Re: Progressing I-Ds Immediately Before Meetings

[John C Klensin <john-ietf at jck.com>]

--On Sunday, July 20, 2008 2:07 PM -0400 Russ Housley 
<housley at vigilsec.com> wrote:

> People seem to be forgetting that all I-D submissions used to
> be processed an a person.  The automated tool is very new.
> When the I-D were processed by hand, the cut-off was necessary
> for the Secretariat to handle the spike of submissions just
> prior to the meeting.  Look at the statistics report by IETF
> chair in their plenary presentations for the last few years --
> the meetings cause a huge spike in I-D submissions.  Remember
> back when the I-D submissions were handled by hand, it could
> be days after the submission that the document actually
> appeared in the repository.
>
> Now that we have the tool, it is a reasonable time to see if
> we still need this cut-off rule.  I have put the topic on the
> agenda for the IESG discussions in Dublin.

Russ,

Thanks.

FWIW, it appears from various notes on this thread that we 
actually have two separate cutoff rules, with one having 
effectively hidden the other.  One of those rules is the two and 
three week hard cutoff originally imposed to protect the 
Secretariat, as you have described above.   That rule came with 
a "AD exception" procedure, which seems to have gotten lost over 
the years ("lost" == "current ADs didn't know about it"). 
Whether that exception model was a good idea or not relative to 
other things ADs could do with their time is a separate question 
-- it did exist and, for better or worse, it was _very_ rarely 
used.

The other rule is an RFC 2418 rule that says "should ... two 
weeks..." (note lower-case "should").  It carries with it 
provision for exceptions by WG chairs and some guidance as to 
whether those exceptions should be granted.

Many of the recent comments in this thread (including, I fear, 
mine) have confused the two.   It appears to me that most of the 
suggestions could be accommodated by looking at the first set of 
rules (the posting cutoffs) only, either eliminating them 
entirely  or cutting them to the point needed to assure a level 
playing field for those who are forced into manual posting by 
deficiencies in the automatic posting tool.

> I see this an opportunity for evolution and incremental
> improvement.

Indeed.  And a very welcome and much appreciated one.

  john



_______________________________________________
Ietf mailing list
Ietf at ietf.org
https://www.ietf.org/mailman/listinfo/ietf


rg
Errors-To: ietf-bounces at ietf.org



--On Sunday, July 20, 2008 2:07 PM -0400 Russ Housley 
<housley at vigilsec.com> wrote:

> People seem to be forgetting that all I-D submissions used to
> be processed an a person.  The automated tool is very new.
> When the I-D were processed by hand, the cut-off was necessary
> for the Secretariat to handle the spike of submissions just
> prior to the meeting.  Look at the statistics report by IETF
> chair in their plenary presentations for the last few years --
> the meetings cause a huge spike in I-D submissions.  Remember
> back when the I-D submissions were handled by hand, it could
> be days after the submission that the document actually
> appeared in the repository.
>
> Now that we have the tool, it is a reasonable time to see if
> we still need this cut-off rule.  I have put the topic on the
> agenda for the IESG discussions in Dublin.

Russ,

Thanks.

FWIW, it appears from various notes on this thread that we 
actually have two separate cutoff rules, with one having 
effectively hidden the other.  One of those rules is the two and 
three week hard cutoff originally imposed to protect the 
Secretariat, as you have described above.   That rule came with 
a "AD exception" procedure, which seems to have gotten lost over 
the years ("lost" == "current ADs didn't know about it"). 
Whether that exception model was a good idea or not relative to 
other things ADs could do with their time is a separate question 
-- it did exist and, for better or worse, it was _very_ rarely 
used.

The other rule is an RFC 2418 rule that says "should ... two 
weeks..." (note lower-case "should").  It carries with it 
provision for exceptions by WG chairs and some guidance as to 
whether those exceptions should be granted.

Many of the recent comments in this thread (including, I fear, 
mine) have confused the two.   It appears to me that most of the 
suggestions could be accommodated by looking at the first set of 
rules (the posting cutoffs) only, either eliminating them 
entirely  or cutting them to the point needed to assure a level 
playing field for those who are forced into manual posting by 
deficiencies in the automatic posting tool.

> I see this an opportunity for evolution and incremental
> improvement.

Indeed.  And a very welcome and much appreciated one.

  john



_______________________________________________
Ietf mailing list
Ietf at ietf.org
https://www.ietf.org/mailman/listinfo/ietf