RE: Versioning and html[5]

Maciej Stachowiak [mailto:mjs@apple.com] wrote:
>Indeed, even OOXML is widely derided as being a "standard" that only
>Microsoft can ever implement, apparently by design. I don't think it
>helps Chris's case to hold it up as a success story.

Indeed, I did not introduce OOXML (or office documents in general) as an example here.  Office document formats, AFAIK, started as proprietary formats, and there is now demand for them to be openly spec'ed, so Office did so.  You shouldn't be surprised that such a spec, for a version-12 product that's been around for 18 years, with a long history of being backward-compatible, has a bunch of foibles and would be hard for a newcomer to fully implement.

_I_ am not the one claiming that we must "make HTML match IE's current behavior exactly".  That claim is being made by others, but at the same time the WHATWG HTML5 spec is not clearly following that goal (and let's be realistic, we're not talking just about HTML behavior).  I'm not trying to move the WG away from that goal either - I'm just saying, you can't be half-pregnant.  If you really want to match IE7, then go for it.  I don't want the WG to have any delusions that we're going to fix any behavior other than crashes or security issues without being told by document/app authors that's what they want, is all.

I personally think the best thing for the standard is to try to be back-compatible with current content, but understand that IE is going to have to offer opt-ins until we can get away from all the bad behavior we have today - and therefore, clearly identifying new HTML is in everyone's best interest.  That doesn't change the "evolve the current web" tenet of this WG.

-Chris

Received on Friday, 13 April 2007 21:59:01 UTC