Re: [xri] Property, rel comparison

[Breno de Medeiros <breno@google.com>]

We can define that URIs are not supposed to be compared using
schema-specific mechanisms but then things will break because people
will use equivalent URI/URLs to refer to the same resource and
interoperability issues will appear.
This is not the case where making implementation easy is the most
important point. The spec should define URI comparison to be based on
the scheme-specific notion of equivalence.

Also, the recommendation to treat URIs as case-insensitive is just crazy.


On Mon, Jan 25, 2010 at 09:20, Eran Hammer-Lahav <eran@hueniverse.com> wrote:
> I am not sure the URI normalization is intended in Web Linking. I don't recall it being discussed, other than the issue about being case-insensitive (which is not how URI comparison works for certain parts like the path).
>
> If you feel strongly about it, please chime in on the IETF Apps-Discuss list in reply to my post.
>
> EHL
>
>> -----Original Message-----
>> From: Scott Cantor [mailto:cantor.2@osu.edu]
>> Sent: Monday, January 25, 2010 9:17 AM
>> To: xri@lists.oasis-open.org
>> Subject: RE: [xri] Property, rel comparison
>>
>> Eran Hammer-Lahav wrote on 2010-01-25:
>> > I think we should incorporate this language because it might not be
>> obvious
>> > to developers who are not likely to study the Web Linking spec. Since
>> > we have a dependency on Web Linking, we will have to wait for it to
>> > get published as an RFC (at least approved by the IESG), and if this
>> > changes,
>> we
>> > will follow.
>> >
>> > BTW, I have an open issue with the above text because I think
>> > "compared as URIs" is poorly defined (is http://example.com/rel the
>> > same as http://example.com:80/rel?).
>>
>> That's why I don't agree with using URL normalization rules to compare URIs
>> that are intended as identifiers. I think binary comparison is much easier and
>> much more likely to produce interop. It has certainly worked that way in
>> SAML.
>>
>> I didn't push on it because I had the impression that everybody wanted URL
>> normalization, but I personally would argue for not doing that where
>> possible.
>>
>> -- Scott
>>
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe from this mail list, you must leave the OASIS TC that
>> generates this mail.  Follow this link to all your TCs in OASIS at:
>> https://www.oasis-
>> open.org/apps/org/workgroup/portal/my_workgroups.php
>
>
> ---------------------------------------------------------------------
> To unsubscribe from this mail list, you must leave the OASIS TC that
> generates this mail.  Follow this link to all your TCs in OASIS at:
> https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php
>
>



-- 
--Breno

+1 (650) 214-1007 desk
+1 (408) 212-0135 (Grand Central)
MTV-41-3 : 383-A
PST (GMT-8) / PDT(GMT-7)