Re: [orca-list] Orca and maths

[Joanmarie Diggs <joanmarie diggs gmail com>]

Hi Michael.

> other maths symbols (such as integral) would be little work, but how
> much work would it be to get orca to support the maths formatting? 

If it's just a matter of adding the symbols, it's super easy.  To be
honest, when I was updating chnames.py, I initially included a bunch of
symbols from the U+2200 table (a.k.a. math symbols, including integral).
The end result was a fairly large file that would be a huge amount of
work for the GNOME translators (much-appreciated volunteers) to
translate. Given the reality that most users don't need those symbols,
the decision was made not to include them all.

That of course leaves the very real issue of what to do for users who
*do* need access to all of those symbols -- or to other unicode
characters we have not included.  I have a couple of ideas on this
front:

1. We need to make it easier for the average user to add symbols to the
dictionary of character names, edit their "pronunciation," and change
their punctuation level.  I filed an RFE based on your original comments
here: http://bugzilla.gnome.org/show_bug.cgi?id=431531 I need to add
some additional thoughts/ideas I've since had, which I'll do in a bit.
You are encouraged to comment on the bug with your own suggestions.
That's what RFEs are all about, after all:  The community designing
their screen reader so that it does exactly what they want and need it
to do.

2. I want to start the creation of unicode tables for Orca.  In other
words take a table like the U+2200 table and generate a file that
contains the stuff Orca needs to add those characters to its dictionary
and to know when it should pronounce those characters.  Such an
orca-ized table could then be placed in the user's .orca directory if
that user needs access to those characters.  In other words, you'd get
all of your math symbols without having to manually add them all
yourself.  As the tables get generated, we could put them on the wiki
for download.  For non-English users, there is still the issue of who is
going to translate these Orca-ized tables for them.  I am hoping that
the community can contribute in this regard: by identifying unicode
tables that are in their language so that we can orca-ize them, and by
providing their own translated files which could be shared via the wiki.

This would not happen over night of course, although perhaps an English
version of the U+2200 table could.  <smile>  

As for MathML, that I would have to try.  The above solution would
definitely cause symbols that you could arrow to in Firefox to be
spoken. I'll generate a MathML file later, see what happens, and let you
know.  If it doesn't work, I'll add an RFE for it.  We should speak
MathML.

> would it be possible to get orca to switch to a more verbose mode in the
> maths environments (as I don't like too much punctuation spoken when
> reading a document, but obviously need all the extra symbols in maths).

This too is easy.  You just specify that the symbols in question should
be spoken at punctuation level "none."  The table of math symbols I
generate will do that.

> How could I help out in getting such support for maths in orca (it

By reviewing the math table I generate and correcting it. <grin>  My
interest in math symbols is not math-based.  I have no background in
mathematics whatsoever.  My interest -- and concern -- is as an
instructor of individuals who are blind.  So my plan is to take the
unicode.org table, combine it with the very little math skills I do
have, and create a list of character names that seem to make sense.  I
will undoubtedly get some of them wrong.

Take care.
Joanie