PODC 2000:
Election of Steering Committee Member at-Large

Long-term planning and implementation of PODC is the responsibility of the PODC Steering Committee. The Steering Committee comprises the following individuals: an elected chair; the current Program-Committee chair; the two previous Program-Committee chairs; the current general chair; the current treasurer; and an elected member at-large. Terms expire at the end of each year's PODC (e.g., currently the Program-Committee chairs for 1998, 1999, and 2000 are on the committee; after the end of PODC 2000, the Program-Committee chair for 1998 will leave the committee and be replaced by the Program-Committee chair for 2001).

The position of Steering-Committee member at-large was re-constituted by the PODC Steering Committee in 2000 and thus there will be an election to fill this position this year. The Steering Committee made this decision because it was felt that, at this time, PODC had some unique opportunities that might be realized with the assistance of a member at-large; these opportunities may be detailed in the positions statements of some of the nominees identified below.

The Steering-Committee member at-large has no specific responsibilities. The PODC community may choose, at its annual business meeting, to elect such a member based on what nominees for the position can offer to PODC. Because of this structure, elections for this position will always include a mechanism by which members can vote for "none of the above". If the "none of the above" vote prevails, the position may be filled in a future year. If the position is filled, then the winner of the election will serve for three years (with no more elections until the winner's term expires).

So far, there is one nominee for the position of Steering-Committee member at-large: Keith Marzullo.

The position will be filled by the candidate receiving a majority of votes at the PODC 2000 business meeting (unless a majority votes for "none of the above"). If no candidate receives a majority during the first round of voting, the two leading candidates (one of which may be "none of the above") will enter a second round of voting.

The nominees for Steering Committee member at-large have been invited to submit position statements. These will appear on this web page when they are available.

Keith Marzullo

A group of us are concerned that there is no general high-quality conference for distributed systems. There are specialized conferences
that seem to be doing well, but their focus is narrow.

PODC is a high-quality conference with a narrow focus, but it has for a long time had an interest in broadening to include more implementation papers. There are several reasons for such interest, including financial. One technical reason, though, is to foster the cross-fertilization of ideas between theoreticians and practitioners. Hence, the PODC program committees usually contain some more practical people and the call for papers request more practical papers be submitted. Few are accepted, though, and PODC has been unable to truly broaden its base. Consequently, I've found myself every now and then in a conversation with some colleague who accuses PODC of not really being interested in broadening. Of course, I don't believe that this is true; instead, PODC is not interested in broadening its base by lowering its quality. We just don't get enough practical papers submitted to PODC.

I believe that we should continue to try to broaden PODC, but we need to take a more radical approach. First, we should agree on what we want the final conference to be like. Should it be like Real Time Systems Symposium that has both theoretical and practical sessions? (I like this idea, and have taken to calling such a conference PPODC, or Principles and Practice of Distributed Computing). Or, should we have two parallel conferences, like a mini-federated conference with a leaner version of ICDCS?

In either case, we would need to expand the PODC community. The only way I see to do this is with old fashioned arm twisting. We need to go to the practitioners who are sending good papers to specialized conferences and convince them to send them instead to the new PODC. We need to make it clear that we're looking for excellent papers and plan to reject 80% or more of the submissions. This means that we need to shoot for at least 50 submissions to have half practical papers. And, we need to back it up with a rock-solid program committee.

We took a first step this year towards this: Aleta Ricciardi created an associated middleware workshop and chose a strong program committee of distributed computing experimentalists. They used a carrot of offering to recommend exceptional papers for consideration to Distributed Computing. They accepted 6 papers (25% accept rate) and did not recommend any papers for DC. The workshop will be integrated with PODC; we hope that there will be some cross-discussions.

The PODC Steering Committee should address the question of what a broadened PODC should look like, and should work towards that implementation. Doing so will involve leading discussion among the PODC community. Then, the Steering Committee should work towards an implementation.

I nominated myself for the Member at Large position because I feel strongly that PODC should change. Although I have served many times on PODC's program committee and consider the conference one of my favorites, my research is not in the core of its current agenda. I think that what is required is someone who will expend some energy talking with people: soliciting input from current PODC community members and twisting the arms of future members. This person should be affiliated with PODC but also with other (more practical) conferences. I am eager to try and also willing to help anyone else who wishes to take on this job.

This page maintained by Gil Neiger.

Last modified: June 27, 2000