- From: Charles McCathieNevile <charles@w3.org>
- Date: Thu, 4 Jan 2001 07:49:38 -0500 (EST)
- To: Kynn Bartlett <kynn-edapta@idyllmtn.com>
- cc: Al Gilman <asgilman@iamdigex.net>, "'w3c-wai-gl@w3.org'" <w3c-wai-gl@w3.org>
Hmm, maybe the answer should be that ChaalzToolz (Pat Pend, (tm), (r), <g>, <nrc>, etc...) uses an architecture that sets the size of an image according to the font size in the page (I have a very clever person thinking about this now, and if they figure it out I hope they tell us all including me <grin/>). In fact I am more likely to include SVG in a page via the use of namespaces. Then I can specify the size of the SVG in terms of em or ex, and it will figure it out for itself. In old browsers and XML browsers without specific SVG support it will produce text versions. This will work in Amaya as soon as it determines the user space from the viewbox (correct behaviour - there is currently a bug or two in teh SVG support, which is still being developed). An alternative that should work is to include the SVG without specifying a size, and then let the SVG inherit its size from teh user stylesheet generated by the browser. But this currently still has implementation issues - I will take them up with the developers of the Adobe plugin at least, and have just been talking about it to the Batik Team (which is the CSIRO and Koala/Jackaroo projects' combined effort). Cheers Charles McCN On Thu, 4 Jan 2001, Kynn Bartlett wrote: At 05:45 PM 1/3/2001 , Al Gilman wrote: >At 08:20 AM 2001-01-03 -0500, Bailey, Bruce wrote: > > If the SVG is on an html page, then the size is fixed by the height and > > width of the embed or object tags. That's standard html behavior for images, plugins, Java applets, etc. >AG:: >So, how should this behave instead? From a user's point of view, I think it's bizarre that a user should need to maintain separate settings for font size based on a distinction that they are unable to grok. Here's an example: (a) The Kynn-O-Matic (tm) web editing program home page is made entirely with XHTML + CSS and no textual images. (There are some screenshots and illustrations, however.) Suzanne comes to the page and changes the size of her text. The "navigation buttons" increase in size the same amount as the text of the page. (b) Next, Suzanne goes to look at the Sean-In-A-Box software's page. That one is built using graphical text -- gif files as buttons. Suzanne also changes her font size and the text on the page increases, but the gif files do not. (c) Finally, Suzanne checks out the ChaalzToolz home page. This one is made using SVG, and fortunately, Suzanne installed an SVG plugin a few months back. She changes the size of the text...and like case (b), the navigation buttons do not change, but the body text changes. Suzanne is confused. She doesn't know SVG from GIF from CSS, nor should she be expected to. Suzanne simply wants to be able to read the navigation buttons. She _definitely_ won't know that she can right-click on Chaalz's navigation buttons and adjust the size (but not the viewport size) -- how on Earth would she know that, and know when to expect it? She can't do it with GIFs or JPEGs and those are far more common. SVG should be able to read the browser's default font size, and if the SVG is built properly (using relative, not absolute font sizes), it should scale appropriately. Now here's a dumb question for you -- can you set the size of a image, plugin, or applet in _ems_? And if you do, and you change your font size, will those "viewports" resize themselves magically? I admit I haven't ever tested this, nor even thought about it, really. Maybe it's a dumb idea, or maybe it's a cool idea. I don't know. If it works, it might be a technique for SVG and other scalable graphics formats. --Kynn -- Charles McCathieNevile mailto:charles@w3.org phone: +61 (0) 409 134 136 W3C Web Accessibility Initiative http://www.w3.org/WAI Location: I-cubed, 110 Victoria Street, Carlton VIC 3053, Australia until 6 January 2001 at: W3C INRIA, 2004 Route des Lucioles, BP 93, 06902 Sophia Antipolis Cedex, France
Received on Thursday, 4 January 2001 07:49:41 UTC