- From: Bailey Bruce <Bailey@Access-Board.gov>
- Date: Fri, 11 May 2007 11:16:43 -0400
- To: "Christophe Strobbe" <christophe.strobbe@esat.kuleuven.be>
- Cc: <w3c-wai-gl@w3.org>
Sorry Christophe to be so harsh here... > I hope this helps. If there is a point to this exercise, it has to be that one *knows* that an HTML version is available. So I am not sure how productive working from Google key word hits will be. > I tested this with PDF, but scenarios with other file types are also conceivable. For the sake of our experiments, I concur that we are try to find HTML versions from PDFs. It is only for the sake of argument that we are making the fairly presumption assumption that the PDFs are not accessible, and the equally outrageous leap of faith that the HTML version are. > turns up the document at > http://ieeexplore.ieee.org/iel5/9716/30679/01421586.pdf >From the title I found this page, which makes me think that this document is available *only* as PDF. http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=1421586 Since I am not an IEEE member, your URL redirected me to a logon page. Is this an example of the same technology we would like to have direct people to accessible versions? http://ieeexplore.ieee.org/Xplore/login.jsp?url=/iel5/9716/30679/0142158 6.pdf > The same search also turns up the document at > http://www.iovs.org/cgi/reprint/48/5/2383.pdf Same problem. > I eventually got to the page at > http://www.bitstream.com/font_rendering/products/dev_fonts/sb_cjk.html, > which contains a link to the above PDF file, but not to an HTML version. Right, which implies pretty strongly than an HTML version does *not* exist. > A Google search for "monotype chronicles" in the domain "www.monotypefonts.com" only finds the PDF version. The URL hacking and search methods all assume than an HTML version exists. If an accessible version does not exist, then the site fails. The question is about how reasonable it is to get to the accessible version from the inaccessible alternate version. The question presumes an accessible version exists, so these failures are not so helpful for invalidating the approach. Likewise, the one success does not prove the techniques acceptable.
Received on Friday, 11 May 2007 15:14:37 UTC