brainstorming -- SVG

I wasn't able to join in on yesterday's IRC, and will be out of town
lots this week and next week. But I did see Don Connolly say something
on the IRC log expressing his disappointment that people were holding back during the brainstorming. This message is, then, I suppose, blamable on him? <humor type='questionnable'/>

1.  SVG - there just has to be a way for HTML and SVG to coexist in the 
same document space without nasty problems. HTML and VML coexist quite 
happily (albeit in one browser only). Not understanding all the gears 
and torque and hydrodynamics, our inability to sprinkle SVG into HTML 
inline consistently across browsers, seems rather silly to me. 


2. The fact that SVG inside <object> only works in two of the three major browsers that current support SVG (IE,FF,OP)is annoying. The workarounds to make it happen inside <frame> are cumbersome. <embed> is currently the only workable cross-browser solution. (Doug Schepers tells me he has it working in <iframe>, and without resorting to histrionics, but I have yet to see his examples.)


3.  Even if <object> support for SVG becomes unbroken in the future, an SVG butterfly cannot fly over an HTML rose without dragging its clumsy rectangular cocoon with it.


4.  I'm not a fan of complicated namespace stuff. It reminds me of IBM 
360 JCL. Not that I really had anything against JCL, but it seemed like 
magical incantation and I neither understood it nor wanted to. I'm 
willing to be persuaded that it cannot be avoided, but... does it really have to be so gnarly?


5.  SMIL is a good technology and while it seems to have been dismissed by WHATWG, the debate seems to have been foreshortened, perhaps prematurely, in my mind. There are a good number of things the HTML world would benefit from. Why wait for HTML6?


6. Much of the specification of WHATWG's <canvas> is redundant with and syntactically inconsistent with SVG. Some of what is there needs to be there for the sake of both HTML and SVG -- the ability to read and write pixel values. Ahhhh.... if we can all work and play nicely together, the flowers will be so fragrant!


7.  If <canvas> is entangled in a patent, as it seems to be, my unlawyerly reading of the W3C patent statement
http://www.w3.org/Consortium/Patent-Policy-20040205/#sec-Licensing <http://rockmail.sru.edu/exchweb/bin/redir.asp?URL=http://www.w3.org/Consortium/Patent-Policy-20040205/%23sec-Licensing>  
suggests that the W3C cannot approve it, if an alternative royalty free technology exists. 


This is really just one idea, in case any one is counting....

David Dailey

Received on Friday, 23 March 2007 04:06:39 UTC