4 ms·

I know absolutely nothing about music notation software (and nearly nothing about making music) but wouldn’t the easiest way of getting notes into any computer

by ugh 16y ago

I know absolutely nothing about music notation software (and nearly nothing about making music) but wouldn’t the easiest way of getting notes into any computer be to connect a midi keyboard and to start playing? I mean, those are dedicated devices, tailored for just that use. They were created for that purpose. I guess you might be unable to play a keyboard at speed but you could still enter the notes without duration with a keyboard (that could be a two step process — press the note and press some key for the duration afterwards). It seems absurd to me that you would enter notes with the normal keyboard or the mouse or by tapping. What am I missing?

spiralganglion 16y ago
A whole bunch of things. First, background: I've recorded over 24 albums, written 5 scores, play over a dozen instruments, and have composed thousands of pages of music in various notational programs using a ton of different input methods. Ok, let's go. A midi keyboard takes up space. It also takes up more space the better it gets. If you want a 3 or 5 or 7 octave keyboard, you need to have that sort of room in your work area. Imagine an entire piano keyboard on your desk. Even a smaller keyboard takes up more space than another regular computer keyboard, and having a second one of those is difficult enough to work with. Space issues aside, there's the issue of workflows. I've yet to find a music notation program with an anything approaching an elegant flow. For each passage of music that you'd like to write, there are a ton of variables that need to be expressed and recorded, and the midi keyboard only handles a few of these. You need to capture pitches of multiple notes (chords), durations, accents, articulations, voicing (for multiple parts on the same staff, as in choral music), dynamics, and many other things. Many of these are optional, and only used when needed, but even the most fundamental ones are often a chore to enter. A midi keyboard is great for entering note pitches, but as you described, playing in time (at speed) seems to be the only way to avoid entering all the note-lengths by hand. Playing in time has its own problems, too. You have to match the data to a grid structure. But the grid structure needs to be strict enough to catch the subtleties of the performance, and flexible enough to compensate for human-error. Otherwise, you'll be back to manual data-entry to correct all the parsing mistakes. These constraints also depend largely on the person playing the keyboard. On the other hand, an alternating two-step process isn't any faster with a midi keyboard than a computer keyboard. Midi keyboards were not created for data entry. They were created for live performance, controlling other dedicated devices that created music, like synthesizers. They've been co-opted to enter data for notation, and while they can accelerate the process somewhat, there are shortcomings on the software side. It is also largely dependent on the type of music being transcribed. I am very much in favour of an iPad notation program. I've been thinking about that since last summer, but I haven't had time to make one myself. If someone tackles this problem and comes up with an elegant solution, they'll have no noteworthy competition on either side of the PC/Post-PC divide. Pun intended.

by ugh 16y ago

spiralganglion 16y ago

A whole bunch of things. First, background: I've recorded over 24 albums, written 5 scores, play over a dozen instruments, and have composed thousands of pages of music in various notational programs using a ton of different input methods. Ok, let's go. A midi keyboard takes up space. It also takes up more space the better it gets. If you want a 3 or 5 or 7 octave keyboard, you need to have that sort of room in your work area. Imagine an entire piano keyboard on your desk. Even a smaller keyboard takes up more space than another regular computer keyboard, and having a second one of those is difficult enough to work with. Space issues aside, there's the issue of workflows. I've yet to find a music notation program with an anything approaching an elegant flow. For each passage of music that you'd like to write, there are a ton of variables that need to be expressed and recorded, and the midi keyboard only handles a few of these. You need to capture pitches of multiple notes (chords), durations, accents, articulations, voicing (for multiple parts on the same staff, as in choral music), dynamics, and many other things. Many of these are optional, and only used when needed, but even the most fundamental ones are often a chore to enter. A midi keyboard is great for entering note pitches, but as you described, playing in time (at speed) seems to be the only way to avoid entering all the note-lengths by hand. Playing in time has its own problems, too. You have to match the data to a grid structure. But the grid structure needs to be strict enough to catch the subtleties of the performance, and flexible enough to compensate for human-error. Otherwise, you'll be back to manual data-entry to correct all the parsing mistakes. These constraints also depend largely on the person playing the keyboard. On the other hand, an alternating two-step process isn't any faster with a midi keyboard than a computer keyboard. Midi keyboards were not created for data entry. They were created for live performance, controlling other dedicated devices that created music, like synthesizers. They've been co-opted to enter data for notation, and while they can accelerate the process somewhat, there are shortcomings on the software side. It is also largely dependent on the type of music being transcribed. I am very much in favour of an iPad notation program. I've been thinking about that since last summer, but I haven't had time to make one myself. If someone tackles this problem and comes up with an elegant solution, they'll have no noteworthy competition on either side of the PC/Post-PC divide. Pun intended.