- From: Dan Connolly <connolly@w3.org>
- Date: Tue, 16 May 2006 20:01:28 -0500
- To: noah_mendelsohn@us.ibm.com
- Cc: Dan Brickley <danbri@danbri.org>, Chris Lilley <chris@w3.org>, Daniel Dardailler <danield@w3.org>, Thomas Roessler <tlr@w3.org>, www-tag@w3.org
On Tue, 2006-05-16 at 20:46 -0400, noah_mendelsohn@us.ibm.com wrote: > I wrote: > > > Indeed, or why not: application/telephoneNumber+xml > > Sometimes I fall prey to thinking about the Web as I intuit it should > work, rather than as it is. On reflection, assigning media types for > little things like phone numbers probably doesn't scale. Well, current practice works both ways. In some cases, you follow a Try <a href="http://chat.service.example/text-voice-chat" >real-time customer support</a> and back comes a file with a mime type that causes your browser to launch a voip doodad, and sometimes the link looks like: Try <a href="sip://sip.service.example/or-something-like-this" >real-time customer support</a> and the dispatch goes via the URI scheme. It's also possible that /text-voice-chat will redirect to a sip: URI, or all sorts of other possibilities. But all that is issue schemeProtocols-NN-i-forget, which is pretty much orthogonal to the .tel issue; all of those mechanisms work without a new TLD. -- Dan Connolly, W3C http://www.w3.org/People/Connolly/ D3C2 887B 0F92 6005 C541 0875 0F91 96DE 6E52 C29E
Received on Wednesday, 17 May 2006 01:01:33 UTC