[LAU] Web-based Notation

[Justin Smith]

On Thu, Mar 19, 2009 at 2:13 PM, andy baxter
<andy at earthsong.free-online.co.uk> wrote:
> David Baron wrote:
>> Folk seeking notation alternatives my check this out!
>>
>> http://www.noteflight.com
>>
> Does it have a midi import feature? I am trying to help out a friend who
> wants to be able to print out music that he has played on the keyboard
> in notation form, partly as a way of learning notation for himself, and
> partly as a way of communicating with other musicians who like to see
> things in this form. (He is a good pianist but mostly self taught and
> doesn't work in this way normally).
>
> I've tried most of the programs available for linux, and none of them
> really do what he wants. The main problems are that:
> - he wants the scores in piano stave notation, with two staves, not just
> on a single stave. This is quite difficult to get working from what I've
> tried so far.
> - the stuff he plays is quite dense and poly/multi/non/rhythmical, and
> programs like lilypond that I've tried seem to get indigestion with it.
> This might be a problem with rosegarden's lilypond export feature though.
>
> Any help with this from you or anyone else on the list would be much
> appreciated, as I've been trying to find something that works for over a
> month now with no luck.
>
> Cheers,
>
> andy
> _______________________________________________
> Linux-audio-user mailing list
> Linux-audio-user at lists.linuxaudio.org
> http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user
>

For dense / polyrythmic / nonrythmic stuff, automatically generated
notation is going to be extremely unhelpful. Unless there have been
huge improvements in midi to score AI, the results are just going to
be messy.

Think about how much work the program would have to do, figuring out
if you are playing shorter notes, of if the tempo is increasing,
whether you just changed key signitures or are throwing in a bunch of
accidentals, etc. etc. Just like with image recognition, there are
many things that are immediately obvious to a human listener that
computers historically have much trouble with.