- From: Al Gilman <asgilman@iamdigex.net>
- Date: Sat, 15 Sep 2001 08:32:30 -0400
- To: love26@gorge.net (William Loughborough), "Jonathan Chetwynd" <j.chetwynd@btinternet.com>, <wai-tech-comments@w3.org>, <w3c-wai-gl@w3.org>, "Sean B. Palmer" <sean@mysterylights.com>
At 07:55 AM 2001-09-15 , William Loughborough wrote: >At 09:34 AM 9/15/01 +0100, Jonathan Chetwynd wrote: >>The semantic meaning or concept underlying images is just as transferable, >>as texts are translate-able. > >Care when using "the" is urged. Yes, it is possible for text to have >differing "semantics" in various situations but with images it is >inevitable. A matter of degree and probabilities. AG:: There is no one way about this. Understanding vs. misunderstanding is determined by comparing the divergence in interpretation with the precision requirements of the use to which the data are being put. Confusion over date formats have killed people in the coding of radiological devices. So text is just as misunderstandable as is an image. In fact, pictures have been added to drivers's licences in the U.S. in order to make the documents more secure: it is harder to misunderstand a comparison of the picture with the person than comparing, say, signatures. I believe that we can, however, say that what the XMLGL says is fact based on a more grubby, mechanistic analysis. The conventional .gif and .jpg representations used for images on the Web are not compatible with widely-deployed repurposing technology that does a reasonable job of recreating intelligibility the way that UniCode text is. This is just a statement of the state of the art in repurposing and media formats. There is a more mathematical theory that argues that text is more removed from experience, more highly encoded, and hence it is natural that it should be more repurposable on the whole, but we don't need to go there. If the test of repurposability is that it can be done so by techniques that it is reasonable to consider the user has available in their client, then the preponderance of image representations are more specific, less repurposable, than are words in text. This is not to say that the objective is not symmetrical. Text is not logically the foundation, experience is. But more indirect representation by means of encoded verbalisations about experience turns out to be more repurposable in the field today that a more direct emulation of visual sensort experience. Al >-- >Love. >EACH UN-INDEXED/ANNOTATED WEB POSTING WE MAKE IS TESTAMENT TO OUR HYPOCRISY >
Received on Saturday, 15 September 2001 08:30:39 UTC