« September 2008 | Main | February 2009 »

November 19, 2008

Conferences as organizations - advising, steering and establishing expectations

While most of us are involved in organizing conferences in some way, we probably do not pay too much attention to the organizational model of these events. This is somewhat surprising, given that conferences are probably the most visible activity of most professional societies, and also entail significant expenditures of money and volunteer labor. While the local pigeon racing club with a $500 annual budget probably has bylaws and statutes, most conferences with hundred thousand dollar budgets operate more by oral tradition than by formal descriptions of responsibilities. In almost all cases, this works just fine, but this informality can lead to misunderstandings or problems when expectations differ among the volunteers or when there is a crisis. Thus, I believe that it is helpful to have clearer models, so that conferences and volunteers can reach a common understanding of what is expected of everybody that contributes their time to the conference, and also who is responsible when things go wrong.

For long-running conferences, the typical conference organization involves four major actors: the sponsoring professional organization, a steering committee, the general chairs and the technical program chairs. However, the roles and reporting relationships seem to differ rather dramatically between different conferences.

I believe that the differences are partially due to the fact that conferences differ in their relative independence in relationship to the related professional society, such as ACM, IEEE or Usenix. In some cases, the conference is just a part of the organization, and has really no independent identity. As an example, the IEEE Communications Society "owns" IEEE Infocom, Globecom and ICC, and it seems highly unlikely that, say, the IEEE Infocom steering committee could just decide to move their conference to full ACM sponsorship. Similarly, the notion that ACM SIGGRAPH would suddenly decide to run its own show seems implausible. As a crude analogy, consider Walmart stores or the company-owned Starbucks outlets, while in the education realm, state universities are probably somewhat similar. In the end, the state government calls the shots and sets the tuition, even if the university president has a fair amount of latitude in running the daily campus affairs.

On the other hand, there are conferences that are much more independent, even if they have been affiliated with the same professional society for years. Such conferences could, and sometimes do, decide to change their affiliation or they might even become their own full-fledged organization. As examples, ACM NOSSDAV and ICNP probably could decide to switch sponsoring society, or at least switch the SIG they ask for sponsorship. These conferences seem much closer to organizations such as private universities or religious congregations, where each such organization is run by self-perpuating group of trustees. These trustees do not own the organization and could not sell it for their own benefit, but they could, for example, decide to disband the organization and sell its assets. Unlike in a business, such trustees are not financially responsible for the organization, but may well face liability for negligence or malfeasance.

Regardless of the structure, I would strongly suggest that long-running conferences that set up a steering committee create at least an informal set of bylaws that addresses a few core issues, such as:


  • Is the conference permanently affiliated with a particularly society or can the steering committee change such affiliation?

  • Is the steering committee the same as the SIG executive committee, overlap with that executive committee or does it report to the executive committee?

  • Does the conference maintain its own budgets and accounts, e.g., to carry forward surpluses from one year to the next? (This only applies to conferences that are independent organizations, such as VLDB.)

  • Does the steering committee just pick the general chairs, or also the technical program chairs?

  • Who gets the final word on adding or deleting major conference elements, such as tutorials or social events?

  • Does the technical program chair report to the general chair or only to the steering committee? (In the analogy used earlier, is the general chair the CEO and the technical program chair the CTO, or are they more like chief operations officer and CTO?)

  • Does the steering committee provide take-it-or-leave-it advice on the technical program committee, leave this completely to the technical program chairs or do they have to get formal approval?

  • Who decides the location of the conference, both the rough location ("New York") and the specific venue ("the Hilton")?

  • Who appoints or elects the members of the steering committee?

  • Are terms on the steering committee members limited or do members stay until they decide to quit or are permanent no-shows? (Many steering committees have members that are no longer actively engaged with the conference; having steering committee members that do not attend the conference is a bad sign. Term limits also avoid the awkwardness of having to ask somebody to step aside.)

  • Are there ex-officio members of the steering committee, such as previous chairs or representatives of the sponsoring SIG or society?

  • Does a steering committee member sit in on the conference calls of the organizers and participate in the mailing list discussions of the organizing committee?

Based on my informal survey among faculty in my department, steering committees across computer science differ dramatically in their responsibilities. In some cases, the steering committee essentially hands the reign to the general chairs for the year, and then does not discuss the conference again until it is time to pick the next chair. In others, the steering committee has much more hands-on involvement.

To further complicate matters, some conferences, such as ICNP, have advisory committees. It is important to distinguish their roles and responsibilities. For example, advisory committees may provide guidance, but may not have direct influence on conference location or chair selection, or may simply provide external "gravitas".

While there is a certain amount of flexibility, the basic conference model, discussed earlier, limits what is feasible. As in all organizations, it is probably a bad idea if a group has responsibility without power, or vice versa. For example, if the SIG budget is on the line if a conference sponsored by the SIG runs a deficit, it is hard to expect the elected officers of the SIG to simply leave the finances completely to the conference organizers or some other organization and hope for the best. Also, given that the SIG cannot simply disown their flagship conference, the SIG leadership has some responsibility to maintain quality and other core attributes of the event year after year. For example, having the organizers dramatically raise registration fees may be considered detrimental to the community represented by the steering committee, and thus, it may have to intervene.

There are many ways to run a successful series of conferences, both technically and logistically, but even successful conferences can do better if all four legs of the conference organization work under a common set of assumptions. As many conferences approach runs of twenty or more years, writing down these assumptions may work better than relying purely on oral traditions.

November 05, 2008

SIGCOMM Elections

SIGCOMM will hold officer elections in early 2009 for Chair Vice-Chair, and Treasurer.

The current set of SIGCOMM officers took office July 1, 2007. Officers' terms are two years, and ACM bylaws provide for their terms to be extended for another two years if the officers agree and the SIG Governing Board Executive Committee confirms them. However ACM bylaws also state that an election must be held if 1% of the SIG members petition for one (SIGCOMM has just over 1800 members). The current officers have agreed to extend their terms, but a valid petition has been filed with ACM requesting an election. Hence we will hold elections in early 2009 for officers with terms beginning July 1, 2009.

ACM guidelines direct the Chair of the SIG to appoint a Nominations Chair, who appoints a committee to propose at least two candidates for each office by December 15. Self-nominations are also possible via a petition process by 1% of the SIG; these must be received by March 15.