[Om3] Being pragmatic about the semantics of, eg, variables and functions

[Robert Miner]

Hello All.

I'm all for being pragmatic, but I think the time has come to consider
the problem at a different level entirely.  Here are the chief
considerations from my point of view:

1) By extortion and pleading, I nagged several of us into making a huge
effort to put time into MML3 work starting in January. We made some
progress, but we all fell behind on sleep and our regular jobs, and the
progress has largely stalled.  One more such convulsion may be possible
(I hope). But at the end of another such push, we have to be done, since
the window of opportunity will have shut to complete MML3.

2) The two-stage translation of qualifiers into a domain of application,
and thence into OM may not be ideal, but we know it is 
  a) as mathematically meaningful as anything else, 
  b) is most compatible with the bulk of the current text in Ch 4
  c) compatible with the large amount of XSLT David has developed over
many years
  d) backwardly compatible with MML2/OM2

3) A change to something such as Michael and James have proposed is
  a) controversial mathematically
  b) requires substantial rewrites of Ch 4
  c) would require extensive changes to David's XSLT, and hence is a
high 
     risk of running into trouble very late in the process when it
becomes
     possible (post changes) to try the proposed mappings on a large
dataset
  d) is not backward compatible, and would introduce a heavy dependency
in MML3
     on OM3, even though the latter shows no sign of being done any time
soon.

To be honest, I can hardly see a path whereby proponent of 3 could
convince me that the work such a change requires could be completed on a
timeline that would lead to successful publication of MML3.  The
weekend's mail between Chris and James and others didn't inspire
confidence in me that the debate was nearly resolved...

Consequently, I think we are left with the following "pragmatic" options
for MML3.

A) Continue with the current proposal, relying on the equivalence of
qualifiers with domain of application.  OM3 can of course extend binding
operators however that community ultimately decides.  However, the
mapping in the MML3 spec won't make use of that new mechanism.  Of
course, once it is fixed, there would be no obstacle to developing and
promoting a new mapping.

B) Allow conditions in Strict Content MathML, and push off the details
of the mapping to OM3 to OM3 itself.  That partially reversed the
decision of making Strict CMML isomorphic to OM3, and having the p2s
mapping within the MML3 spec.  But, it would allow us to go forward on a
timeline compatible with a successful MML3.  I don't personally favor
this, since it is still a good deal more work than A, and I don't think
it has a huge benefit.  But it might be more palatable to some of you.

At any rate, that is where I am at the moment.  On Thursday's call, I
believe we need a firm decision on a plan for finishing Ch 4 that
results in a date certain for a publishable draft in May.  There's just
no point in throwing effort away on anything that will take longer than
that.

--Robert