MuseScore vs Lilypond for Alternative Notation Development
I recently heard a podcast featuring Thomas Bonte discussing MuseScore. In it he mentioned LilyPond, indicating some communication and perhaps borrowing between the two projects.
Is there any relationship between the MuseScore and Lilypond organizations?
As a participant the Music Notation Project [musicnotation.org] I've been working on alternatives to Conventional Music Notation, and on software to produce such notations.
Lilypond and Musescore, being open platforms, are both candidate bases for such development.
Are there reasons we should prefer to develop for MuseScore, or not?
Ought we to expect some kind of "merger" or compatibility in the future?
Joe Austin
DrTechDaddy.com
Comments
Hey Joe,
The MuseScore team has no plans to work on alternative music notation formats. Focus remains on CMN. That said, if a developer demonstrates the capabilities and passion to implement an alternative notation format with a considerable market adoption like Jianpu, we'd be happy to support this effort. In the end, MuseScore is open source under GPLv2 so feel free to fork it.
As for MuseScore's connection with Lilypond, there are two things which currently connect us. One is the use of the Emmentaler/Feta font in MuseScore which is originally created by the Lilypond founders. The other one is the open source development approach. Besides that, as both softwares are following a total different software architecture design, there isn't much overlap between the two projects in terms of code sharing. We like to support Lilypond as they are an ally with us to bring open source software to the market, and meet up with members of the the Lilypond community every now and then.
I hope this answers your questions.
Maybe my comment is a bit "Off Topic", but I can't do without stating that while I *love* the appearance of the scores generated by Lilypond, I find the textual approach a tough one indeed, particularly when dealing with complex scores. The strength of Lilypond, in my humble opinion, lies in its ability to control each and every detail with great freedom and precision. But the costs of achieving that high degree of control is huge in terms of "learning curve"! This is where the visual approach of MuseScore comes to the rescue! :)
As a result of what I just stated, as a teacher I find MuseScore is a much better solution and I suggest it to my students (most of them love it as much as I do, and can learn to use it with good results in a matter of a few weeks).
Should I be a professional publisher, my opinion COULD be different.
In reply to OT? by Aldo
You should try lilypond with denemo. Furthermore I think musicxml is a good evolution for a standard music notation. Also the way lilypond makes it's reserved words is to complex. a Staff for instance should always have the same tag "new Staff", with a TYPE- attribute, instead of having for each type of Staff another tag...
Why I still use lilypond when I make music ( as an amateur), is the total freedom at input-time (using denemo or text). I already saw on the musescore forum that a lot of musicians have the same problem with musescore as I do. When I am creating some music, I first put in the pitch-line, without considering the rythm (even though I already have it in mind). Once that is done, I start putting the rythm as I have it in mind. All this is done without measures or all the stuff that keeps me from going on with my song in musescore. The point is that in denemo (based on lilypond), I do have this TOTAL FREEDOM to go on with my music. So, if you tell me that musescore is better, I say "yes, it is for some ways, but lilypond is for all ways".
Please programmers of musescore, when I change the length of a note in the middle of my score, do not replace it with a rest, but shift in the notes on that staff (or give me the choice to replace or shift), because shifting is what I (and many other musicians) intend to do. I read something like hiding bars and measuring, but found out that this is not the solution to the (frustrating) problem. I can't think of any technical obstacle to implement a total free input (where I can use my ears and intuition ), and keeping the check for the end. Replacing with rests is like taking over my music and preventing me from writing it the way I want it to be.
In reply to You should try lilypond with by Jowannes
You already have the choice to shift - that's what cut and paste accomplish.
The problem with trying to do this automatically is that there is no way MuseScore can no how many notes you want to shift - only you know this. Based on the way you describe your method, you want literally all notes to the end of the score moved, but that would be almost unhrewad of in normal use - more often people would want only the next note, or all notes to the end of the measure, or to end of a phrase of some sort, to be moved. And very often, when people change a duration, they don't want *anything* to move. Since only you know how many notes you personally to move, the best way to do this is with cut and paste.
That said, the next major version of MuseScore *will* feature a mode that allows rhythms to be specified after pitches. But it can't read minds, so it's still not going to be exqactly what everyone wants all the time - that;s an impossibility.
In reply to You already have the choice by Marc Sabatella
To be brief : is it not far simpler to shift in at the place I want, and insert any measure farther a rest if necessary ? Everybody pleased. I'm speaking out of practice ( I mean this is what I every time encounter when I use musescore ( and as I can read in the forum, others with me). Maybe a user should be able to choose the way he wants the input to behave? Denemo also has the nice feature of splitting/merging measures by hand. I am a programmer and was always opposed to programs which take the user by the hand while he is still concentrating on his job, constantly intervening (read interrupting) the user with annoying messages or prohibiting him to arrange things, even at a time when the score in his totality is unknown . I realy like the way denemo gives me that possibility to be busy with my business. Having said this, if I am exposing my point of view over here, then it means that I do believe in the way musescore is evolving.
I don't understand what you mean about cut/paste? "Cut" simply replaces the cutted note with a rest, it does'nt shift... am I missing something? ( I'm using musescore 2.0.3)
In reply to To be brief : is it not far by Jowannes
Yes, as I said, for your particular usage, having everything else to the right shift automatically makes sense, but as I explained, this is *not* how most people use notation software, so it is *not* how most people would want things to shift. More often, they would want only one note, or just a few, to move, if any. So no matter how we do it, it is exactly what some people want at some some times but worse than nothing for everyone else the rest of the time. There simply is no pleasing everyone - no one behavior is what everyone wants all the time.
Cut and paste work to mvoe things just as they do in any other program. That is, "cut" does *not* just replace notes with rests; it also copies those notes to the clipboard - this is the standard meaning of "cut". And you can then click some other spot and "paste" the contents of the clipboard there, thus achieving the effect of shifting the notes to the new position.
BTW, you can split and jopin measures by hand in MuseScore as well - see the Edit / Measures menu.
In reply to Yes, as I said, for your by Marc Sabatella
I feel I have to second that from my user's perspective. Such an input mode would drive me nuts.
Let me ask you a question: If we had no score writer and you'd have to compose using paper and pencil like Beethoven: How would you manage to compose with your work sequence? Using the eraser all the time? Or writing your pitch sequence, then write it on a fresh sheet *again*, this time in rhythm?
I say Musescore behaves more like paper and pencil than your proposed mode of input. Meaning at least to me that it is "intuitive".
In reply to Yes, as I said, for your by Marc Sabatella
I feel I have to second that from my user's perspective. Such an input mode would drive me nuts.
Let me ask you a question: If we had no score writer and you'd have to compose using paper and pencil like Beethoven had to: How would you manage to compose with your work sequence? Using the eraser all the time? Or writing your pitch sequence, then write it on a fresh sheet *again*, this time in rhythm?
I say Musescore behaves more like paper and pencil than your proposed mode of input. Meaning at least to me that it is "intuitive".
In reply to I feel I have to second that by azumbrunn
Do you realy think Beethoven did not use an eraser?
If it was my intention to write down music as they did a few centuries ago, I definitely would use paper and pencil.
To have the measures and note-lengths on paper exactly as I have them in mind, often requires more time and patience from me. And no, it is not my intention to rewrite the score, but to be able to model it with a good helping hand from my ... software.
In reply to Yes, as I said, for your by Marc Sabatella
'Cut' in a normal editor does not replace te 'cutted slice' with spaces, but shifts in. Musescore does replace cutted notes with rests. And yes, the cutted elements are placed on the clipboard and I can paste them somewhere else (but this is not the issue)
Lilypond gives the possibility to shift in (or replace with something else) on the staff itself or on the group of staffes : marvelous!
Indeed, I can merge and split ( nice!), but the cut-problem stays the same.
In reply to 'Cut' in a normal editor does by Jowannes
Cut, then paste, is the key thing.
In reply to Cut, then paste, is the key by Isaac Weiss
I've been trying to explain something for nothing. Till yet none of you gave a straight answer to the problem : Deleting or cutting a note replaces it with a rest instead of shifting in. Whatever you explained, it is not the issue. Can any guy having the same problem, please stand up and explain the problem out of another point a view? And please, leave Beethoven out of it, try Chopin.
In reply to I've been trying to explain by Jowannes
I am not understanding your question. As we have explained, of course cutting a note (or region containing multiple notes) replaces with silecne - aka rests. But if you then immediately do a paste operation, the notes you just removed are then placed at the rewquested location. The effect is *exactly* as if they had simply shifted - no difference at all. And htis is exactly how cut and paste works in most other programs - not just notation programs, but computer programs in general. I am not sure how to make this more clear. Perhaps pictures would help.
Original:
After cut:
After paste:
As you can see, cut and paster works exactly as in any other program to shift the selected region to the desired location (in this case, to the beginning of the measure, but I could just as easily have chosen ny other location).
In reply to I am not understanding your by Marc Sabatella
As my final reply, here is the problem I encounter. This is the way it works for me ( as an example I take the 3 first measures of the score 'Annemarieke':
1. first I note the pitches without considering time, measures ... etcetera (because the way I type it, I hear the correct timing)
2. I try to fill in the right note-lengths : (doesn't work (works in denemo) )
![f2]( "f2")
this is what it becomes in denemo (without complaining):
![f3]( "f3")
3. Having the right accents in mind, I find the right measure, which is 6/8 with an up-measure
![f4]( "f4")
4. Having this, while playing what I already have in the score, I can hear in my head the left hand piano pitches and type this line on the left hand staff, again without considering time, measures and so on... :
![f5]( "f5")
5. Now I have to give each note the right length and synchronize the left and right hand staffs,... this is the result ( the rests are messing up my score, (trying to keep both staffs in line??) :
![f6]( "f6")
This is what denemo does for me :
Conclusion : Why am I telling you this ? Because musescore has other pro's .
Using the best of both (musescore/musicXML - denemo/lilypond) importing trough midi is at this moment my best solution.
I'm curious: do you actually have statistics on how "most people" use music notation software?
And is that because most music notation software does not support using it in any other way?
I use software most ofter to create a "personalized arrangement" of a published score,
typically a piece that I am learning. For example:
1. I change the arrangement, or the layout.
I simplify the arrangement to match my performance capability,
or change the key, or replace repeats with duplicate measures,
or match the arrangement the director is using.
2. I convert "round notes" to "shape notes" (for singing)
3. I like to change the layout so lines break where lyrics (phrases) break,
and notes on the same beat in successive lines are vertically aligned,
rather than squeezing as many notes as possible onto a page.
BTW, few software systems make this easy, especially if the phrase breaks in the middle of a bar. It often involves creating invisible bar lines and time signatures which typically (but not in lilypond) messes up the measure numbers.
So basically I'm transcribing scores.
Since I'm not very proficient at the keyboard, if I enter notes via playing them (MIDI)
the timing comes out all wrong and I need to adjust it. What I really want, after all pitches have been entered, is to specify the duration (or measure and count) for each note, with a minimum of clicks or keystrokes, and have the system take care of the rest. Cut-and-Paste is quite tedious in this case.
Specifying the exact time for individual notes used to be possible in early Cakewalk versions by editing the MIDI file view, but later versions seemed to drop MIDI support.
Now I realize I may not be "most people", but perhaps more people would use notation software if it were more usable. I don't think "pencil and paper" convenience should be the target, but the baseline from which we seek significant improvement. The goal should be to make it as easy to create and edit a score as it is to create and edit a text document with a good text editor.
In reply to I'm curious: do you actually by DtTechDaddy
You have the same problem as I do (thank you fate). What I forgot to mention all the time till now, is that I'm working with piano ( == grouped staffs).
Since all of you are doing the effort to brainstorm this thing, I myself will also make an example that exposes the problem. The example of cut/paste does clarify possibilities but the issue still remains.
For the example I'll take the arrangement of 'Annemarieke' .
I'll be back... ;o)
In reply to I'm curious: do you actually by DtTechDaddy
Nothing you describe in terms of your use case seems unusual. What I don't understand is why you don't simply enter the durations while entering the notes, rather than entering them with the wrong durations and then change them later? It obviously isn't easier to do it your way. If you enter durations as you go, you will find it *is* as easy to create and edit a score as to use a text editor.
If for some reason you must enter notes without durations and then figure out durations later, probably better to use a MIDI seuqnecer and not a notation program at first. Then once the durations are the way you want, simply export to a standard MIDI file and import the results into notation software. Probably one could also use a text-based system like LilyPond or ABC as part of this same type of workflow.
BTW, I don't know that anyone has stats on what "most people" currently do, and obviously no one knows what people would do if things were different. But as I said, absolutely nothing you describe about *what* you are trying to do sounds at all unusual, except for *how* you have chosen to go about it.
Also, note that splitting and joining measures is very easy in MuseScore - see Edit / Measures - and you can make the measure numebrs do whatever you want via Measure Properties.
BUMP