On Wed, Jan 23, 2013 at 04:52:48PM +0100, Stephane Bortzmeyer wrote:
> On Wed, Jan 23, 2013 at 04:47:17PM +0100,
>  Sebastian Kiesel <[email protected]> wrote 
>  a message of 50 lines which said:
> 
> > > Earlier versions explicitly had a reference to RFC 3958, explaining
> > > that ALTO discovery requires an URI as output.
> > 
> > this is still in section 1, third paragraph of our draft.
> 
> No, it is not, the reference to S-NAPTR (RFC 3958) is not
> present. (There is a reference to RFC 3958 later but for a differant
> reason.)

our draft states that we have to yield and deliver an URI.

The U-NAPTR spec, RFC 4848 which is cited, has an introduction section
with rationale why S-NAPTR is not sufficient for all applications.

let me cite from that:

   This U-NAPTR is effectively a modest extension to S-NAPTR, to
   accommodate the use of URIs as targets, without allowing the full
   range of possible regular expressions in NAPTR records.

> BTW, to return a URI, there is also draft-faltstrom-uri

we have all we need in RFCs, so why build on drafts that might still
be subject to change?

Sebastian
_______________________________________________
alto mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/alto

Reply via email to