- From: <S.N.Brodie@ecs.soton.ac.uk>
- Date: Fri, 5 Jul 1996 13:59:44 +0100 (BST)
- To: Hermanus@iafrica.com (Melt van Schoor)
- Cc: sja20@hermes.cam.ac.uk, j.wallis@wlv.ac.uk, www-html@w3.org
Melt van Schoor wrote: > > >No, I meant that you'd have an external application work on it to a > >temporary file or area of memory, and then the browser would pick the > >result up and display it in its own window. > > > >James > > Sorry for me-tooing, but this is a great idea. I can imagine that different > people would attempt to create graphics utilities especially for this > purpose, and in the end, it would lead to a vast improvement in speed & quality. > > This approach could also enable the web to contain more than 3 graphics > formats for www pages, if the utility would simply convert it to .GIF (or > .PNG) before passing the results to the browser. Mabey we would even be > able to use vector-based graphics on www-pages, and there are many > advantages to this. This idea reduces the size of the browser by a long way. My web browser has always relied on external tools to convert incoming images into the native sprite format. As a result, support for new graphic formats is trivial, plus I can use virtually any native file format for inlined images. In effect, I suppose this is similar to the effect that plugins give you with Netscape. If a document arrives which isn't text/html or text/plain, then a message is broadcast to all tasks requesting that the object be rendered. I have an option in my web browser to choose which tool is asked to translate GIF images. One of them will do Floyd-Steinberg dithering and is slow because of that, the other does not dithering and is therefore a lot faster although the quality is lower. The dithering and other sophisticated graphical transformations are then under control of a tool other than the browser. The only bad experience I have had with this is that it takes longer to import the images this way, and progressive image display is much harder, and thus animated inlines are very hard. -- Stewart Brodie, Electronics & Computer Science, Southampton University. http://www.ecs.soton.ac.uk/~snb94r/ http://delenn.ecs.soton.ac.uk/
Received on Friday, 5 July 1996 08:59:34 UTC