- From: Boris Zbarsky <bzbarsky@MIT.EDU>
- Date: Wed, 18 Feb 2009 11:24:24 -0500
- To: janina@rednote.net, HTML WG <public-html@w3.org>, W3C WAI-XTECH <wai-xtech@w3.org>
Janina Sajka wrote: > Which realization raises in my mind the question, "Why? Why Canvas?" I > don't see the benefit, either for a11y or for the web ingeneral. Why canvas at all, or why canvas as the basis for a text editor? The latter is covered by the Bespin web page linked earlier in this thread. The main concern seems to be that other web technologies are not sufficiently performant, as implemented, to achieve the desired effects. The right solution would seem to be to see what can be done to fix that, possibly including adding new web technologies of some sort. That's pretty vague, but it's hard to say more without hard performance data and use case descriptions. For the former, the point of canvas is as stated in the spec: an image that can be mutated by the web page. It's no different in concept from sending data to the server in a query string and having it generate images, then mutating an image on a web page by repeatedly setting its .src to URIs with said query strings. The big difference is one of performance and usability. One would tend to use canvas for situations where one wants... an image. I've seen it used for things like fractal visualizations, Wordle-clones in javascript, that sort of thing. All are no more or less accessible per se than would be a static image. I agree that if one is using canvas as a rendering layer for an application then accessibility of that application goes out the window. The solution to that problem is to figure out why people would want to do that and provide them more accessibility-friendly ways of achieving their goals. > So, I'm of the opinion that solid value needs to be demonstrated for > canvas. I'm not generally inclined to reinvent the wheel, or our > perfectly good existing web technologies just because someone has a cool > idea. I don't thing existing web technologies do a good job of the visualization use cases I mentioned above. -Boris
Received on Wednesday, 18 February 2009 16:25:21 UTC