Re: revised mailto spec

Tim Berners-Lee (timbl@w3.org)
Mon, 09 Dec 1996 15:47:44 -0500


Message-Id: <32AC7AF0.5EC5@w3.org>
Date: Mon, 09 Dec 1996 15:47:44 -0500
From: Tim Berners-Lee <timbl@w3.org>
To: Larry Masinter <masinter@parc.xerox.com>
Cc: uri@bunyip.com, jwz@netscape.com
Subject: Re: revised mailto spec

If this is the same proposal as has been around for a while
to make a mailto: URL a way of extracting data by email
reply, then I would oppose that.

Cocneptually, a mailto: URL is a point in th eemail address
space.  It is a well defined and importamt space.  Mail addresses
behave in certain ways.   There are things I can do with them.
(For example, I can display a list of all messages I know of recently
to and from that address).

If you try to hook on a space of information objects retrievable
by email application, you get  quite different space and a quite
different behaviour. (You will want to cache results, and track
requests).  Muddling the two would be a mistake -- it would
confuse users, and mean that a whole extra set of behaviour would have
to be taken into account by the software.  It would also be
a very messy architectural decision.

Tim Berners-Lee


Larry Masinter wrote:
> 
> I want to propose recycling 'mailto' to "Proposed Standard" and
> suggest extending it to include mail headers. Some browsers do (some
> of) this anyway, and it's a generally needed functionality. We were
> going to make it 'mailserver' for a while, which flopped because
> there's really no need for another URL scheme.
> 
>  ftp://ftp.parc.xerox.com/pub/masinter/draft-hoffman-url-mailto.txt
> 
> It's looking more and more like there SHOULD be a URL working group
> (re)constituted, but we'll need a chair, a charter, milestones, etc.
> 
> Also, volunteer to take minutes at our (very short, 30 minute) BOF.
> 
> Larry