That is very helpful, Moira. Kat's original query was:
> If the glyph for No. (as outlined in Wikipedia: > http://en.wikipedia.org/wiki/No.) is used, should this be in an > abbreviation element to explain it? It is an abbreviation, isn't it?? > > What do screen-readers make of this particular glyph, if anything? > > Or should it be kept as No., which is quite common, and wrapped into an > abbreviation element with a class of contraction, and a title of number? In theory, all that's needed to represent No. is to use № - the HTML Entity (decimal) for the Unicode Numero symbol. This gives all browsers, screen readers, search engines and various bits of software every chance to interpret the information correctly. Screenreaders can clearly be set to make decisions about what to say (as in Moira's O circumflex example), so can be instructed by their publishers to pronounce this as "number". You shouldn't need to wrap an ABBREV around it, which is good because while it /is/ an abbreviation of the word "numero", it isn't an abbreviation of the concept to be conveyed, ie the word "number". It also means there should be no confusion about "x" as "times", "by", "unknown", "Roman numeral ten" or the letter "x" as each has its own Unicode and HTML Entity. Sighted people can judge the meaning by the context while screenreaders can pronounce the appropriate meaning. However. At least one widely used screenreader says nothing when confronted by the Unicode for "numero". It also isn't supported by all font sets, although the most widely used seem to include it. Google doesn't seem to translate HTML entities at all, treating them simply as character strings. It may therefore be practical to wrap the Numero sign in an element that describes it. The ABBREV element seems to be the best candidate. So the answers for Kat are yes, no, probably nothing and possibly. Ricky > > I'd still like to know if using character entity references and/or > > unicode for symbols and special characters will actually convey web > > content more clearly to people who use screen readers. > > Sometimes, sometimes not. I have JAWS 6.1 and 7.0 installed - > both know about some Unicode characters and not others. JAWS > reads a short name for the character, not usually the full > description from the standards, e.g. for U+00D4 (Ô) > JAWS says O circumflex (not "Latin capital letter o with circumflex"). > > & some more examples... > > for U+00D7 (×) JAWS says times > for U+2665 (♥) JAWS says nothing. > for U+2116 (the numero sign №) JAWS says nothing for U+2122 > (™) JAWS says trademark > > JAWS Unicode support was meant to have improved significantly > in version 7.0 and might be better again in 8.0, but lots of > people who use screen readers don't have the most recent > version. Hope this helps, > > Moira Clunie > Accessible Formats Developer > Royal New Zealand Foundation of the Blind Awhina House, 4 > Maunsell Road, Newmarket, Auckland Private Bag 99941, > Newmarket, Auckland DDI +64 9 355 6938 Fax +64 9 355 6960 > [EMAIL PROTECTED] ******************************************************************* List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm Help: [EMAIL PROTECTED] *******************************************************************
