Firstly, I'll observe that this is outside the strict scope
of the Secretariat SOW, since it covers the process cradle-to-grave,
including WG, IESG, IANA and RFC Editor actions.
Secondly, yes, "dashboard" metrics are a good idea, and are on the
Tools team agenda, but often the devil is in the details and it's
only by looking at specific cases of apparently stuck drafts
that we can understand why things are moving slowly.
Brian
Spencer Dawkins wrote:
Bernard/All,
Ack on Bernard's note.
I know that speed isn't the only thing that matters, but if we move
slowly enough, the other stuff that matters won't matter.
I'm remembering from previous discussions (sometime around the time of
http://www3.ietf.org/proceedings/03mar/134.htm? or
http://www3.ietf.org/proceedings/04mar/981.htm?) that the states we
track in the ID tracker are sometimes overloaded, so it's hard to tell
who has the token and exactly what is happening with the draft, and
there are limits on what we've been able to do with metrics in the past.
It's definitely worth thinking about this from a metrics perspective.
Spencer
In thinking through the Statement of Work (SoW), I think that an
important
component is to provide the IETF with sufficient information on how well
the organization is performing.
There are many metrics for that, but an important one is the time
taken in
various stages of the IETF process.
Unfortunately, it is not clear to me that we are currently collecting
this
information in a form that makes it easy to analyze. We are also not
analyzing the data on a regular basis, using it in a systematic effort to
improve IETF performance (or at least to prevent it from deteriorating
further).
Researchers such as Tim Simcoe of the University of Toronto have studied
metrics of IETF performance and have come to some interesting
conclusions.
For example, it appears that time from an initial -00 to RFC publication
varies considerably by area, as well as by designation (Information,
Experiemntal, Proposed). In the process of developing this research, Tim
has also had to do significant work to adjust the data to make it
suitable
for analysis.
My suggestion is that the IAOC needs to start thinking about what data
and reports are needed to enable the IETF to measure and improve its
performance.
References
----------
Simcoe, T., "Delays and de Jure Standards: What Caused the Slowdown in
Internet Standards Development?", UC Berkeley Haas School of Business,
April 2004.
Simcoe, T., "Standards Setting Committees", J.L Rotman School of
Management, University of Toronto, Decmeber 2005.
Available at: http://www.drizzle.com/~aboba/IAB/simcoe/
_______________________________________________
Ietf@xxxxxxxx
https://www1.ietf.org/mailman/listinfo/ietf
_______________________________________________
Ietf@xxxxxxxx
https://www1.ietf.org/mailman/listinfo/ietf
_______________________________________________
Ietf@xxxxxxxx
https://www1.ietf.org/mailman/listinfo/ietf