Music Notation on the Web

I would like to propose a discussion in this group of what might/should
be done to standardize and promote the use of music notation on the Web.
I talked about this a bit at the recent TPAC meeting and a number of
people responded positively.  I then posted a couple notes on AC-Forum
and there was some resulting conversation.  Here is some of that
conversation (from people who have given me explicit permission to
quote):

 

Roger (in response to the announcement of work on the Audio Group
Charter): I would like to say again that it seems to me that work on
music markup would be more in the spirit of the architecture of the Web.
By this I mean that the current norm in publishing music is to put an
image of the music, often PDF but there are other possibilities, onto a
Web site as a binary image.  If you then want to USE the music in any
kind of modified form (for example, if you want to transpose it into a
different key, a VERY common use case) you need to use a scanner and
some sort of music OCR to get it into an editable form.  Let me tell you
from experience, music scanning software gives a VERY approximento
result, so in practice you have to spend literally hours on the simplest
score fixing the errors.

 

Wouldn't it be nice to be able to publish music in a way that the
CONTENT, as opposed to the formatting, could be picked up and used?
Isn't that kind of in the spirit of the Web?  And isn't it also kind of
in the spirit of the Web to worry about the content before you start
messing around with binary streams and images?

 

Charles McCathieNevile: As has been mentioned a few times, I think it
would be great to have a  

musical *notation* format for the web. Indeed, one of the most exciting


use cases I see for the audio API which kicked off this discussion is  

being able to do Web apps that interact with real music.

 

There are various initiatives for marking up music for the Web - my name


is somewhere in an ISO spec (it doesn't say so, but I think I was  

outstanding for not actually contributing anything all to the
discussion).

 

I agree with Ian though. The Audio API is a piece of work that has clear


developer interest, and work on that line is a separate issue. But I
would  

like to see people who want to lead a group working on music notation,
and  

look forward to the proposal for an XG. While I cannot make guarantees,


there are a few people in Opera who are very interested in different
kinds  

of music and might be able to participate with more regularity and  

importantly far more knowledge than me.

 

Roger:  I'm a bit disappointed that no one has commented on the central
question that I am asking - which, if you like, is a gentle objection to
a proposal for a new W3C activity.

 

Let me try again, perhaps a bit less gently.  There is certainly well
over 500 years of experience in representing the content of music, and
there are entire industries that depend on this capability.  This is not
a subject that has not been thoroughly studied and used.   This is, in
any reasonable sense of the term, a mature technology.  Sadly, the Web
is blind to the content of music.

 

Is it not consistent with the fundamental architecture of the Web to
worry about the content of the information before trying to figure out
schemes to transmit the bit streams that result from representations of
that content?  Is an audio streaming activity really appropriate in the
absence of standards for transmitting the content from which those
streams are derived?  Have I missed something really fundamental here?

 

Dan Brickley: My guess is that music markup, like geneaology, is one of
those topics

where many of us around here have a personal rather than professional

interest, and where a W3C Community group will be able to benefit from

a lot of evening/weekend enthusiasm from Web standards old-times. Not

that I'm calling anyone old! :)

 

[NOTE - There ARE companies involved with music notation and publishing,
but they are not in the W3C.  I'd really like to reach out to them and
try to involve them in this effort, and I have some ideas how to make
that attractive, or at least how NOT to make it UNattractive - Roger]

 

Dan Brickley:  I agree with the importance, but suggest perhaps MathML
is the best analogy.

 

Getting support for maths in Web content was similarly backed by

centuries of cultural heritage; but that alone wasn't enough to make

it a major priority for all browser-makers. While Javascript isn't a

magic cure-all, the flexibility and extensibility of modern browsers

(esp. given richer post-HTML5 APIs eg. canvas/svg and the upcoming

Audio work) make it much easier for 3rd parties to bolt such

functionality on, even if the browser teams are too busy with other

features. I expect much the same for music. Also btw I'm assuming the

audio APIs work will explore sound synthesis, which should help

motivate people to work with notation formats.

 

Charles McCathieNevile: 

>> I'm a bit disappointed that no one has commented on the central
question

>> that I am asking - which, if you like, is a gentle objection to a  

>> proposal for a new W3C activity.

 

Let me clarify that aspect of my comment: I see no conflict between the


API activity and Music notation - they are complementary.

 

>> Let me try again, perhaps a bit less gently.  There is certainly well


>> over 500 years of experience in representing the content of music,
and

>> there are entire industries that depend on this capability. This is

>> not a subject that has not been thoroughly studied and used. This is,

>> in any reasonable sense of the term, a mature technology.  Sadly, the

>> Web is blind to the content of music.

 

Agreed. Indeed, over a decade ago my first attempt at building an XML  

vocabulary was a language for representing music, because it seemed  

obvious to me that this was something the web should get - no musical  

representation is monolithic and universal, but I believe common
notation  

styles are as widely understoood than any written language, and across  

many different 'linguistic cultures'.

 

>> Is it not consistent with the fundamental architecture of the Web to


>> worry about the content of the information before trying to figure

>> out schemes to transmit the bit streams that result from

>> representations of that content?

 

Probably not. It is consistent with a sound architecture of the Web to  

separate the two so they can be developed independently, rather than  

depending tightly on one another. Indeed a read/write audio API would
not  

only allow generation of music from a score, but would enable clever  

people to generate a score from an audio stream (realising a high-school


physics project I did, but in a far more useful way ;) ).

 

It is also consistent with the development of the Web to work on things


where there is clear interest in "mass-market" implementation. I don't  

mean that 'if it isn't on the public web it isn't relevant' - I happen
to  

consider that an extremely short-sighted approach. But I do mean that  

willingness to invest in implementing the work of a group is an
important  

criterion for deciding whether such a group is likely to produce
something  

tangible, or a recommendation for what other people should do that will
be  

made without the significant real-world testing that a good web standard


requires.

 

>> Is an audio streaming activity really appropriate in the absence of

>> standards for transmitting the content from which those streams are

>> derived?

 

Sure. The two are, and should be, orthogonal.

 

> Have I missed something really fundamental here?

 

People are implementing audio APIs now. That means it makes sense to  

standardise, so we get one common one and can build things that reach
the  

world, not just the google-verse or the mozilla-verse or the
opera-verse.

 

If the music publishing industry, or the browser industry, or some other


significant segment, were to put there money where our desires are and  

start seriously working on this, it would make sense to do it. Until
then,  

it is a part of the full potential of the web that is not yet realised,


and W3C would be premature in trying to drag us there without real-world


support.

 

As I did a decade ago, I still hope that this will in fact happen. But  

like trying to create regulations for Nuclear Fusion-based power plants,


or personal petroleum refineries, or a standard sign-language notation
for  

the Web that can be used to drive avatars, I think today we are still
too  

far ahead of the demand curve to begin this work. I'd be happy for that
to  

change tomorrow, but it takes more than good ideas.

 

Noah Mendelssohn:

> Isn't this discussion just one aspect of a more general topic -- the
trend in the W3C away from declarative markup and towards an API-driven
Web?

 

No, in this case I think it's more fundamental, because the API and the 

markup in question are dealing with completely different levels of 

information.  The API is for waveform samples at, say, 44KHz.  The
markup 

is for conveying musical scores, typically with features like key 

signatures, notes with specific durations, rests, etc.