Re: Browser implementation of active images
Wed, 13 Jul 1994 17:19:17 +0200

Murray Bent writes:
> According to the HTML+ draft of Nov 1993,
>>Images can be made active in one of three ways
>> 1. The whole image can be made into a hypertext link
>> 2. Mouse/Pen clicks on the image can be passed to a WWW server
>> 3. Shaped hypertext buttons can be overlayed on the image
>For 2,
>How will browsers detect that an image can have multiple mouse clicks
>and mouse dragging resulting in a sequence of coordinates sent to server?

Good question. I just detect how long the mouse has been down, before
turning it from a click into a mouse-drag sort of submit. Semi-easy.

>For 2 and 3,
>Have any browsers already implemented this?

The Emacs-w3 browser has support for this, but it is broken right now
because of a limitation of lucid emacs. It sends the results (x,y) in
relation to the entire emacs frame, not the actual image. Next version of
Xemacs will hopefully have this feature built in. But the groundwork is
there, so it will take about 3 minutes to add 'real' support once emacs
supports the necessary primitives.

I imagine that Pei Wei could add this to Viola-WWW pretty quickly if
he hasn't already. Working in an interpreted language is so much
nicer/faster for prototyping things like this.

>Sorry if this has been asked before.

Don't think it has - good point though. Perhaps we need another
tag? ISDRAG to complement ISMAP?