- From: Richard Kaye <R.W.Kaye@bham.ac.uk>
- Date: Sat, 20 Jan 2007 10:11:11 -0000 (GMT)
- To: R.W.Kaye@bham.ac.uk
- Cc: www-math@w3.org
> By the way, there's another embarrassment for someone, possibly > not unrelated: it seems that IE7 (with or without MathPlayer) > fails to display http://www.w3.org/ At least that's what I > discovered. I cannot explain this one either. I can now, partially. When I uninstalled MathPlayer I did not think to clear IE's disk cache. So after unistallation IE was still displaying the old cached page (which was incorrect) and not the new one. So the problem is in MathPlayer2.1beta which (at my suggestion, I shamefully add) alters the content negotiation to look for and accept application/xhtml+xml. This works fine for pages that MathPlayer alters, but when it doesn't and passes the document straight through as XHTML, the browser fails to display it properly. I am back at the point I was last year when I saw no easy resolution to the problem of serving web pages with MathML to both Firefox and IE as well as the many web-search bots simultaneously. Possible solutions are 1. Wait for IE to support XHTML (this will take too long) 2. Configure the web server to send different pages according to the user-agent (this is morally very wrong, and 99% of authors do not have this control over their web server anyway) 3. Abandon any hope that search engines should index ones pages properly 4. Redesign content negotiation and/or mime types so that XHTML+MathML can be identified as a specific documtn type to be requested/served (out of my control, and in any case there are too many potential XHTML+Something formats to have a mimetype for each) 5. Author pages for one browser only and receive massive amounts of flak from readers using some other browser. (Been there, done that .. believe me the flak *is* massive. Don't forget these readers have no technical knowledge, and cannot follow the simplest instructions on installing of configuring software, and completely deny the existence of any problem at their end.) Very fed-up... Richard
Received on Saturday, 20 January 2007 10:12:13 UTC