The 'tag' URI Scheme
RFC 4151

[Ballot comment]
I believe we're following the wrong procedure for this document.  I do
not believe this document meets the criteria for URI registration
using an informational RFC.  In particular, I do not believe the
scheme is in wide use.  However I'm not willing to block on this issue
for this document because I believe this document is well written and
meets the quality we would require of a proposed standard.

[Ballot discuss]
I'm still not sure whether this document is targeted for provisional
registration or for what status it is intended.  I'll also hold a
discuss on the lack of an IANA actions section as there does not seem
to be one on that yet.

If this document is intended as a permanent URI, I would need to do
somewhat more complete review.

[Ballot discuss]
The "Further Information and Discussion of this Document" section
  needs to be deleted prior to publication as an RFC.

  Section 2.4 (Equality of Tags) basically says that tags can
  be compared with strcmp or memcmp.  And then, section 3
  (Internationalisation) says that tags can be IRIs.  The IRI
  document has a section on comparison that is inconsistent with
  the simple comparison described in section 2.4.

[Ballot discuss]
The "Further Information and Discussion of this Document" section
  needs to be deleted prior to publication.

  Section 2.4 (Equality of Tags) basically says that tags can
  be compared with strcmp or memcmp.  And then, section 3
  (Internationalisation) says that tags can be IRIs.  The IRI
  document has a section on comparison that is inconsistent with
  the simple comparison described in section 2.4.

[Ballot discuss]
Word wrapping in the ABNF has introduced syntax errors.  Although this is a grand tradition, starting with RFC 2234 itself, I think it should be discouraged.

[Ballot comment]
Reviewed by Joel Halpern, Gen-ART

His review:

This draft is on the right track for publication as an Informational
RFC, but I think it has an open issue, described in the review.

Specifically, it is using informational publication to create a new
URI scheme. My reading of the scheme definition documents indicates
that informational is for use with existing schemes, and PS is to be
used for new schemes?

Nits: In listing the alternatives, I am sure there is a good reason
why URNs won't work for this problem. (It may be as simple as ~they
look confusing to humans~. Whatever the reason, shouldn't it be
stated?

In the security section, the discussion of malicious parties could
use more work. I can not tell how the second sentence (using
reputable assigning authorities) helps with the threat of someone who
simply lies and makes up tags from a space he does not own. I
suspect that there is an assumption about how tags get associated
with resources that mitigates this, but I can not determine the
assumption from this document.

[Ballot comment]
On page 6:
  Examples of tag URIs are:

      tag:timothy@hpl.hp.com,2001:web/externalHome
      tag:sandro@w3.org,2004-05:Sandro
      tag:my-ids.com,2001-09-15:TimKindberg:presentations:UBath2004-05-19
      tag:blogger.com,1999:blog-555
      tag:yaml.org,2002:int
      tag:herv%C3%A9.example.org,2004:r%C3%A9sum%C3%A9

I kind of understand why these examples are not accoring to our rules
and guidelines for examples. If we want to keep them as is, then maybe
a note as to why we do not follow the rules makes sense.

IANA Comments:  Is this document also expecting the IANA to add this URI scheme to the registry?  There is no IANA Consideration section requesting this registration.

State Changes to New Version Coming                                from AD to write --Don't publish                      by paf

State Changes to New Version Needed (WG/Author) from Reading List                                      by Patrik                    (email update)

State Changes to Reading List                                      from AD to write --Don't publish                      by Steve Coya

State Changes to Reading List                                      from Ready for Telechat                                by Steve Coya

State Changes to Ready for Telechat                                from AD to write --Don't publish                      by Steve Coya