W3C home > Mailing lists > Public > www-svg@w3.org > November 2013

SVG fonts: [ was RE: minutes, SVG WG TPAC F2F 2014, day 2]

From: David Dailey <ddailey@zoominternet.net>
Date: Fri, 15 Nov 2013 06:39:42 -0500
To: "'Cameron McCormack'" <cam@mcc.id.au>, <www-svg@w3.org>
Message-ID: <006e01cee1f7$602285d0$20679170$@net>
ChrisL   
 ... I propose that we get rid of SVG Fonts completely from SVG
    2

    cyril: what is problematic with removing this is making them on
    the fly

    ChrisL: have you seen roc's font workshop page?

    cyril: no...

Am not sure if the font workshop page resolves Cyril's question or not. Does it?

Losing the ability to define new fonts on the fly, client-side would present a major crippling of current functionality.

See part 5 of the instructions concerning http://cs.sru.edu/~ddailey/Parisien.html 

If svg-in-opentype gives some mechanism for providing for client-side generation of fonts that include color, gradients, animation, patterns, (namely the whole ball of wax that SVG fonts provides) then I'm cool, but I think the resolution to drop them may have been premature given that one person voiced a question and another asked for a link, implying that the above question of loss of functionality may not have been clearly answered prior to the call for a vote. Some folks will be very unhappy with a withdrawal of major functionality from SVG1.1 to SVG2 . Of course the oral discussion that ensued may have resolved the issue satisfactorily.

Btw -- It was me who mentioned to Doug about superpath maybe not offering the solution to drawing efficiency, since some hardware vendors (I think like NVidia) are implementing (and presumably caching) SVG paths in the GPU. 

I will check into that further to see if I heard right, but at the same time there are other savings offered from superpath: 1. Conservation of overall document size for complex shared edges (as in maps) 2. Ability to animate shared borders (as in the partition of Poland 1772-1939) or 3. Ability to animate unshared parts of paths -- making Mickey Mouse's arm move without having to redraw (and stipulate) the entire d attribute. Just the d attribute of the arm subpath of the superpath could be animated, saving lots of compute cycles.  Nonetheless I will see if I got my facts straight on the above.

Cheers
David


-----Original Message-----
From: Cameron McCormack [mailto:cam@mcc.id.au]
[snip]
Received on Friday, 15 November 2013 11:40:19 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 8 March 2017 09:47:34 UTC