From: William F. Hammond <hammond@csc.albany.edu>

Date: 18 Apr 2002 15:05:06 -0400

To: www-math@w3.org

Message-ID: <i7r8lcg74t.fsf@pluto.math.albany.edu>

Date: 18 Apr 2002 15:05:06 -0400

To: www-math@w3.org

Message-ID: <i7r8lcg74t.fsf@pluto.math.albany.edu>

Paul Topping <PaulT@dessci.com> writes: > > And how accessible would they be to people who rely on text readers? > > Not very. However, if someone went through the resulting page adding ALT > attributes with some kind of text for the equation (e.g. "x^2 + 3"), that > might help. I've not tried to do this, though. Clearly, if the math is very > complicated it isn't going to work well. I'm not sure what "text readers" means. If it means a web browser such as "lynx" that performs through vt100-level terminal connections, I've not seen much done with MathML in that class of browsers so far. Given the terminal condition, however, it does strike me as sensible that such a browser undertaking to render MathML might reasonably do so with pseudo TeX, using braces to remove ambiguity. Another approach would be "ascii art" of the type used in a terminal interface for a computer algebra program. If such a browser handles CSS, then a user might consider providing an early-in-the-cascade style sheet, and content MathML might be more easily handled that way than presentation MathML since the former markup is closer (both on the author side and on the reader side) to meaningful human mathematical thought. -- BillReceived on Thursday, 18 April 2002 15:05:11 GMT

*
This archive was generated by hypermail 2.2.0+W3C-0.50
: Saturday, 20 February 2010 06:12:51 GMT
*