- From: Ian Hickson <ian@hixie.ch>
- Date: Thu, 8 Feb 2001 18:44:46 -0800 (Pacific Standard Time)
- To: Nic Ferrier <nferrier@tapsellferrier.co.uk>
- cc: <www-style@w3.org>
On Tue, 6 Feb 2001, Nic Ferrier wrote: > > I don't want to start a flame war (like in August) but I'd like to ask > some questions about the current acitivities for adding behavioural > specification to styling. Just the word 'behaviour' is enough to start a flame war! :-) > As far as I understand it there are 2 main proposals: > > - CAS > - CSS behaviours (CSS3) commonly: BECSS > > Is this right? or has CAS not been accepted for review? There are three contenders (as opposed to proposals): CAS, HTC, and XBL. CAS are a NOTE on the W3C site. HTCs are described on microsoft.com and implemented in IE. XBL is a recent addition to the race and is described on mozilla.org and partially implemented in Netscape 6. The BECSS are a hybrid of HTCs and CAS (the only contenders at the time BECSS was being written). > Is there any current work on integration of behaviours with the XForms > work? Some work on this recently restarted. > Now some thoughts of my own: the reason I'm asking these questions is > because I'm about to start work on a web browser and I hope to support > behaviours using the selector syntax in one form or another (I'm not > religious about which proposal I use, they both look like they could > do the job adequately). You should also look at XBL. However it is important to note that no behaviour proposal has been given the W3C stamp of approval. All of these technologies are proprietary (including the one developed by a free software project). As a browser maker you basically have three options right now. 1. Pick one of the existing proprietary solutions, either HTCs, as implemented by Microsoft, CAS, not implemented by anyone, or XBL, as partially implemented by mozilla.org. If you pick a solution developed by another organisation, then you gain the benefit of an existing userbase (especially in the HTC case), the downside of locking yourself to following a specific vendor, and the risk that the specifications will change (especially likely in the XBL case). Of course, the technical merit of each spec should be considered too (I won't give my opinion since I am obviously biased...). 2. Wait until the W3C agree on a spec for behaviours. This could be anything from a few months from now to a few years. There is reason to believe that it will happen `soon'. But the track record on this issue is not good (talks breaking down, etc). 3. Invent your own. Here you have full control over the situation, but of course unless you have a secret weapon to increase your market share dramatically, it is unlikely that your solution will see widespread adoption. Microsoft and mozilla.org have picked option 3. Opera Software and Konqueror, so far, appear to be picking option 2. My advice is to concentrate on adding support for the existing, published W3C, ECMA and IETF specs (HTTP 1.1, XML1, XML namespaces, CSS2, PNG, MNG, MathML, DOM2, and ECMAScript, for instance) and only then to start considering adding the more esoteric capabilities like element behaviours and bindings. But what do I know... -- Ian Hickson )\ _. - ._.) fL Netscape, Standards Compliance QA /. `- ' ( `--' +1 650 937 6593 `- , ) - > ) \ irc.mozilla.org:Hixie _________________________ (.' \) (.' -' __________
Received on Thursday, 8 February 2001 21:44:59 UTC