Re: Proposed CG agenda changes

Hi Joe,

Thanks for your thoughtful reply.  I'm glad that we're all in agreement on
the eventual goal of a future notation standard with an excellent
specification, and now that I understand the larger context better, I can
support not attempting a high-quality specification for MusicXML 3.1.

That being the case, however, could you say a few more words about what the
short-term agenda item

> 	Build an initial MusicXML specification

means?  What would such a specification consist of?  What would the criteria
for success or completion be?  What would we want to learn from the process?
I would really like to see some discussion of those questions before
starting, but perhaps you're already working on proposed answers and I'm
jumping the gun.  However, if you'd like a straw man, here's one:

The "initial MusicXML specification" should just be the existing commented
DTD, with work limited to these tasks:

* Minimal alterations required for thoughtful SMuFL integration.

* Correction of typos and other outright errors.

* Additions that do not involve any changes or additions to the existing set
of elements or attributes -- i.e., limited to additions of new alternatives
for attributes or CDATA values currently chosen from fixed sets.

* Identification, in detail, of those specificational issues that would have
to be addressed if one wanted to produce a high-quality specification (one
meeting my four criteria).  (A simple example: "Specify what elements, if
any, can intervene between the notes in a chord.")  No effort should be
devoted to proposing design changes that might mitigate any issue.

I think the last of these is worth doing because I believe it will be very,
very helpful in identifying problematic areas for future design
consideration, without taking positions on how to fix them.  That is the
response to my "learning" question.

						L Peter Deutsch

Received on Thursday, 12 November 2015 09:07:41 UTC