Yeah, sorry if I wasn't clear enough on this.

If a browser supports only the XSLT 1.0 spec, you have to use the 1999/XSL/Transform namespace and the version number. If your browser does not support the 1.0 spec, you'll have to use the TR/WD-xsl spec and the version is optional.

Browser / ver-------Spec supported
IE 5&5.5--------------Working draft only
IE6--------------------Working draft and 1.0
Mozilla---------------1.0 only

What is the typical browser used in Linux? Can someone test that this works? I don't think it's true that browser support isn't there. I've only seen one person say it didn't work in Mozilla, but then I found and posted the fix.

I'd also disagree that it's too complex for people to be able to edit. If one can understand HTML, one can understand XML / XSLT. At least the basic stuff that would be used to process this data. Also, if you didn't want to even look at xml, you could still always bite other people's stuff and probably be very happy.

Also, I don't think we should trash this idea based on the subject line of the thread. Wouldn't it be a good thing for it to be easily ported to other applications and work equally well on browsers?

It looks like we definitely want to have xml access, so I'll put something together as an empeg-bbs RFC. It'll specify how the calls would be made and a possible structure of the xml. Remember, that the real structure hardly matters, as the xslt would be able to interpret it however it was built. There are, of course, "better" ways to structure as compared to others.

It also appears clear that others don't want to (or have to) use the xml / xslt model and still want to customize the html. I leave it to someone else to come up with a possible spec for that.

Chris