Re: [apps-discuss] Review of draft-snell-atompub-tombstones-14

[James M Snell <[e-mail]>]

Yves Lafon <[e-mail]> wrote on 02/07/2012 12:46:46 AM:

[snip]
>
> Review of draft-snell-atompub-tombstones-14
>
> On Mon, 6 Feb 2012, Yves Lafon wrote:
>
> It's always better with a proper subject line :)

Yes, quite... my spam filter flagged the first one because of the lack of a
subject line ;-)

>
> > Dear all,
> >
> > I have been selected as the Applications Area Directorate reviewerfor
this
> > draft (for background on appsdir, please see
> >
http://trac.tools.ietf.org/area/app/trac/wiki/ApplicationsAreaDirectorate)
> >
> > Document: draft-snell-atompub-tombstones-14
> > Title: The Atom "deleted-entry" Element
> > Reviewer: Yves Lafon
> > Review Date: 6 Feb 2012
> >
> > Summary: This draft is almost ready for publication as Informational
RFC,
> > there is a minor issue with the ref. section, and a clarification
> request on
> > a MUST-level requirement that is non-blocking.
> >
> > First, the clarification request: I noted that in 3/ "The
at:deleted-entry
> > element"
> >
> > It says:
> > <<
> >   An Atom feed MAY contain any number of at:deleted-entry elements, but
> >   MUST NOT contain more than one with the same combination of ref and
> >   when attribute values.
> >>>
> > then later
> > <<
> >   Implementors should note that the at:deleted-entry element is
> >   informative in nature only and may be ignored by Atom processors.
> >   The presence of an at:deleted-entry element does not guarantee that
> >   the atom:entry to which it is referring will no longer be available.
> >>>
> >
> > If it is informative only, why in the first paragraph there is a
> MUST NOT on
> > duplicate at:deleted-entry with same 'ref' and 'when', especially as
two
> > delete may happen in the same second while being different.
> > What is the rationale for having a MUST NOT instead of a SHOULD NOT?
> >
> >
>

While SHOULD NOT would likely work here, the motivation for the MUST NOT is
that I was unable to find any valid cases where multiple deleted-entry
elements for the same entry at the same time would appear in a single feed.
If such duplication did occur, it is either certainly an error or a
possible attack vector.

The Informative nature is purely a recognition that, as an extension to
Atom, there mere presence of the deleted-entry element in a field does not
guarantee that Atom implementations will be able to actually do anything
with it.

> --
> Baroula que barouleras, au tiéu toujou t'entourneras.
>
>          ~~Yves
>