- From: Mark Baker <distobj@acm.org>
- Date: Fri, 4 Jul 2008 13:05:02 -0400
- To: "Erik Wilde" <dret@berkeley.edu>
- Cc: public-geolocation@w3.org
On Sun, Jun 29, 2008 at 2:29 PM, Erik Wilde <dret@berkeley.edu> wrote: > hello. > > i guess i am missing some context here. where did that discussion start? i > can't remember seeing it on the public-geolocation list, but i probably just > missed it. The thread started here; http://lists.w3.org/Archives/Public/public-geolocation/2008Jun/0001.html If you read further on from there you'll learn that different people had different ideas about what "DOM API" meant. I used it to mean an interface based on the DOM, and others thought it meant - AFAICT - any old Javascript API. > there also is/was > https://datatracker.ietf.org/drafts/draft-daviel-html-geo-tag/ which is dead > now, but not for a long time. it has been around for a while, but never made > to a standardized format. the last version was -09, i think. Right, I forgot about that. It too could be used, though I expect > repeating my appeal for a well-defined position model > (http://lists.w3.org/Archives/Public/public-geolocation/2008Jun/0093.html), > i think it would be a really good idea to have a position model that is > consistent across embodiments in various technologies (such as a DOM API or > an HTML version or a URI or HTTP headers). > > i know that this somehow contradicts the major goal of the geolocation WG to > get something out of the door quickly. but i also think for that "something" > (the DOM API) to be well-designed, it needs to have a well-designed position > model anyway, so why not just do it once and in a way that is reusable > across specifications? The position model does indeed seem an orthogonal concern. A true DOM based API could support any model I expect. Mark. -- Mark Baker. Ottawa, Ontario, CANADA. http://www.markbaker.ca Coactus; Web-inspired integration strategies http://www.coactus.com
Received on Friday, 4 July 2008 17:05:45 UTC