Re: HTTP: T-T-T-Talking about MIME Generation

On Fri, 16 Dec 1994, John Franks wrote:

> The fastest way to get all the text is to send it first but it can't be
> displaye until layout information like the size and shape of all images is 
> known.  This is the point of the Netscape multiple connections.  They get
> the first few bits of each image which contain the size information.

Yes, but ... the document authors can contribute by having meaningful
alternatives to graphics and folks with slow connections can choose to
ignore the graphics and run in text mode (wibni authors could rank
images importance to content and give users a chance to ignore the
warm feely stuff but get content graphics like weather maps or
economic trends or whatever).

But beyond that,  NETSCAPE has proposed HTML extentions to code image
size in the <img> tag so that would assist the rendering.  And if
HTTP were extended in a minor fashion, the browser could query shape
of objects to speed up the process. An agressive browser like NETSCAPE
could then use two connections, one to pull in the base text and the
other to obtain the shape data.  Dynamic evaluation of the RTT could
lead to more connections if the apparent bandwidth would accept it.

Also, there is really no reason why the browser can't revise the
rendering as more information becomes available.

Received on Saturday, 17 December 1994 01:14:33 UTC