Re: updated mailto draft

Previous Topic Next Topic
classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view

Re: updated mailto draft

Mike Brown-3

Martin Duerst wrote:
> Dear URI Experts,
> There is now a new mailto draft at

Dates on the draft weren't updated. (published Feb., expires Aug.?)

> This takes into account a lot of comments, but we have unfortunately
> not yet had the time to deal with comments in the thread starting
> at,
> and also some comments from Bruce Lilly. We plan to deal with
> them in the next version (planned for after the upcomming IETF).
> In the meantime, any further comments are highly appreciated.

Section 3 says 'A mailto URI designates an "internet resource", which is the
mailbox specified in the address.'

Since multiple addresses/mailboxes can be put into a single mailto URI, this
statement seems less than ideal. What is the 'resource' represented by a
mailto URI that contains 5 addr-specs?

Since addr-specs are comma-separated, why require the use of "%2C" instead of
a raw "," to separate them?  It seems to go against the conventional wisdom
that percent-encoded reserved characters constitute part of the represented
data while raw reserved characters constitute part of the structure of the
URI.  The comma is being used as a sub-delim rather than data, and it isn't
used anywhere else, so why not say "," has the reserved purpose (in this
scheme) of separating addr-specs? Then, since RFC 2822 makes it pretty clear
that "," would never be found in an addr-spec, you could add that "%2C" is
also valid as a separator, thereby implying that mailto:addr1,addr2,addr3 and
mailto:addr1%2Caddr2%2Caddr3 are equivalent when using scheme-based
normalization (RFC 3986 sec. 6.2.3). So I suggest:
  to = [ addr-spec *(("," / "%2C") addr-spec) ]

Apologies if this has been discussed already; I didn't pay much attention
to the previous posts on this topic.