Sigcomm 2005 European Shadow PC Information
The Shadow PC is an experiment with many different aspects. Its main goal is to
educate all participants about the review process of a competitive conference
and the criteria that makes a paper a good one. Participation in the Shadow PC
process has no impact at all on the normal review process!
Motivation for a Shadow PC
What is a Shadow PC
Benefits of a Shadow PC to the authors
Benefits of a Shadow PC to the Shadow PC members
Benefits of a Shadow PC to Sigcomm
Criteria for Shadow PC members
Members of the Shadow PC
Motivation for a Shadow PC
The hope is that Shadow PC experience can address some of the following
concerns:
- Most Sigcomm authors and PC members are either at institutions in the US or
have strong ties to US research labs such as Sprint or AT&T.
- Some folks consider Sigcomm as cliquish. The chance of getting a paper
accepted is proportianal to one's distance to the in-crowd.
The impression is that even the double blind submission and review process
does not prevent this. The Shadow PC experience allows people outside the "inner"
core to experience a Sigcomm review process (the shadow PC) first hand.
- Expand the set of potential reviewers and PC members from Europe.
- ...
What is a Shadow PC
A shadow PC is one that will not have **any impact** on
which papers are going to be presented at the conference. Yet the shadow PC
will be run just as the real PC in the sense that PC members will:
- paper authors decide if they want to participate in the experiment or not.
The decision to participate in the Shadow PC process has **no** impact on
the real PC.
- be asked which of the 240 papers participating in the experiment they would
like to review based on the abstracts of all submitted papers
- be assigned roughly 5 papers to check if they should be subject to
rejection in the first stage of the review process.
(this means that the shadow PC should not consider this paper
further, because it is out of scope or because it is obvious that the
paper has no chance of being accepted). Not all papers suggested for
rejection at this stage are automatically excluded!
- be assigned roughly 15-20 papers (including the above 5) for
rigorous review
- attend a PC meeting in Munich probably in the second half of April
(just after the actual PC meeting). Everyone participating should attend
this meeting.
- the reviews will be sent back to the authors in a separate email clearly
marked as reviews by the shadow PC
In contrast to a "normal PC" meeting we are planing to extend the PC meeting
to two days. The first day will be the "regular PC meeting" in which all
highly rated papers are discussed and a "final program" is selected. The
second day is focused on reviewing and better understanding the process. This
includes:
- that every participant of the shadow PC will be able to see the anonymized
reviews of some of the real PC members for the papers that they reviewed
(Since reviews can contain sensitive information real PC members can opt to
not share their reviews with the shadow PC.)
- that the shadow PC has a chance to compare the two final programs (of the
real and the shadow PC)
followed by:
- a discussion of the differences in the reviews.
- a discussion of the differences in the final program envisioned by the
shadow PC vs. the accepted papers by the real PC.
Benefits of a Shadow PC to the authors
- Additional reviews for their papers.
- The shadow PC will be helpful in characterizing the differences of opinions
on papers for people from varied backgrounds (actual PC vs. Shadow PC).
- A more transparent review process. (The Shadow PC includes people that
are not at the inner core of Sigcomm. These might have a different view of the
material and a different perspective.)
- Differences between the Shadow PC and the real PC opinions can be
investigated to see if some material of quality was systematically being
missed by the actual PC. This might help with future Sigcomms.
- ...
Benefits of a Shadow PC to Shadow PC members
Given that the shadow PC does not have any impact one might wonder why someone
should volunteer to participate. Here are some of the reasons. The main one is
that it is a learning experience!
- participants get to know how a PC is run and how it operates.
- participants get to know the other PC members.
- participants will get to experience the process SIGCOMM uses to select papers.
- participants have a chance to read top notch papers in their area of expertise.
- participants will see good quality papers as well as bad quality papers
(helpful for calibrating expectations) and see how they are viewed by
others. The latter is probably the most important aspect.
This should motivate everyone to think about what aspects make a paper a
good one and what aspects have to be addressed in order to succeed in
getting a paper accepted at such a selective conference as SIGCOMM.
- participants that submit high quality reviews are possible candidates for future
PCs.
The Shadow PC experience should help members in the following ways:
- Members will get to see good quality papers as well as bad quality
papers. This helps with calibrating the overall expectations.
- Members will have a chance to review a bunch of papers and therefore get
motivated to think about what aspect of the paper make it a good paper and which
parts make it a bad paper. Therefore they might learn something about what are
the parts are needed in order to succeed in such a selective conference.
- Through the review process members will see that the double blind process
usually works.
- Based on the quality of the reviews of the shadow PC members it is possible
to identify possible candidates for future PCs.
- ...
Benefits of a Shadow PC to Sigcomm
- The shadow PC will benchmark the work of the actual PC.
- Identify qualified European PC member candidates.
- Make the SIGCOMM review process more transparent.
- Make SIGCOMM more accessible to non-US based people.
Criteria for Shadow PC members
In order to ensure good interactions at the PC meeting as well as high quality
reviews the following criteria apply to shadow PC members:
- # of PC members:
around 35-45 (about 2 times the number of PC members) to reduce the workload
for each member but to ensure that each member has read a significant
number of papers and can therefore participate in the discussion at the PC
meeting
- level of PC members:
I have contacted folks at different European institutions and I asked
them to nominate possible members. Each shadow PC member should already
have some experiences in terms of reading papers and have some foundation
in his area. Ideally the shadow PC member is a senior graduate student
(1-2 years before graduation) or a post doc or a recently appointed
faculty member that has not yet participated in SIGCOMM, SIGMETRICS, IMC,
or INFOCOM PCs..
Shadow PC members
| Name |
Institution |
Country |
| Vinay Aggarwal | TU Muenchen | Germany |
| Marcelo Amorim | Universite Pierre et Marie Curie (Lip6) | France |
| Dominique Barthel | France Telecom | France |
| Marc Bechler | TU-Braunschweig | Germany |
| Augustin Chaintreau | Ecole Normale Superieure | France |
| Gyoergy Dan | KTH | Sweden |
| Viktoria Fodor | KTH | Sweden |
| Holger Fuessler | Universitaet Mannheim | Germany |
| Thomas Fuhrmann | Universitaet Karlsruhe | Germany |
| James Gross | TU Berlin | Germany |
| Andrei Gurtov | University of Helsinki | Finland |
| Horst Hellbrueck | Universitaet Luebeck | Germany |
| Gisli Hjalmtysson | Reykjavik University | Iceland |
| Rebecca Isaacs | Microsoft Research | UK |
| Thomas Karagiannis | Intel Research | UK |
| Roger Karrer | ETH Zuerich / Rice | Switzerland / USA |
| Christian Kreibich | University of Cambridge | UK |
| Lambros Lambrinos | University of Cyprus | Cyprus |
| Lars-Ake Larzon | Uppsala University | Sweden |
| Arnaud Legout | Inria Sophia Antipolis | France |
| Vincent Lenders | ETH Zuerich | Switzerland |
| Olaf Maennel | TU Muenchen | Germany |
| Laurent Mathy | Lancester University | UK |
| Martin May | ETH Zuerich | Switzerland |
| Marco Mellia | Politecnico di Torino | Italy |
| Pietro Michiardi | Eurocom | France |
| Richard Mortier | Microsoft Research | UK |
| Colin S. Perkins | University of Glasgow | UK |
| Kave Salamatian | Universite Pierre et Marie Curie (Lip6) | France |
| James Scott | Intel | |
| Gwendal Simon | France Telecom | France |
| Giovanni Stea | Universita di Pisa | Italy |
| Renata Teixeira | UC San Diego | USA |
| Steve Uhlig | Universite catholique de Louvain | Begium |
| Klaus Wehrle | Universitaet Tuebingen | Germany |
| Michael Welzl | Leopold-Franzens-University of Insbruck | Austria |
| Joerg Widmer | EPFL | Switzerland |
| Eng Keong Lua | University of Cambridge | UK |
| Tatsuya Mori | NTT Labs | Japan |
| Jeyashankher SR | Lucent | India |
| Huzur Saran | IIT Delhi | India |
| Yongqiang Xiong | Microsoft | China |
Last updated February 2005.