that we must admire the people who have put MuseScore together for producing any progress at all on a path that might be hoped to deliver all things for all people eventually, and all who have tried it, as I, ought to thank them for sharing it. If you find it useful, use it. If you find it somewhat useful, use it somewhat, as I do. If you find it almost useful, adapt if you can, but be patient if you decide to wait for, or persuade, the developers to deliver what will work for you. The problem space addressed is so huge, so many different kinds of sounds, scales, instruments, notations, markups, styles, media and formats, performance spaces, input devices and playback systems, personal preferences, human factors, human languages, ...
Readit News
My dad wrote an opera using LilyPond in vim, though I believe these days he's actually doing more with supercollider, which skips sheetmusic and goes right to sounds: https://supercollider.github.io/
Also as a programmer, Lilypond is programmable, and nearly infinitely customizable (similar to tex). I love that I can write some style information in a "header" and include it in every score. In my current project I am writing a lot of SATB choral songs in book format. But I also want to do a melody only edition. Just tweaking a header file, I can hide all the other voices, no problem.
(I also use Musescore, which I prefer for composition or simple projects. But when I want to output a large finished product, I always go back to Lilypond.)
There's definitely room for something that bridges the advantages of the different approaches, but it's a difficult problem and it's a $0 billion market, so having passionate people make MuseScore better is probably the best path forward.
On the editorial side, LilyPond is an incredible tool for creating modern editions of earlier music, specifically music in mensural notation. It comes with common glyphs and has full ligature support, and the separation of content and presentation means that one can reproduce the look of early manuscripts and prints while generating a modern score from the same source file. This has been really great at reducing transcription errors by easing the mental burden of transcription, transposition and note reduction, which can all be handled automatically by LilyPond. It's not perfect, but miles ahead of any alternative.
https://www.staffpad.net/
It requires a tablet and pencil/stylus, and it’s a bit pricey. But I’ve found the handwritten interface, once I got used to it, to be fairly intuitive.
- [1]: https://abcnotation.com
- [2]: https://obsidian.md
- [3]: https://github.com/abcjs-music/obsidian-plugin-abcjs
(But Lilypond is hard, harder than writing documents in LaTeX for example.)
https://abjad.github.io/
from this pursuit, I've come to realize that "sheet music" is just the only technologically available way to store the music back in the day. a practice which had an entire publishing and printing industry around it; this is how composers in the 19th century made money
BUT THEN, sound recording technology made that be somewhat obsolete, and then synthesizers and computers came and made music writing notation be less and less capable of keeping up the pace of musical technology innovation.
all in all, I'm considering that what if I say that PCM is a writing and start 'handing down music' directly as 0s and 1s in at a 16 bit depth (two bytes per sample) at a 44100 samples per second?
then again that's cumbersome, so the super collider code (and everything needed to run it) IS now the 'music notation' which includes a description of the instrument as well as the song/the music. (traditional written sheet music notation does not really include a description of the instrument beyond a reference by name)
this also "turns" the music interpreter into a digital to analog converter. lol
A composition, to varying degrees depending on the composer, leaves various degrees of freedom for the performance. So the PCM encoding of a performance is entirely different to the composition ... conceptually.
However, a good musician may be able to listen to the PCM encoding (aka "digital audio") and from that frame their own understanding of the composition, thus leading to a new performance.
So .. it's complicated.
It's very much like with kerning - once you see a sheet with decent alignment, you can't unsee it.
His video on how Sibelius' UX is a pile of shit remains one of my favourite vids on yt: https://youtu.be/dKx1wnXClcI
Have things cooled down now?
The products are completely separate
You can get it without charge from just about any and every Linux distribution.
You can get the source code without charge from ardour.org/download (or from our mirror repo on github).
You can get it (legally) without charge from anybody else who already has it.
The only thing you cannot do is to get a ready-to-run binary without charge from ardour.org itself. You can opt to pay as little as US$1 for that, however.
Also the "locking downloads behind a paywall" only applies to scores that users have uploaded to musical score social media sharing site musescore.com, but the desktop program MuseScore binaries are available for free on Win, Mac, Linux, BSDs via https://musescore.org/en/download. You can anyway post your scores anywhere for free. And there may be proprietary VSTs that you can pay to download via MuseHub, but MuseScore's baseline audio synthesis (as well as any free soundfonts you can freely download) are free.
Musescore the company is a disaster. They took my money for a Pro subscription, without delivering the actual benefits. Their support organization is, as far as I can tell, nonexistent. After six months of trying, I have yet to find a way to get them to respond to my request for support.
That said, I never noticed any input lag in the first place. Though I use an M1, not sure what the parent uses...
I have no background in authoring sheet music so my sense is that this is just me and my whole mental model is probably wrong.
Did anyone else experience this? Or did anyone else find it to be the opposite: super intuitive and smooth?
Probably, if you mean 4+ which is significantly intuitive with certain big changes. 3 and earlier was a lot more unituitive.
Note that all the industry-standard alternatives (except the relative newcomer Dorico) are much much less intuitive, and not because of the domain complexity of sheet music, but because of bad UI.
Pro tip: With your left hand centered on asdf, you can easily access the most important stuff (note pitches abcdefg and note durations 123456).