Hash: SHA1
Craig Partridge wrote:
> Hi Joe:
>
> I cannot tell from your note if we are roughly agreeing or disagreeing.
>
> So let me try upleveling and see if we agree or disagree.
>
> * I think we agree that COMSOC and the TC needs to sunset conferences and that
> long-established meetings sometimes don't get enough scrutiny.
I don't think the term "sunset" is accurate - that implies that existing
meetings are approved until explicitly stopped. The Comsoc and TCCC both
evaluate every meeting every time.
> * I think we disagree about what kind of scrutiny is appropriate.
> I, based on my IEEE and ACM experience, find that the simply tracking
> if the conference sticks to its usual review process and character
> and assuming that if interest is lost then finances fall short is
> a slow and not always effective way to sunset conferences.
The key question is "under what conditions does the Comsoc or TCCC
endorse/sponsor meetings?" The current metrics for endorsement focus on
meeting quality (number of independent peer reviews, diversity of the
TPC, etc.). The current metrics for sponsorship (involving financial
support) focus on viability (surplus/loss).
It's hard to understand why those metrics would apply differently to
ongoing meetings than to new meetings - in both cases, we sometimes take
a chance (allowing new meetings, or "off" years of existing meetings).
> (I can name a number of IEEE conferences and some ACM conferences that
> everyone I knew felt lingered too long).
>
> * I think we agree that whatever metrics we use for scrutiny should
> be rigorous and measure a relevant feature. We have some disagreement
> about whether citations or other impact ratings should be part of
> the process. Note that I'm NOT arguing this should be the sole
> measure -- there are reasons for meetings that don't get reflected in
> citations/impact numbers. But I'd like us to start requesting some
> consistent across conferences and independent (and/or hard to fudge)
> metrics beyond simply finances and attendence and, if we are requiring
> them in IEEE (we're not in ACM), I'd like to encourage your panel to
> examine if we're using the right metrics.
The panel's purpose is to examine the costs of current conferences, and
to explore future ways to contain costs. Metrics for sunsetting are out
of scope of that group.
If we want to discuss such metrics, we can start on the mailing list. I
don't know whether a panel would be useful in that process - we didn't
need one to converge on the metrics for endorsement that were recently
approved, e.g.
That said, as an individual contributor to this discussion (i.e., chair
hat "off"), I doubt there are appropriate objective conference metrics
other than those that currently exist for quality and viability. In
specific, using citation indices presumes that a key purpose of a
conference is to publish papers to be cited - that, as I noted before,
is a key purpose of a journal, but conferences have other very important
metrics. E.g., even if no papers are ever cited from a given meeting,
the feedback researchers receive that makes their work better for future
publication (e.g., in a journal) would be more than sufficient to
warrant a conference's continuation.
Joe
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (MingW32)
iEYEARECAAYFAktBFwMACgkQE5f5cImnZrvL8QCeJdw4UA1epjuUYNFc2EHb0XRs
Y+cAmwdQCBrYFZuQ6avFUj9+M6bscpnZ
=QA3D
-----END PGP SIGNATURE-----
_______________________________________________
Tccc mailing list
Tccc@lists.cs.columbia.edu
https://lists.cs.columbia.edu/cucslists/listinfo/tccc
Posts
No comments:
Post a Comment