Minutes of 25 April 2013 SVG WG telcon

Minutes of today's meeting are here: 
http://www.w3.org/2013/04/25-svg-minutes.html
and in text below:


     Topics
         CSS 3 UI Last Call review
         hasFeature
         RDFa and SVG
         marker-segment and marker-patterns
         meeting with ePub people at the F2F
         Meeting
     Summary of Action Items

<trackbot> Date: 25 April 2013

<krit> Fore everyone: Conf room is 26631

pass code is invalid ?

<krit> Cyril: 26631

<scribe> scribe: Cyril

<scribe> scribeNick: Cyril
CSS 3 UI Last Call review

heycam: Chris emailed us recently to point out the Last Call

<ed> http://www.w3.org/TR/css3-ui/

heycam: the summary says there are new values for the cursor property, 
the nav property
... some removed because of lack of implementation

ed: which were attributes in Tiny 1.2

<heycam> http://dev.w3.org/csswg/css-ui/

heycam: perhaps they were added to the CSS as properties because of the 
SVG attributes
... it's mentionned in the feature at risk that nav-*
... are at risk
... do you think this spec needs close review by us ?
... if so, by who ?

Cyril: is there anything related to zooming and panning?

heycam: not that I'm aware
... 4.1 applies to form, views, control
... 4.2 are for form things
... it doesn't look like they apply to SVG
... Doug wanted them to apply to SVG content to have focus highlight for 
instance
... there is also the text-overflow property that we want to have

krit: I would like to review it

<scribe> ACTION: Dirk to review CSS 3 UI spec [recorded in 
http://www.w3.org/2013/04/25-svg-minutes.html#action01]

<trackbot> Created ACTION-3490 - Review CSS 3 UI spec [on Dirk Schulze - 
due 2013-05-02].

ed: is this the right version of the spec to review?
... it seems quite old, from january

heycam: I'm looking at the dev.w3 one
... I assume that's the one we should be reviewing

krit: if someone else wants to review as well, that would be better
hasFeature

<ed> http://lists.w3.org/Archives/Public/www-svg/2013Apr/0038.html

heycam: what are our current plans in SVG for feature strings
... Anne wants to unconditionally return true
... my view is that if you're using script to call hasFeature, you 
should be calling that feature directly

krit: there is the method isSupported for SVG elements

ed: never heard of that one

heycam: it seems to exist in DOM2 Core

<heycam> http://www.w3.org/TR/DOM-Level-2-Core/core.html#ID-1950641247

<birtles> 
http://www.w3.org/TR/DOM-Level-2-Core/core.html#Level-2-Core-Node-supports

heycam: there is a slight difference with hasFeature (global) and 
isSupported
... there is also a method to convert an element into another element

krit: there is also the check of each version we have: 1.0, 1.1, 1.2
... you can also search for animations ...
... feature and versions
... I have no idea how much it is used
... but it returns the right values
... at least in WebKit

heycam: in Anne's email
... there was a survey from Opera to check if the hasfeature was used
... and got very few results

krit: how did he check that?

heycam: I don't know the methodology

krit: I would trust if it was with Opera mobile
... but just Google is not sufficient

<heycam> http://lists.w3.org/Archives/Public/www-svg/2013Apr/0038.html

heycam: I think he used a large collection of documents
... from http://webdevdata.org/
... right, how many of the top 10000 web sites use SVG scripting

krit: I would be fine if no user was using it
... we could ask Google
... if they could check how much that feature is used

<heycam> krit is suggesting to instrument Blink to record how often it 
encounters has Feature

krit: I can ask to instrument Blink to do that and after 3 months, we 
can rediscuss

<scribe> ACTION: Dirk to ask on the Blink mailing list to record usages 
of that feature [recorded in 
http://www.w3.org/2013/04/25-svg-minutes.html#action02]

<trackbot> Created ACTION-3491 - Ask on the Blink mailing list to record 
usages of that feature [on Dirk Schulze - due 2013-05-02].
RDFa and SVG

Cyril: the question was raised in some ePub discussion about whether SVG 
could use RDFa

heycam: in tiny 1.2, some attributes were used but they've not been 
ported to SVG 2
... what's the status in HTML

Cyril: not sure, but I think RDFa is supported in HTML

heycam: if RDFa is an extension spec for HTML, they could do the same 
for SVG
... not sure that would require a spec for that

Cyril: there is also the difference between RDF and RDFa

ed: yes RDF is XML so you can use it with namespace
... RDFa is a lot of attributes
... I don't think they conflict with SVG attributes but ...
... SVG 1.1 and SVG 2.0 do not say anything about those attributes, but 
Tiny 1.2 does

Tav: what's the difference between RDF and RDF a

heycam: RDF is heavily namespace based, and RDFa uses attributes in no 
namespace for better HTML integration

ed: that's my understanding too

Cyril: is it what the 'a' is for ? attributes ?

heycam: no idea

krit: yes

Tav: Inkscape includes RDF

heycam: for licenses?

Tav: yes

heycam: if we did want to make it explicit that we can use RDFa in SVG 2
... I'm not sure if it should be in the SVG spec
... or in an extension spec to SVG
... they are content attributes, you can put them on any element in the 
markup

<heycam> http://www.w3.org/TR/rdfa-in-html/

<thomassmailus> Did we change the phone #?

<shepazu> (I don't think RDFa should be in SVG, or even in an SVG 
extension spec… it should be defined somehow so it "just works")

heycam: what Doug is saying is what I was thinking too
... it should just be possible, we don't want to select RDFa or microdata

ed: I agree, it should be possible to use it

<thomassmailus> Um, how do I "see topic"

heycam: we should look at how we could have RDFa in SVG in the same way 
as it is done for HTML

krit: I would like to continue discussing the marker discussion
marker-segment and marker-patterns

<krit> marker-segment: 50% url(...)

heycam: didn't we decide tjhat % on marker pattern refer to the length 
of the segment
... in my original proposal, it would take a single value
... just like marker mid
... but I don't remember for patterns
... it would make sense to refer to the whole

krit: I would like to discuss marker-segment
... in my example, would it be always at the middle ?
... if you don't specify 50% it would be 0
... start of the segment

heycam: I don't know if that's the right level of complexity
... if you wnat to put 1 marker on a segment, you would want it at the 
middle
... without the ability to put more than 1 marker on a segment, 
specifying the position is not really useful
... if you ahve 2 markers, you could have 40% and 60%

krit: I think we already agreed to have more than 1 marker

heycam: why are you asking that? to merge the properties ?

krit: I dont' know yet
... we might want to have the same syntax

heycam: make the 2 more similar

krit: maybe merge, I don't know

heycam: I'm not completely against the idea, but I'd like to see the 
complete set to see if we need to increase hte complexity or not
... it's a bit hard to see in isolation

krit: pattern is probably more useful

heycam: but some simple examples will be common cases
... and I'd like that to work
... if you want to write up a proposal and send it to the list, please do

krit: yes, I can do that

heycam: we did not settle on that last time so it'd be good to hear some 
thoughts

<heycam> https://www.w3.org/2002/09/wbs/19480/SVGTokyo2013/
meeting with ePub people at the F2F

heycam: I had suggested that some of the ePub people come and visit us 
on the monday
... it was going to be the plan
... but someone said that they could have some space on tuesday
... so it wasn't clear
... we wouldn't meet for the all day, something like tuesday PM2
... if people don't have strong opinion, we'll organize what's best for them
... I don't think it's going to be far away

Cyril: do we have any what's going to be discussed ?

heycam: how much time would it take to go there?

birtles: it's not so far, maybe 1/2 hour
... but I'm a bit concerned that it's going to be only 1 and 1/2 day for SVG
... we could ask them to come or go there only in the evenging

heycam: I wouldn't like to spend more than 1/4 of a day
... I'll reply to them and see if it's not feasible for them to come on 
monday

birtles: how many are they?

heycam: on monday probably only a few
... I'll see how late their meetings go
Meeting

birtles: because the meeting is so early, can we send a mail at least 12 
hours before

ed: was the agenda sent at the right time

birtles: yes

heycam: normally I try to send it 20 hours before
... if we collect our agenda items on the wiki, it should be feasible
Summary of Action Items
[NEW] ACTION: Dirk to ask on the Blink mailing list to record usages of 
that feature [recorded in 
http://www.w3.org/2013/04/25-svg-minutes.html#action02]
[NEW] ACTION: Dirk to review CSS 3 UI spec [recorded in 
http://www.w3.org/2013/04/25-svg-minutes.html#action01]

-- 
Cyril Concolato
Maître de Conférences/Associate Professor
Groupe Multimedia/Multimedia Group
Telecom ParisTech
46 rue Barrault
75 013 Paris, France
http://concolato.wp.mines-telecom.fr/

Received on Thursday, 25 April 2013 21:29:26 UTC