W3C home > Mailing lists > Public > www-svg@w3.org > February 2006

Re: [SVGMobile12] SVGT12-207: event "alias"es

From: Ian Hickson <ian@hixie.ch>
Date: Mon, 20 Feb 2006 01:57:13 +0000 (UTC)
To: Jon Ferraiolo <jonf@adobe.com>
Cc: www-svg@w3.org
Message-ID: <Pine.LNX.4.62.0602200145480.28514@dhalsim.dreamhost.com>

On Sun, 19 Feb 2006, Jon Ferraiolo wrote:
> We agree that the term "alias" is insufficiently defined and have now
> added an explanation for the term in the specification:
> --------------
> The event aliases described in this section allow the use of an
> alternate event type when registering an event listener. For example, it
> is possible to register an event listener for the resize event (i.e.,
> {"http://www.w3.org/2001/xml-events", "resize"}) using the alias
> {"http://www.w3.org/2001/xml-events", "SVGResize"}.
> --------------

Given that this new text has no normative content, this does not in any 
way resolve my issue. (The text above consists of a statement of fact -- 
something allows something else, without saying how -- and an example, 
neither of which are testable normative conformance criteria that 
implementors could use to actually do anything with.)

Please make the spec normatively state what the above means.

If it is the intent of the specification to make, as Bjoern asked in a 
separate e-mail, the following two statements to be equivalent:

   element.addEventListener('resize', handler, false);
   element.addEventListener('SVGResize', handler, false);

...then that is absolutely not acceptable to me. The SVG spec cannot 
change the behaviour of DOM Events implementations, that is highly 
inappropriate. Nor would scoping such a change to the SVG spec make any 
sense given the existence of mixed SVG/HTML/MathML/XUL/etc user agents 
that have a single DOM Events implementation and a single scripting model.

Even allowing for the extremely poor wording of the proposed text above, 
it would _still_ leaves unanswered a number of questions about what the 
model being described is, e.g.:

   * is the UA to fire a 'resize' event or an 'SVGResize' event,
   * what happens when an author fires the other event,
   * what happens when events of those types bubble from non-SVG content,
   * what happens when events of those types bubble into non-SVG content,

...and so forth.

Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'
Received on Monday, 20 February 2006 01:57:22 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 8 March 2017 09:47:07 UTC