Re: [apps-discuss] WEIRDS WG report

On 2014/03/05 06:41, Murray S. Kucherawy wrote:
> WEIRDS met today and discussed what we believe are its remaining issues.
>
> The group discussed the outcome of a design team meeting that tried to
> address two issues at once: the issue an imposed URI structure
> (draft-nottingham-uri-get-off-my-lawn) and the methodology by which to find
> an authoritative registry that can further delegate.  Both issues are being
> addressed by allowing a key-value table that supports longest-match lookups
> resulting in  URI templates that are then used to construct a query that
> will either return an answer or a redirection.  The WG will continue this
> discussion on its mailing list and select a path forward.  Participants
> were asked to open tracker items for any remaining issues on the open
> documents (query, redirection, bootstrapping) so they can be driven to
> resolution.

Regarding this point, Mark sent a mail to the W3C TAG (Technical 
Architecture Group) a while ago, at 
http://lists.w3.org/Archives/Public/www-tag/2014Feb/0068.html, and I 
replied at
http://lists.w3.org/Archives/Public/www-tag/2014Feb/0079.html, switching 
to defend the (former?) WEIRDS position. There was only one reply, 
agreeing with me
(http://lists.w3.org/Archives/Public/www-tag/2014Feb/0080.html) but 
probably better worded.

Looking at the above "key-value table that supports longest-match 
lookups resulting in URI templates that are then used to construct a 
query that will either return an answer or a redirection", this to me 
smells like over-engineering even just when reading it (not to speak of 
having to implement it).

Of course, if the WEIRDS WG is happy with that, I won't complain any 
further, but I just wanted to mention this because I earlier supported 
Mark and Tim on this list (and would still continue to do that if there 
were no indirection at all in WEIRDS, i.e. if e.g. WEIRDS was requiring 
top-level directories with fixed names).

Regards,   Martin.

> The discussion of the publication of the table (IANA being the logical
> point) ventured into the complications that come with the IETF creating
> domain registries and imposing policies potentially outside of its
> purview.  This issue has not been completely resolved but we believe there
> are ways forward through careful crafting of the language and keeping the
> IANA section to an absolute minimum.
>
> Finally, a report was given of interoperability tests performed prior to
> the IETF meeting.  No specification issues were discovered and a fair
> number of server as well as client implementations interoperate.  A couple
> of issues were identified and we are at or close to consensus on resolution.
>
> The chairs believe that the work of the group may conclude between now and
> Toronto, so this may have been our final meeting.
>
> -MSK and Olaf, co-chairs, WEIRDS
>
>
>
> _______________________________________________
> apps-discuss mailing list
> apps-discuss@ietf.org
> https://www.ietf.org/mailman/listinfo/apps-discuss
>