Standards - WTF?

Daniel Glazman reports on an IRC discussion where someone complained about the work he did in the HTML Overlays proposal. Now there’s lots wrong with the proposal, which I’ll come onto later, but the complaint seemed to be that this trivial bit of script somehow undermined the W3C like the WHAT-WG and the XFN folk. As people know I really don’t think the WHAT-WG are doing useful work, (see my other posts) but they’re right to be working on HTML 4, they should, and I believe could be doing it within the W3C if it wasn’t for the way they’re going about things, but that’s a seperate matter.

The XFN and this proposal though, I’ve seen no-one complain about on undermining the W3C simply because they’re not! They both work in areas the W3C don’t go near in HTML, XFN because representing human relationships in HTML isn’t something that needs standardising, and HTML Overlays - because it’s a joke technology that cannot degrade.

The HTML Overlay proposal has many problems, it doesn’t degrade, this isn’t just a problem with the implementation (which is poor: little error protection, syntax errors in all non ES3 browsers, sync http requests in javascript, universal browser read popup boxes) but it’s actually impossible to do useful degradable replaced content, since it removes almost all the benefits, you have to end up putting the fallback to the replaced content into the main page - so you might as well have put it all in there.

Daniel says there are 2 sorts of non-JS users - those with genuine accessibility or similar needs, and those who disable javascript because they believe it’s the worst invention of all time (He was quoting Bert Bos well out of context here). Which is ridiculous, there are other very important categories - the non-PC device (and there are no non-PC devices that have the xml httprequest object) and the corporate security policy. The second of these especially is a market you cannot ignore - lots of people spend more time accessing the net at work, than at home, if the corporate security policy blocks activexcontrols - and they generally do - I’ve yet to work on a corporate targetted site that specced activex control access from script as a required feature, it simply wouldn’t be allowed.

The problem with the HTML Overlays proposal is not that it’s not from the W3C, it’s just that it’s really not very good, or even original - but then most of the stuff we see isn’t actually original, it’s just a rehash of another idea, so we shouldn’t criticise people for that.

Comments

  1. karl Says:

    And nobody, nobody has said that HTMLOverlay was wrong technically speaking. As you said… nothing to worry there.