W3C home > Mailing lists > Public > public-fx@w3.org > July to September 2010

Re: How does the svg element handle CSS border and background-color?

From: Doug Schepers <schepers@w3.org>
Date: Wed, 25 Aug 2010 13:13:22 -0400
Message-ID: <4C754F32.5050101@w3.org>
To: public-fx@w3.org
Hi, Nico-

Domenico Strazzullo wrote (on 8/25/10 6:52 AM):
>
> I suppose the related thread is “Should the base svg tag receive
> events?”. In that thread several arguments have been exposed. As I
> understand it, two distinct and correlated points emerged:
> 1) registration of pointer events on the outermost <svg>, or other <svg>
> or <g>.

Naturally, pointer events ('mousedown', 'click', 'mousemove', etc.) can 
be registered on container elements.  However, they will only be 
dispatched on the graphical children of those containers.  This was the 
design from the beginning of SVG, and I believe it is honored by every 
user agent.  The exception seems to have arisen, because of confusion 
and possibly ambiguous language in the SVG spec, around the root <svg> 
element, which in some user agents is allowed to dispatch events as if 
it were a graphical element.

This issue is unresolved.  Many people, myself included, have stated 
opinions, but we haven't yet discussed it as a group, and don't yet have 
consensus.


> 2) the eligibility of the above elements to assume the status of
> event.target, otherwise expressed in the DOM specification as event's
> target.

In DOM3 Events, I express this as the "proximal event target"... in 
other words, the element on which an event is dispatched.


> The inconsistency of the svg specification on the subject, as well as
> some inappropriate language and wording, have been pointed out with
> multiple arguments, the least of which not being the contradicting
> interference with other specifications, the DOM in particular.

We will try to correct the inconsistency at our next SVG F2F in Paris, 
just after SVG Open.  We'd be happy to talk with you more about the 
subject then; maybe you could even join us for part of our SVG WG F2F to 
discuss the issue.  As an experienced content author, it would helpful 
to have you there.


> Questions and requests for confirmation on some sensitive points have
> not been acknowledged. By not acknowledging, that doesn't make the
> *crushing* evidence of some of the arguments less crushing.

I mentioned this earlier, but you may have missed it.  The SVG WG is 
presently on hiatus (because of various vacations, implementation 
crunches building up to product release, presentation and demo 
development, conference travel, and other disruptions) until SVG Open. 
So, there's only so much bandwidth we have to respond.

<note>
Just as a clarifying note: even when we are active, the SVG WG may be 
slow to respond to comments on the list.  It's not because we don't 
value the feedback, is that there are only a few of us, and we are all 
doing many other things.

Just to bring this into perspective...
In addition to helping manage the SVG WG, I do the same for the very 
large WebApps WG; I monitor many SVG community lists, exchange quite a 
lot of private SVG emails, act as a sort of developer outreach person as 
the activity lead for our Rich Web Clients activity, and deal a lot with 
W3C members' issues; I write charters and try to help foster new work 
and new vendor participation (I'm trying to get web education, audio 
read-write, and touch interface activities going at W3C right now); I 
review a lot of specifications, and I'm editing at least 4 
specifications right now, including the rather demanding DOM3 Events, 
which we are trying to bring to Last Call; I also try to write blog and 
twitter posts to bring the larger community into the conversation of 
creating and maintaining SVG; and I've been speaking at several 
conferences to try to spread SVG goodness.

Among the things that Chris Lilley does in addition to serving as the 
walking SVG history library, he is Team Contact for the CSS WG and the 
new WebFonts WG (new groups often take a lot of energy); he's also 
involved in many other duties I neither know about nor understand; and 
he's a father, which is, as I understand, not a small investment of time.

Chris and I each have several long teleconferences per week, and have 
W3C Team duties around improving W3C that we are currently involved in.

The rest of the group consists largely of implementers, who are busy 
trying to make SVG implementations, and do many other things at their 
own companies.

Each of us in the SVG WG also do too much SVG WG administrivia that is 
necessary but not very visible, file bugs, write tests, edit the specs, etc.

And I'm not complaining!  I like my job.  But I hope you understand that 
we can't always give as much attention to every task as we would like.

There's this idea of "context switching" [1], which is a computer term 
for something I'm sure has a psychology equivalent.  It takes time and 
resources to switch from one task to another, to remember the details, 
to get into the right frame of mind, and to put aside the other tasks 
you're doing.  In the midst of everyday multitasking, a request for an 
immediate response on an issue is an "interrupt", and there are 
sometimes that an interrupt is disruptive to the other active threads.

You are all busy too, I know, creating content or working on 
implementations (or even editing specs, poor bastards)... I'm sure you 
could list an equally improbable list of chainsaws you're juggling.  You 
have devoted time to this issue, and you want the closure of having it 
resolved; there may be some urgency to have it done in time to meet 
deadlines (like browser shipping schedules).  I get that; honestly, I 
do!  I'm often on that side of the equation.

But like I've said before on this thread, the SVG WG will be discussing 
the issue soon, taking into account the contents of these threads, and 
we will come back to the community with our resolutions.  If there is 
some problem at that time, we'll discuss it and work to resolve it.

I'm asking, in general, for a little patience on this and other issues.
</note>

So, all that said, I hope you understand that we simply aren't prepared 
to answer your question right now.  I hope to have you in on the 
discussion in person, where you can clarify some of your points (I'll be 
honest, I found it hard to follow some of your email... maybe the 
language barrier), and we can come to a conclusion that we all feel 
meets the needs

[1a] http://en.wikipedia.org/wiki/Context_switch

Regards-
-Doug Schepers
W3C Team Contact, SVG and WebApps WGs
Received on Wednesday, 25 August 2010 17:13:27 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 25 August 2010 17:13:28 GMT