W3C home > Mailing lists > Public > www-dom@w3.org > October to December 2010

Re: [DOM Level 3 Events] MouseEvent.getCoordsAt(element) [ISSUE-136]

From: Doug Schepers <schepers@w3.org>
Date: Tue, 05 Oct 2010 22:06:31 -0400
Message-ID: <4CABD9A7.50509@w3.org>
To: Jacob Rossi <jrossi@microsoft.com>
CC: Jonathan Watt <jwatt@jwatt.org>, "www-dom@w3.org" <www-dom@w3.org>
Hi, Jacob-

Jacob Rossi wrote (on 10/5/10 5:28 PM):
> Jonathan Watt wrote:
>> On 05/10/2010 21:04, Jacob Rossi wrote:
>>>
>>> seeing as you have demonstrated a simple workaround and D3Events is in
>>> Last Call, I think this might better be suited for a future
>>> spec--especially if it might better be implemented as an API for more
>>> than just SVG elements. Thoughts?
>>
>> The API I proposed isn't intended to be SVG specific, that just happens
>> to be what I demoed it with.
>
> I believe similar workarounds can be achieved for non-SVG elements. The
> existence of relatively simple workaroundsgives us the flexibility of
> wrapping up DOM 3 Events without this API.

I don't agree that the workaround is simple.  It's short, true, but it's 
rather complicated and obscure.  It is one of the chief pain points 
people report in working with SVG, and with transforms being added to 
CSS, even more people will bump up against this very common problem.


> I agree such an API would be nice to have, I just don’t think it’s
> necessary that we work it into D3Ebut rather a future spec.

I'm inclined to add this to DOM3 Events.  With common problems, the 
solutions should be built into implementations, especially when it is 
relatively easy to implement and relatively hard for authors to work out 
or discover; otherwise, the web platform will be too brittle.

I'd rather have this in place before transforms are in widespread use. 
It would be more cumbersome to specify this later, when people have 
already had to work around it... then they have to do feature detection 
and include the javascript fix even where the current browsers support 
the getCoordsAt() method.  If the fix were not well-defined, that would 
be one thing... it might be good to let the best solution emerge from 
the community; but in this case, we already know the problem, and the 
solution, from years of development in the SVG community.


Regards-
-Doug Schepers
W3C Team Contact, SVG and WebApps WGs
Received on Wednesday, 6 October 2010 03:41:15 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 22 June 2012 06:14:06 GMT