- From: Charles McCathieNevile <charles@w3.org>
- Date: Tue, 30 Jul 2002 08:57:20 -0400 (EDT)
- To: jonathan chetwynd <j.chetwynd@btinternet.com>
- cc: David Woolley <david@djwhome.demon.co.uk>, <w3c-wai-ig@w3.org>
Yes, my example is not really optimised at all. The big things are that I didnt't bother to compress the SVG, and I don't think the original jpegs that it incorporates were clipped to just the area that the clipped part comes from. In a production-quality tool both of these should be fairly eaasy to do, so if you have 10 5k jpeg images that you want to use you should be able to get the SVG into another 4 or 5 k (zipped - but SVG browsers can handle zipped SVG as a rule) including paths and animation of the sprites. (no, I haven't tried celinea yet. It looks good, and I have been looking for a simple tool like this for a while, but I don't have a windows system to run it on at the moment.) Cheers Chaals On Tue, 30 Jul 2002, jonathan chetwynd wrote: >charles, > >not sure where you got the file size from. > >My estimate goes something like this 50k / 10 images = 5k >however svg will have problems fitting into this, unless the file size is >~1k which does seem feasible >your example is something nearer 60k which is fine for a single image, but >may not work so well for sprites, have you thought this thru? >or perhaps youre not too concerned for 56k users? > >did you give celinea a try out? >I can't get it to function at all, but that may well be me >http://www.celinea.com/ > >jonathan > > >----- Original Message ----- >From: "Charles McCathieNevile" <charles@w3.org> >To: "jonathan chetwynd" <j.chetwynd@btinternet.com> >Cc: "David Woolley" <david@djwhome.demon.co.uk>; <w3c-wai-ig@w3.org> >Sent: Tuesday, July 30, 2002 1:14 PM >Subject: Re: do vector graphics enhance our concept of self? > > >> Well, jpeg itself doesn't have transparency, nor allow animation of parts >of >> the picture. So it isn't going to help you, and rather than trying to >rebuild >> jpeg from the ground up for technical reasons it seems smarter to use SVG >> (the specification requires support of jpeg, so you don't lose anything >that >> jpeg already provided). In fact if you want to convert an SVG to a jpeg >(and >> lose the transparency and other neat features - I'll try to make an >animated >> example) there is a tool available as part of the Batik toolkit. >> >> I think you should be moaning about the fact that your browser doesn't yet >> ship with SVG. (Some browsers do, of course). >> >> (It is possible in some versions of HTML to have a plugin downloaded >> automatically, but unfortunately you don't get to choose which of the >systems >> that can handle a particular format is installed. Fine for most people, >but >> it means that someone can't easily choose a more suitable plugin for their >> needs. Further development - see http://www.w3.org/TR/CX if you're >> fascinated) >> >> Cheers >> >> Chaals >> >> On Tue, 30 Jul 2002, jonathan chetwynd wrote: >> >> >Chaals >> > >> >Well it does look impressive, >> >so I'm stuck moaning about the failure of an easy way to degrade to jpeg. >> > >> >thanks again >> > >> >----- Original Message ----- >> >From: "Charles McCathieNevile" <charles@w3.org> >> >To: "jonathan chetwynd" <j.chetwynd@btinternet.com> >> >Cc: "David Woolley" <david@djwhome.demon.co.uk>; <w3c-wai-ig@w3.org> >> >Sent: Tuesday, July 30, 2002 12:48 PM >> >Subject: Re: do vector graphics enhance our concept of self? >> > >> > >> >> I have attached an SVG image. The file size is a bit bigger than the >> >listed >> >> 3.2k, because it downloads two jpeg images. >> >> >> >> But it shows photo-realistic transparent jpegs which can be animated to >> >move >> >> around like sprites. >> >> >> >> This was generated by Jim Ley's annotation tool - you load up a jpeg >and >> >> trace around the bit you want, then say who it is. >> >> >> >> I found it by looking through his search tool - >> >> http://www.jibbering.com/rdf/foafwho.html >> >> >> >> The one modification I made to the code was to change the opacity >value - >> >it >> >> had opacity="1" for both photos, and for one I made it opacity="0.5" (1 >> >means >> >> not at all transparent, 0 means completely transparent). It is not >> >difficult >> >> to adapts Nick's tool to do this by default, or as an option. >> >> >> >> What this shows is that with SVG you can easily select a piece of a >jpeg >> >you >> >> want, and use it as a photo-realistic transparent animated sprite. It >adds >> >> around 1-2k per image. It works in the Adobe SVG plugin (available for >> >many >> >> browsers - I used it in iCab on MacOSX but it works in Explorer, >Netscape, >> >> and compatible browsers. I haven't tested this yet in other substantial >> >SVG >> >> browsers such as Batik but expect it to work fine). >> >> >> >> Cheers >> >> >> >> Chaals >> >> >> >> On Tue, 30 Jul 2002, jonathan chetwynd wrote: >> >> >> >> > >> >> >Unfortunately, given our obsession with identity, if SVG is not >capable >> >of >> >> >photorealistic representation, in a small file size, then we might >then >> >seem >> >> >to need a (jpeg like?) high compression format, with transparency, and >> >> >capable of scaling. Alternatively if it were capable, perhaps a >camera, >> >or >> >> >at least a conversion tool would be extremely popular. >> >> > >> >> >In terms of images, bandwidth and accessibility, this may well be one >of >> >the >> >> >critical areas for development in the near future. >> >> >many images are just not accessible. >> >> > >> >> >jonathan >> >> > >> >> > >> >> > >> >> > >> >> >----- Original Message ----- >> >> >From: "David Woolley" <david@djwhome.demon.co.uk> >> >> >To: <w3c-wai-ig@w3.org> >> >> >Sent: Monday, July 29, 2002 10:33 PM >> >> >Subject: Re: do vector graphics enhance our concept of self? >> >> > >> >> > >> >> >> >> >> >> > Can anyone point to realistic vector portraits, with a small file >> >size? >> >> >> >> >> >> That's really a research topic for low bandwidth telephones, >although >> >> >> the games people may also have something, but I suspect it is >> >> >> proprietory and may still require a texture map, as JPEG etc. >> >> >> >> >> >> It seems to me, though that this mailing list is not a good place >for >> >> >> asking the question. If you are lucky, someone will know, but you >> >> >> are not targetting the question well. >> >> >> >> >> >> I'd try the BT Research web site, as they were interested in this >sort >> >> >> of thing at one time. TV broadcasters may be as well. >> >> >> >> >> >> >> >> > >> >> > >> >> >> >> -- >> >> Charles McCathieNevile http://www.w3.org/People/Charles phone: +61 >409 >> >134 136 >> >> W3C Web Accessibility Initiative http://www.w3.org/WAI fax: +33 4 >92 >> >38 78 22 >> >> Location: 21 Mitchell street FOOTSCRAY Vic 3011, Australia >> >> (or W3C INRIA, Route des Lucioles, BP 93, 06902 Sophia Antipolis Cedex, >> >France) >> >> >> > >> > >> >> -- >> Charles McCathieNevile http://www.w3.org/People/Charles phone: +61 409 >134 136 >> W3C Web Accessibility Initiative http://www.w3.org/WAI fax: +33 4 92 >38 78 22 >> Location: 21 Mitchell street FOOTSCRAY Vic 3011, Australia >> (or W3C INRIA, Route des Lucioles, BP 93, 06902 Sophia Antipolis Cedex, >France) >> > > -- Charles McCathieNevile http://www.w3.org/People/Charles phone: +61 409 134 136 W3C Web Accessibility Initiative http://www.w3.org/WAI fax: +33 4 92 38 78 22 Location: 21 Mitchell street FOOTSCRAY Vic 3011, Australia (or W3C INRIA, Route des Lucioles, BP 93, 06902 Sophia Antipolis Cedex, France)
Received on Tuesday, 30 July 2002 08:57:25 UTC