SIGPLAN FY'04 Annual Report
July 2003 - June 2004
Submitted by: Michael Burke, SIGPLAN Chair
This year ACM SIGPLAN has continued its active sponsorship of many conferences and workshops as well as its two newsletters.
SIGPLAN's present financial situation is still strong, but we have been forced to address some unfortunate trends. Our fund balance still comfortably exceeds the required minimum, but has declined substantially in the past three years. OOPSLA , our largest conference (and primary source of past surpluses), incurred a significant financial loss for the third consecutive year. Our other conferences incurred slight financial gains.
We have become more selective with funding worthwhile projects such as student travel, funding these at about one half the level of recent years.
A good resource for monitoring our activities is our web page, found at http://www.acm.org/sigplan/
We sponsored six annual conferences last year, OOPSLA, POPL (with SIGACT), PLDI, ICFP, PPDP, and LCTES. We also sponsored PPoPP, PEPM, and ISMM, which are held approximately biannually. We also sponsored SOFTVIS (with SIGGRAPH and SIGSOFT), Haskell, and SAS.
Of these conferences, PLDI, ASPLOS, POPL and PPoPP appear in the Citeseer top 15 of more than 1200 Computer Science publication venues, based on their citation rates.
PLDI, PEPM, LCTES, PPoPP, SAS, and SOFTVIS were all part of FCRC 2003. ICFP, PPDP, and Haskell were co-located and collectively called PLI. Additional workshops were co-located with POPL, PLDI, and PLI.
Financial results for our conferences other than OOPSLA were positive. There was an upward trend in conference attendance, in particular at the conferneces that were part of FCRC.
Conferences continue to receive far more submissions than we can accept, and our major conferences continue to be extremely selective.
We now have separate steering committees for all of our conferences.
SIGPLAN publishes two newsletters on a regular basis.
SIGPLAN Notices is sent to all SIGPLAN members monthly. Eight of these issues contained conference or workshop proceedings. The editors, Cindy Norris and Jay Fenwick, of Appalachian State University, have continued to arrange for SIGPLAN Notices to run very smoothly.
For FY 2005, we plan to limit the number of issues containing proceedings to six issues.
Our other regular newsletter, Fortran Forum, published three issues, as scheduled. Michael Metcalf served as its editor. The editor for FY 2005 will be Ian Chivers.
John Backus received the 2004 SIGPLAN Programming Languages Achievement award. He is one of the founders of the field of programming languages, and is most widely known for conceiving of and leading the team that developed FORTRAN.
Ron Cytron received the 2004 SIGPLAN Distinguished Service Award for his many contributions to the SIGPLAN community, including five terms in all positions of the SIGPLAN Executive Committee.
Godmar Beck received the 2003 SIGPLAN Doctoral Dissertation Award for his thesis entitled "Isolation, Resource Management and Sharing in the KaffeOS Java Runtime System" (University of Utah, May 2002).
Hans Boehm received the award for the 1993 Most Influential PLDI Paper Award for his paper "Space Efficient Conservative Garbage Collection".
IV. Activities and Programs
We continue our Professional Activities Program, PAC, which was instituted a number of years ago, primarily to provide funding to graduate students who participate in a SIGPLAN sponsored conference. This year we limited grants to $10,000. We awarded 17 grants, all to support travel by student authors of papers being presented at SIGPLAN-affiliated conferences or workshops. We capped individual awards at $500, to extend the allocated $10,000 over more grants. This worked out well: we didn't have to turn down any grant requests, and awarded all but $450 of the allocated funds. Most of the grants (14/17) were for this $500 maximum amount. The large majority (14/17) of the grants were given to students at US Universities, but we also were able to help students in Denmark, Singapore, and Canada. No individual institution had more than two students receiving grants.
We continued but reduced funding to sponsor travel and attendance of professors from 2-4 year colleges at the OOPSLA educator's symposium.
V. Membership Changes
For FY 2005, we instituted several changes to SIGPLAN membership and our dues structure. Membership rates for FY 04 were $35 a year for professional members and $15 a year for student members. For FY 2005, the rate of the online SIGPLAN membership, which includes an electronic (instead of a print) subscription to SIGPLAN Notices, is $25 per year for professionals and $15 a year for students. The print membership has been increased to $50 per year for professionals and $40 a year for students. In the past we have subsidized SIGPLAN memberships with surplus conference revenue. Since we have not had such a surplus for the last three years, we can no longer continue this pratice. We do not lose money on members who choose the online option. In that the cost to SIGPLAN of a print subscription to SIGPLAN Notices is at least $60 a year, SIGPLAN is still losing money on print memberships. The rate for print memberships will probably have to be increased again for FY 2006.
Starting in FY04, an annual SIGPLAN CD-ROM became a member benefit. The SIGPLAN 2004 CD was enclosed with the May 2004 issue of SIGPLAN Notices. Covering the year 2003, the CD contains the proceedings of all SIGPLAN sponsored or co-sponsored conferences, all issues of SIGPLAN Notices, all volumes of TOPLAS, and all SIGPLAN-sponsored workshops that were published by the ACM.
VI. Key Issues for Next 2-3 years
The OOPSLA 2003 financial shortfall is a major concern to both the SIGPLAN Executive committee and the OOPSLA Steering committee. This was the third consecutive year that OOPSLA incurred a significant loss.
OOPSLA 2004 has been budgeted for a minor loss, but poses a challenge in that many commitments predate fall 2001, and assumed a larger conference. OOPSLA 2005 will be budgeted to break even and we expect it to do so. For 2005 the hotel space/room commitment does not assume an increase relative to recent OOPSLA attendance. Further, the SIGPLAN EC and the OOPSLA SC have worked together to produce a set of requirements for the OOPSLA 2005 budget, which will decrease expenses.
We hope that electronic memberships will make it possible to avoid losing money overall on memberships. If this does not happen, we will need to rely on future price increases for paper subscriptions to achieve this goal.
SIGPLAN's fund balance allows us to continue to fund our existing programs in the short term at about one half the level of recent years. We clearly need to move towards a balanced budget, and our FY2005 budget is close to that.
The relative lack of new SIGPLAN members continues to be an issue. We will be exploring ways to attract new electronic-only members.
We are receiving significantly more requests for SIGPLAN proceedings publication in SIGPLAN Notices than we can handle with our current budget. We will move some of these to electronic-only supplements, which will be included in the Digital Library and on the year-end CD-ROM.
ACM's prestigious conferences and journals are seeking top-quality papers in all areas of computing and IT. It is now easier than ever to find the most appropriate venue for your research and publish with ACM.
ACM Queue’s “Research for Practice” is your number one resource for keeping up with emerging developments in the world of theory and applying them to the challenges you face on a daily basis. In this installment, Dan Crankshaw and Joey Gonzalez provide an overview of machine learning server systems. What happens when we wish to actually deploy a machine learning model to production, and how do we serve predictions with high accuracy and high computational efficiency? Dan and Joey’s curated research selection presents cutting-edge techniques spanning database-level integration, video processing, and prediction middleware. Given the explosion of interest in machine learning and its increasing impact on seemingly every application vertical, it's possible that systems such as these will become as commonplace as relational databases are today.