W3C

- DRAFT -

SVG Working Group Teleconference

13 Jul 2010

Agenda

See also: IRC log

Attendees

Present
ed, anthony_w, Shepazu, +33.9.53.77.aaaa
Regrets
Chair
Erik
Scribe
anthony

Contents


<trackbot> Date: 13 July 2010

<scribe> scribenick: anthony_w

<scribe> scribe: anthony

Last Call Comments Status

ED: I saw Chris sent an email stating his status on his LC comments

<ed> http://lists.w3.org/Archives/Public/public-svg-wg/2010JulSep/0027.html

ED: doesn't think there is anything there that can be closed

<ed> http://www.w3.org/Graphics/SVG/WG/track/products/24

ED: I closed off mine which was gaussian blur

<ed> http://www.w3.org/Graphics/SVG/WG/track/issues/2345

ED: Today is the final day of the Last Call period

TB: There are some problems where CSS style don't have a unit

DS: We should put that in the tracker
... or fix it

ED: We should do both
... can you raise an issue

<shepazu> http://www.w3.org/Graphics/SVG/WG/track/

ED: I'm not sure if it is a simple as adding the units
... I think there should be some explanation as well
... Does the feImage issue need changes in the spec?

TB: I need to think about if so
... may need to clarify it
... somethings need to be changed in the Inkscape implementation

ED: I have done some work on the testsuite. I've added some tests from Microsoft
... so there are some new tests to review
... I will have two weeks to finish remaining actions then I will be on holidays
... might not be able to have all comments ready before SVG Open

DS: I spoke to PLH about this
... we need a patent exclusion period as well
... we could do a pole as an alternative, but we need all the AC reps to vote
... would take some effort
... we'll almost be done by the SVG Open anyway
... will be in Proposed Rec

ED: From Chris's email the Mime Type registration will be slightly urgent

DS: Not urgent for us
... It's urgent for the group doing the registration

ED: No other updates. Doug do you have any actions on LC?

DS: I think I have one
... There was the whole idea of the security problem

ED: Probably possible to put something in
... The function that causes the problem is not part of SVG
... So I think the suggestions made by ROC are ok
... the first part is fine. The second part needs some work with the details

DS: ED, why don't you and I correspond on this
... and put something in the spec next week

ED: I think we have all the last call comments with actions on them

Catmull-Rom curves

<shepazu> http://schepers.cc/?p=243

DS: This is just a loose proposal

TB: Inkscape has spiral curves which are similar

<shepazu> http://www.levien.com/spiro/

DS: Dave Crossland suggested Spiro

<shepazu> http://schepers.cc/?p=243#spiro

DS: If you look at that link. If you compare his spiro curves and mine
... his look a lot better
... not sure how he does that
... the advantage of the Cutmull-Rom curve if you just give a set of points
... with every number corresponds to a number.
... It looks like with Spiro curves points have a different characteristic

TB: Can have arrows, squares etc
... I think Font Forge implements up to the 4th order continuity
... Inkscape up to the 2nd order

DS: It is possible to specify a syntax
... Depending on the tension between the points. It will either have a looser or tighter loop between the points
... Cutmull-Roms are not so straight forward
... with Spiro we can specify it with 3 points. x,y, and magnitude/tension

TB: In Inkscape. This is what's called a live path effect
... Inkscape has it's own name space
... it has it's own path and then it creates an SVG bezier line

DS: That's what I did. I'm looking for Cutmull-Rom curves then converting them to bezier curves
... I'm doing essentially what Inkscape is doing
... but that's not what we want to do
... when we decide on new functionality we should have the syntax supported

TB: There are two things
... In the Inkscape name space there is "d" that keeps track of the curve
... similar to SVG
... then there is a Sodipodi node type

DS: Chris seemed to like this idea
... the problem with extending path is it is no longer backwards compatible

<Tav> inkscape:original-d m 320,360 c 55.9322,38.13559 159.32203,-116.77966 200,-160 40.67796,-43.22034 160,40 160,40 l 0,120

<Tav> sodipodi

DS: if I had one of these shapes and older SVG UA would not recognise it

ED: It would up to a point

<Tav> sodipodi:nodetypes

DS: It wouldn't render the same

<Tav> sodipodi:nodetypes cscc

ED: Would need some scripted solution to make it render the same

<ed> ed: so in older UA:s at least you could count on rendering up to the point where the old UA finds an "error" in the path syntax

DS: If we add parameters and some other things, we already break backwards compatibility

<Tav> d: M 320,360 C 334.57304,317.36037 361.51706,279.02944 396.70395,250.87993 431.89083,222.73043 475.2014,204.85768 520,200 c 55.8061,-6.05126 113.60698,8.39896 160,40 l 0,120

TB: I think Inkscape would go along with it
... if you turn off the live path effect that is what you would see
... Inkscape is using the SVG syntax to store an generate the point

DS: I think there is some data missing
... can you send something to the list
... we need to figure out what the syntax would be
... but the larger question is, is this something we are interested in doing

ED: I don't know how the spiro curves are to hand author, but if they are we should take a good look at them

DS: We need to talk to Ralph Levien
... about using his method
... we need to find out if it can be a royalty free algorithm that we can put in SVG

<scribe> ACTION: Doug to Contact Ralph Levien about his algorithm for rendering Cutmull-Rom curves [recorded in http://www.w3.org/2010/07/13-svg-irc]

<trackbot> Created ACTION-2823 - Contact Ralph Levien about his algorithm for rendering Cutmull-Rom curves [on Doug Schepers - due 2010-07-20].

ED: Do these curves have any license issues related to them?

DS: I didn't find any problems with their licencing
... The arc syntax is hard to use and it would be nice to have another arc syntax that did something else
... that works on degrees or angles
... if we're going to change path we should make it as user friendly as possible
... including nerbs

SVG Front Page

ED: The news updates there is HTML code?

DS: Static HTML

<ed> http://www.w3.org/Graphics/SVG

DS: nothing interesting about the code I put up there
... someone was complaining to it on SVG dev
... then it was linked to on slashdot
... so I put something static together

ED: If you want twitter updates that is possible with client side script
... you can insert twitter updates into the page

DS: With java script?

ED: Yes

DS: That is a bit of a hack

ED: There are plenty of existing scripts that do that

DS: Not very friendly to people without java script
... and accessibility people
... because it's client side
... it something to consider definitely
... I added links to most of our specs
... going to be ongoing discussions with comms team about how they want us to do this

Blog and Twitter

DS: I sent out information for the accounts
... I encourage you guys to say stuff that is of interest to the SVG community
... I posted an interview with Microsoft (Patrick) about their implementation
... I'm interested in talking with other implementers
... so TB would you like to be interviewed?

TB: Sounds good to me

DS: I will send you some questions ED

<scribe> ACTION: Doug to Send out first round of interviews to different implementers [recorded in http://www.w3.org/2010/07/13-svg-irc]

<trackbot> Created ACTION-2824 - Send out first round of interviews to different implementers [on Doug Schepers - due 2010-07-20].

DS: Any blog posts that you guys make, it might be a good idea to aggregate them

http://lists.w3.org/Archives/Public/public-svg-wg/2010JulSep/0028.html

DS: If you open up a text editor and you look at that path

<shepazu> <path style="fill:none;stroke:red;stroke-width:1px;" d="m 320,360 c 55.9322,38.13559 159.32203,-116.77966 200,-160 40.67796,-43.22034 160,40 160,40 l 0,120"/>

TB: Inkscape keeps track of paths. If the curve is a smooth curve when you change the 3rd point it will change the point in the next curve
... if you change the curvature of one curve
... to keep it smooth you need to change it on the next one

DS: To get the visual effect that we get from the "d" which is what is rendered in browsers
... we modify it using Sodipodi node types
... There is nothing that says what node type in that example

<shepazu> http://www.levien.com/spiro/

<ed> http://en.wikipedia.org/wiki/Kochanek–Bartels_spline

Summary of Action Items

[NEW] ACTION: Doug to Contact Ralph Levien about his algorithm for rendering Cutmull-Rom curves [recorded in http://www.w3.org/2010/07/13-svg-irc]
[NEW] ACTION: Doug to Send out first round of interviews to different implementers [recorded in http://www.w3.org/2010/07/13-svg-irc]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.135 (CVS log)
$Date: 2009-03-02 03:52:20 $