- From: Yvette P. Hoitink <y.p.hoitink@heritas.nl>
- Date: Thu, 4 Dec 2003 14:15:57 +0100
- To: 'Claus Thøgersen' <thoeg@get2net.dk>, <w3c-wai-gl@w3.org>
Hi Claus, In the examples I gave in my article (http://www.heritas.nl/wcag/language.html) which examples do you think would need to be marked up? How can we give clear guidelines so a developer knows which foreign words have to be marked as such? My personal estimate is that the Shell website would be OK as it is (page marked up as Dutch even though there are some English words) where the LogicaCMG website would need a lot of markup to identify English texts. Marking up foreign sentences would fall into the 'must do' category for me, whereas marking up occasional foreign words would be a 'should do' or even just 'could do' depending on how easy it is to recognize the word when pronounced in the main language of the text. "Sitemap", even when pronounced in Dutch, can be easily recognized. "Email" on the other hand, becomes a different word with a different meaning when pronounced in Dutch (thanks to Ineke for the example). There will always be cases where there just isn't a solution within the current HTML specs, even if you want to go out of your way to mark up every foreign word. Cadeaushoppen is an example of this: the word 'cadeau' is a Dutch gallicism, 'shoppen' is a Dutch anglicism but the combination will probably never be in the official Dutch word list. It would be wonderful if the HTML standard could incorporate a mechanism to phonetically describe how to pronounce such words, but that's just blue sky thinking. Working within the current standards, we can only attach one language to each word. Also, could you please clarify what you mean by contracted braille? I am not familiar with this term but would deduce that it involves language-specific abbreviations to speed up braille reading, is that a correct assumption? Yvette Hoitink CEO Heritas, Enschede, The Netherlands E-mail: y.p.hoitink@heritas.nl > -----Original Message----- > From: w3c-wai-gl-request@w3.org > [mailto:w3c-wai-gl-request@w3.org] On Behalf Of Claus Thøgersen > Sent: donderdag 4 december 2003 13:45 > To: w3c-wai-gl@w3.org > Subject: the need for marking up change of language > > > Hi, > > Screen readers like JAWS with multilingual synthesizers like > Eloquence are beginning to be able to swich language on the > fly. In JAWS it has worked since before version 4, or it mayh > have been added in version 4, and with the release of JFW 5 > in september JAWS kan now identify the language if it is > marked up. The question is not tto markup every word that may > originate from other languages e.g. English but to markup a > block of text written in a another language, and then hope > that the screen reader has an installed synthesizer that can > speak this language, and you never know on the devvelopper > site of the webpage but that is life! > > Also remember that for braille devices the same applies for > reading contracted braille. >
Received on Thursday, 4 December 2003 08:16:11 UTC