Minutes: MathML intent meeting, 29 April

Attendees:

David Carlisle

Sam Dooley

David Farmer

Patrick Ion

Paul Libbrecht

Louis Maher

Murray Sargent

Neil Soiffer

Laurence Zaysser

Moritz Schulbot

Deyan Ginev

Bruce Miller

Steve Noble

Charles LaPierre

Thanks to Louis for once again taking notes.

Regrets:

Agenda:


1.  Announcements/updates

NS: will send out information on Zacum for taking W3C minutes.

2. How to make the

 list of known semantics
<https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdocs.google.com%2Fspreadsheets%2Fd%2F1EsWou1K5nxBdLPvQapdoA9h-s8lg_qjn8fJH64g9izQ%2Fedit%23gid%3D0&data=04%7C01%7C%7C1a8f7e03a7f64b4ab09a08d90a0ed410%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637551881066443761%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=GmTICiK29MKW4X6BXCrWpmK6isuiMCp64zLMR46MNLk%3D&reserved=0>


table easy to implement

LM: LM will work with DG to make DG’s table accessible for A.T.

LN: can not access this table online.  He must download and let Excel alter
it.  Perhaps this is due to the use of pivot tables.

We discussed how easy it should be for people to alter this table. This
table could be accidentally damaged.   The table is backed-up.  However, if
an error were introduced into the table, and was not noticed for months,
then the original content of the table might not be recoverable.

Level one is for items not in the original MathML specification.  It could
be in an appendix to the specification. Level three will be open-ended and
not part of the spec.  The spec can grow too large such that people will
not implement it.

NS: The MathML primary goal is accessibility. The secondary goal is
searchability and perhaps computability.  You  should be able to feed
MathML into a program doing computations, and it should work.

SN: Perhaps we should supply use-cases to demonstrate what we mean by
computability.

NS: We must produce documents like the specification and use cases.  We
also should produce documents explaining why the specification is being
written.

We discussed having a long form and a short form for speaking something
that is repeated throughout a document. For example, if you had a character
spoken as “big O” and if it would be repeated, should you later pronounce
it as just “O”?

LZ: This would be too complicated for A.T. to implement.  It would not be
accepted.

DF:  O(n) could mean the orthogonal group or the asymptotic class of
functions.  Intent should specify which.  Pronunciation may be a different
issue.

LM: From an A.T. point of view, the default behavior should always be to
read the expression as it is.  The A.T. listener would never know when the
speech change was due to using a short form of speaking, or be caused by
encountering a new character. It should be remembered that the A.T.
listener may not be able to see the ambiguous character; and, therefore,
could not determine for themselves what the correct interpretation is.

NS: A.T. does  not see CSS information. It just sees the MathML. You may
not be able to globally change speech behavior with CSS technology.

NS: Should we say that the A.T. must look at CSS as well as MathML? (this
is often not the case or not doable thus far)

The next meeting will be a WG meeting on May 13. It is expected that
everyone will be on board.

Those who need to be on the WG, and do not have an organization to sponsor
them, should send the W3C a request to be an invited expert.

The link is:

https://www.w3.org/participate/invited-experts/