NAACL-2001 General Chair Lori Levin We planned this conference as a small, high-quality gathering. Having picked a date that was only one month away from ACL and EACL, and not being a joint conference, we expected attendence to be between 300 and 400 people. We had three tactics to draw more people: co-locate with EMNLP, focus on diversity (try to attract papers on areas not well represented at ACL meetings), and have an industry exhibits session. Attendance turned out over target -- 460 people are currently pre-registered. Detailed reports from the sub-chairs are attached. Here is a summary of the highlights and problems. Submissions and Attendance: --------------------------- The number of submissions was low for the main session, but attendance is higher than expected. Low submissions could be attributed to getting a late start on publicity, or just that the submission deadline was at an unusual time. High attendance could be attributed to several factors: workshops, tutorials, and EMNLP; a huge number of computational linguists, speech, and machine learning people in living in Pittsburgh; a large number of demos; a huge amount of contact with industry (even though they were reluctant to sponsor or exhibit); quality of the program; quality of the web site; CMU's reputation (#2 in Computer Science). Budget: ------- When we saw that submissions were low and that sponsors withdrew (because of the economy), we raised the registration fee in order to make sure that the budget balanced, and are now showing a surplus in the budget because of high enrollment. Publications: ------------- Some of the publications (the companion volume and program brochure) were put together at the last minute, due to lack of planning on my part. As a result, the companion volume is not very beautiful, but at least contains all of the relevant information. I hope that the program brochure does not contain mistakes. The production of the CD ROM was also rushed. On-Line Registration: --------------------- The publicity chair report below contains some suggestions for making it easier to implement on-line registration in coming years. (It is in the publicity report because it was handled by the web master.) Exhibits Session: ----------------- Organizing the exhibits session and attracting sponsors was much harder than we expected. We came up with the idea of an industry exhibits day during NAACL 2000 where industry made a very strong showing and was recruiting heavily. The idea of the exhibits session was for the academic ACL crowd to find out about the companies that were springing out of the woodwork, started by people that are not in our field. There were two problems. One is that the economy went bad and many companies had to cut back or even close. The other was that some companies weren't as interested in meeting us as we were in meeting them. If they weren't recruiting, or if their target customer base wasn't attending, they didn't want to come. One company asked me to invite venture capitalists and even drafted a letter for me to send to them. (I apologized and didn't do it.) Sponsorships: ------------- For the same reasons, sponsorships were lower than expected. NAACL 2000 had over $50,000 in sponsorships. This year we went under even the ACL 99 level ($18,000 in 1999) with $15,000 in sponsorships. In addition to the $15,000, the CMU School of Computer Science organized the opening reception and will pay for it on an as-needed basis. The Student Research Workshop also brought in an NSF grant. EMNLP's budget is also relevant in that if EMNLP has a surplus and NAACL doesn't, EMNLP's surplus will go to NAACL. (Currently both are showing a surplus.) EMNLP has a $6,000 sponsorship and a free reception. MAIN SESSION SPONSORSHIPS -------------------------- IISI (Cornell) $5000 Mitsubishi Electric $1000 SRA $1000 AT&T $1000 Nuance $1000 Lingomotors $1000 Lexis-Nexis $1000 ISL (CMU) $1000 LTI (CMU) $1000 Trados $1000 Clairvoyance $1000 ------- $15,000 School of Computer Science, CMU, will fill in money as needed. EMNLP SPONSORSHIPS ------------------- IISI (Cornell) $6000 Whizbang free reception at Whizbang STUDENT RESEARCH SESSION SPONSORSHIPS ------------------------------------- NSF ???? In addition, the AT&T sponsorship for the main session is earmarked for student use. ****************************************************** Report from the Student Research Workshop Chairs Krzysztof Czuba (CMU) and Lisa Michaud (University of Delaware) The co-chairs of the NAACL-2001 Student Research Workshop, Krzysztof Czuba (Carnegie Mellon University) and Lisa Michaud (University of Delaware), were nominated by last year's co-chairs, Donna Byron and Peter Vanderheyden, and approved by the NAACL executive committee. We decided to continue last year's extended presentation format with feedback from assigned panelists, in emulation of many Doctoral Consortium programs at other conferences. We also added to this program a poster session, in order to give our presenters a broader exposure to the conference at large. We selected a Program Committee consisting of 13 students and 5 non-students recruited from last year's NAACL papers, student participants, and personal recommendations. The 5 non-student members, including 3 academics and 2 from industry, were approved by the NAACL Executive Committee. Fifteen papers were submitted to the workshop and were subjected to review panels consisting of at least two students and one non-student reviewer (some papers had three student reviewers). We originally accepted seven papers for presentation in the workshop, but one of those papers was withdrawn after the author informed us that it had been accepted for presentation at another conference. The six authors presenting at our workshop are evenly divided between men and women and are affiliated with universities in the United States, the United Kingdom, and Japan. They will each speak for 25 minutes, after which two panelists assigned to each paper will have time to deliver prepared feedback. The floor will then be opened to general questions. In addition to this specific panel format, we will be holding a general career panel at the conclusion of the workshop. We have invited all of our panelists to participate and to offer their perspectives on work in academia, industry, and government in order to aid the student audience as they prepare for their future careers. The concluding poster session will take place near the demonstrations and exhibits and will allow general conference attendees additional exposure to the student work. We have applied for an NSF grant to cover the participation costs for the workshop presenters and organizers. The grant budget includes the registration, travel and housing costs as well as the banquet ticket costs. In addition we have requested funds to cover a part of the costs of producing the proceedings. The grant has been recommended for full funding but it is still being processed by the NSF. In addition to the NSF grant, we have obtained funds from the conference organizers to cover part of the Student Business Lunch and the lunch for the SRW participant during the workshop. We would like to acknowledge the assistance we have received from the conference organizers that helped us to prepare a successful workshop. Total number of submissions: 15 Submissions by Country (based on authors' first affiliation): US 6 40.00% Canada 2 13.33% Malaysia 2 13.33% Algeria 1 6.66% Estonia 1 6.66% Japan 1 6.66% Portugal 1 6.66% UK 1 6.66% Submissions by Geographical Area: North America 8 53.33% Europe 3 20.00% Asia 3 20.00% Africa 1 6.66% Total number of papers accepted: 6 Acceptance Rate: 37.5% North America 66.6% Europe 33.3% Asia 0% Africa Acceptances by Geographical Area: North America 3 50% Europe 2 33.3% Asia 1 16.6% Submissions by Gender (subject to error): 47.05% Male 52.95% Female Acceptances by Gender: Male 3 50% Female 3 50% Submissions by Topics: 1 Syntax and parsing 15% 2 Knowledge acquisition/extraction 5% 3 Generation - 4 Statistical language processing 10% 5 Natural language systems 10% 6 Discourse and pragmatics - 7 Grammar 5% 8 Semantics 5% 9 Corpus Analysis 5% 10 Machine translation 10% 11 Speech - 12 Other (Document Categorization) 10% 13 Other (Question Answering) 15% 14 Other 10% Accepted papers by Topics: 1 Syntax and parsing 22.2% 2 Knowledge acquisition/extraction 11.1% 3 Generation - 4 Statistical language processing 22.2% 5 Natural language systems - 6 Discourse and pragmatics - 7 Grammar - 8 Semantics - 9 Corpus Analysis 11.1% 10 Machine translation 11.1% 11 Speech - 12 Other (Document Categorization) 11.1% 13 Other (Question Answering) 11.1% 14 Other - **************************************************************** NAACL 2001 WORKSHOPS REPORT Lillian Lee, Cornell University Five workshop proposals were submitted, and all were accepted. After the fact, two (one on WordNet and one on extending and customizing lexical resources) were merged, after some discussion initiated by the SIGLEX community. [This does not count the Student Research Workshop, which the student organizers handled independently]. The pre-registration numbers are as follow (courtesy of Alon Lavie): [one-day] Automatic Summarization: 56 [one-day] MT Evaluation: 17 [one-day] Adaptation in Dialogue Systems: 55 [two-day] WordNet: 99 Overall, things went relatively smoothly, thanks in large part to the efforts (and patience!) of Bob Frederking, Jodi Hauck, Alon Lavie, and Priscilla Rasmussen. A few suggestions for the next time: 1. As suggested by Scott Miller (previous workshops chair), it is best to have clear as early as possible how much is budgeted for each workshop, what page limits/cost factors come into play for the workshops proceedings, and what the various deadlines (such as camera-ready due dates) are. PDF files were requested for the CDROM, which seems like a good idea, but notification of the request came a bit late. 2. On the various registration forms/media, it might be good to make clear whether or not registration for the workshops without registering for the main session is "allowed". 3. A multiple submission/publication in workshops and "main conferences" policy should probably be put in place (this year, the ACL exec decided that double publication in an NAACL workshop and ACL is not permissible. Should this or something else be made a permanent policy?) I understand this is a controversial topic, though. -- EMNLP -- http://www.cs.cornell.edu/home/llee/emnlp.html This meeting combined SIGDAT's two usual annual meetings (the Conference on Empirical Methods in Natural Language Processing and the Workshops on Very Large Corpora) into a single two-day conference preceeding NAACL 2001 and co-occurring with the workshops. The program committee contained over 40 members, including the Co-Chair, Donna Harman, and Publication Chair, David Yarowsky. Partial support came from a very generous donation from the Intelligent Information Systems Institute (IISI), which is a joint Cornell University/Air Force Research Labs organization. The pre-registration count was 163 people. The issue was raised about the co-scheduling of EMNLP with the NAACL workshops, with the concern being that this could potentially hurt the workshops. I am not sure what the effect was. The number of proposals and perhaps the registration numbers seems to have been down this year, but this may have been affected by the closeness in timing to the ACL meeting in Toulouse. ***************************************** Report from Tutorial Chair Dekang Lin, University of Alberta Proposals for tutorials for ACL 01 were solicited by a public announcement, and by a couple of private invitations. 11 proposals were received by January 2001. 9 were from the North America, 1 from Europe and 1 from Asia. The general topic areas of the proposals were as follows: Machine Translation: 2 Speech: 4 Language Technology Start-up: 1 Logic and Symbolic NLP: 2 Statistical Language Processing: 1 Question Answering: 1 Four proposals were selected: 1. "How May I Help You?": Automated Customer Service via Natural Spoken Dialog. Alicia Abella, Allen Gorin, Guiseppe Riccardi, Tirso Alonso, Jerry Wright, AT&T Shannon Laboratory 2. Empirical Methods in Natural Language Processing: What's Happened Since the First SIGDAT Meeting? Kenneth Ward Church, AT&T Labs-Research 3. Building Synthetic Voices. Alan W Black and Kevin A. Lenzo, 4. Open-Domain Textual Question Answering. Sanda Harabagiu and Dan Moldovan, Southern Methodist University *********************************************************** Report from the Demonstrations Chair Ronnie W. Smith, East Carolina Univeristy In the spirit of outreach to related communities, my philosophy with regards to acceptance of demos was to be very open-minded to work related to issues studied in Computational Linguistics (provided the work appeared to be of academic interest) in order to broaden participation in the conference. Consequently, all 29 submitted demos were accepted for presentation, however we have had two withdrawals up to this point (May 25). Approximately 50% of these demos are related to speech and/or dialogue. Twenty of the 27 demonstrators are from North America (the two who have withdrawn were from Europe), but I am encouraged that there are several demonstrations coming from Europe even with ACL being held in France. In terms of initial recruiting of demos, I attended the AAAI Fall 2000 Symposium on building dialogue systems for tutorial applications and extended invitations at that meeting. Several attendees from that symposium are giving demonstrations. Otherwise, normal advertising techniques were used (i.e., society mailing lists). A few personal invitations were extended, some submitted while others did not. In terms of scheduling, the decision was made to run the demo sessions roughly in parallel with the first 9 regular sessions of the conference. In order for there to be equity among the sessions, they have all been set at 75 minutes in length, and in order to facilitate setup time, some sessions overlap slightly with break periods. All demos are offered as informal walk-in, small group format. In terms of logistics, it is important to solicit information from the demonstrators early and frequently, and to keep a steady flow of communication with local arrangements to maximize the probability that the demonstrations will go smoothly. We of course will not know the outcome of our efforts for another two weeks, but I have nothing but the highest regard for the work of Local Arrangements chair, Alon Lavie, with regards to coordination of logistics for the demos. To summarize, I believe the key logistics issues for planning demonstrations at future conferences are the following: * Availability of large monitors to improve displays of the notebook * and laptop machines that most demonstrators bring * Availability of Internet connections (a small but significant number * of demonstrators require this capability) * Frequent communication between Demonstrations Chair and Local * Arrangements Chair with regards to planning * Proximity of demonstrations area to the regular presentation room. * We are fortunate for this conference to have them very near each * other, facilitating ease of movement for conference attendees * between paper presentations and live demonstrations. *********************************************************** NAACL-2001 INDUSTRIAL EXHIBITS Lynn Carlson, Department of Defense A new focus for this year's conference was the idea of giving a prominent role to industrial exhibits, by sponsoring a full day event designed to showcase the latest commercial trends in language technology. The Process I worked very closely with the Sponsorships Chair, Kurt Godden. This arrangement was very beneficial for both efforts. We not only shared all our contacts and leads, but also made companies aware of the opportunity to sponsor as well as to exhibit. By doing so, we were able to attract some companies originally intending just to sponsor to become exhibitors, and vice versa. I concur with Kurt Godden's recommendation that the sponsorships and exhibits committees be combined in the future, with two co-chairs. Kurt and I originally sent out a solicitation letter for sponsors and exhibitors to about 100 recipients in October 2000, including organizations that had previously sponsored ACL events. A number of reminders were then sent out periodically over the next several months to the same general mailing list. In addition to this broad distribution email list were the following approaches: 1) personal emails and phone calls to known colleagues 2) "cold calls" to companies, with requests to speak to the marketing manager 3) an on-line application form for the exhibits session on the NAACL Web site 4) emails sent to the "info@" addresses of companies that advertised in Multilingual Computing In the end, we were able to attract the following companies, academic institutes and book publishers to the Exhibits Day Session: Applied Technical Systems Apptek Blackwell Publishing Cambridge University Press Carnegie Mellon University, School of Computer Science Cornell University, Intelligent Information Systems Institute Intel John Benjamins Publishing Company Kluwer Academic Publishers LexiQuest LingoMotors Multicorpora R&D Inc. Nuance Tata Infotech, Ltd. Trados Corporation Transclick Lessons Learned The amount of time invested in preparing for this exhibits session was considerable. I would advise future Exhibits and Sponsors committee members to start early and be persistent. In my experience, the least effective results stemmed from the generic emails. People are extremely busy, and these emails are easy to ignore. Phone calls and personal emails were much more effective, and the on-line application form also drew in a few applications from companies we had not originally contacted. Surprisingly, we acquired two exhibitors from the emails sent to the "info@" addresses of Multilingual Computing advertisers. Follow up emails and phone calls were an essential part of the process. The turndown in the economy had an effect on our ability to attract exhibitors, as well as sponsors. When the idea for the Exhibits Day session was first conceived (summer 2000), it was hard to imagine that we would have any problems attracting companies. However, the reality was that many companies that originally expressed keen interest in both the exhibits and sponsorships had to pull back for financial reasons. The economy also had an impact on the number of book publishers we were able to attract this year. It is my hope that the success of this year's event will be a catalyst for an even more successful event next year. ********************************************************** Report from the Sponsorship Chair Kurt Godden, Justtalk Sponsorship report: The sponsorship committe consisted of Arendse Bernth, Jeff Allen, and Kurt Godden (chair). As of this writing, we have actually received money from only five Bronze sponsors, at $1k each. These five are: MERL, Nuance, SRA, AT&T Labs, and Clairvoyance. We *expect* to receive additional contributions from: LingoMotors (Bronze, $1k), TRADOS (Bronze, $1k), and Lexis-Nexis (Bronze, $1k). Clairvoyance originally planned to be a Silver sponsor, but had to scale back due to economic reasons. This was actually a strong theme for the year, with two companies originally expecting to be Gold sponsors, but pulling out altogether due to the weak economy, and several other companies having similar cutbacks. I am disappointed, since I set a goal for myself to obtain more sponsorship money than last year's conference, but ended by obtaining far less. All of last year's sponsors were solicited to become repeat sponsors, even to the point of personal communcation, but with little success. Some of this may be locale related, since, for example, Microsoft and Boeing were sponsors last year when the conference was in Seattle, but are not repeat sponsors this year. I believe it is very helpful to have the exhibits and the sponsorship chairs work together, as Lynn Carlson and I did this year, beginning with our original solicitation announcement. Offering an exhibits discount to sponsors is also helpful. Some organizations originally contacted Lynn to become exhibitors and ended up also becoming sponsors, so I would encourage future chairs to continue working together. It may even be beneficial to have co-chairs of ONE committee on sponsorhips and exhibits, but I would not ask one person to perform both duties. It also certainly helps to send repeated reminders to contacts, once they have expressed an interest in sponsorship, although you have to be careful not to be perceived as a spammer. We sent out one mass mailing, then a second reminder. After that, it was limited to people who expressed an interest. You can give a deadline to sponsors to sign up and contribute, but the deadline is not enforceable! And if it were enforceable, you would end up with fewer sponsors because organizations wait until the last minute to actually send their money. One difficulty involves communications. Even though my instructions to sponsors clearly stated that I wanted them to inform me when they sent their checks to Priscilla, no one did. Therefore, I had to continue asking Priscilla to let me know who had sent her money, and how much. As a way to improve this, I would suggest that we accept the fact that sponsors will not bother to inform the chair of their payment. Rather, we ask Priscilla to send a quick email to the chair notifying them as checks are received. She ends up doing this anyway when the chair has to bother her, and since she is so remarkably efficient anyway, I am hoping this would not be perceived as a burden by her. ******************************************************** NAACL-2001 Publicity Robert E Frederking and Ralf D Brown, Carnegie Mellon University 5/28/2001 Website and Related Issues -------------------------- The conference webmaster (REF) was kept fairly busy keeping the website up-to-date. I believe this was successful, and that it was also reasonably attractive and easy to navigate. I made a serious attempt to make sure the website was equally usable under Netscape and IE, and from PC, Mac, and Unix platforms. Due to concern about possible negative pre-conceptions about visiting Pittsburgh, we produced a "Local Attractions" page early in the process, highlighting the nice qualities of the area, and the current lack of a steel industry. A web form was created for the "Call for Exhibitors" page, to accept and mail contact information to the Exhibits chair. Fairly late in the process, a shorter alias for the webapge was created, at the request of the Carnegie Mellon publicity person (Anne Watzmann). I believe the only serious difficulty involving the website was the Online Registration. This was very late coming online (after the end of Early Registration) for several non-technical reasons, plus two technical ones: the changes required to move from one conference to another were very significant, and the NAACL webmaster did not have direct access to the ACL website (this latter added about three weeks to the delay). If this package is going to continue to be reused for ACL-sponsored conferences, it would be worthwhile to redesign it to be more modular. (I have added notes to the NAACL version with suggestions. The main problem is that, currently, four separate source files need to have corresponding extensive changes made to them for events, dates, and prices; these might all be auto-generated from a single database in a redesigned system.) Otherwise a full student volunteer for just online registration will be needed for each conference, and they should be given direct access to the ACL website. In addition to conference-specific information changes, I: -- Stripped out the Javascript due to security and modifiability concerns; I moved validation into radio-buttons and Perl, and moved the early/late distinction into separate webpages, requiring a manual link change at the end of Early Registration and at the end of Late Registration. -- Changed the turquoise-on-gray color scheme. Also made the form look reasonable on non-Microsoft platforms. -- Changed the CGI mechanism to work with the CGI package installed at Carnegie Mellon, for testing (which then had to be converted back again at ACL). A second recommendation for future conference Webmasters (and Sponsorship chairs): it would be good to acquire high-resolution logos from the sponsors at the same time as the logos for the website are acquired. Website logos need only 72 pixels per inch, whereas any other publicity use (involving printing) requires much higher, typically at least 300 pixels per inch. This might be obvious, but was not obvious to us. Other Publicity --------------- Carnegie Mellon School of Computer Science development person (Matt Hughes) and publicity person (Anne Watzmann) helped in some fashion, but I (REF) was not directly involved in most of it. (Anne Watzman sent out a press release to local and national news agencies. So far the NY Times and the Boston Globe have called asking for stories about language technologies, though they won't cover the conference.--LL) Ralf Brown posted the conference announcement on mailing lists and news groups that were not on Priscilla's list. We had a glossy conference brochure that was handed out at ACL, ICSLP, and AMTA. We also handed something out at HLT. For ICSLP we added a special note on the cover of the brochure, "speech and language papers welcome". For AMTA the note said "MT papers welcome". ========================================================== NAACL-2001: Local Arrangements Report Alon Lavie, Local Arrangements Chair Local arrangements for NAACL-2001 started about 12 months in advance of the conference date, and have generally progressed smoothly throughout the year of planning and preparation for the conference. Preliminary planning focussed on putting together a local team of area coordinators, determining the general and preferred venues for the conference events, and setting up a working relationship with the office of conference services at Carnegie Mellon. At the point that our proposal to host NAACL-2001 at CMU was accepted, we had already determined that all conference sessions would take place at CMU's University Center (UC), and, based on availability constraints for the UC, that the proposed dates for the conference would be the first week of June. We then held a preliminary meeting with Lorrie Safar - the director of CMU Conference Services, and decided to hire the services of her office as a central logistics coordinator with all internal and external venues and vendors for the conference. The main advantages of working through CMU Conference Services were: - Organizational: central coordination with all needed vendors by a knowledgeable professional familiar with the vendors and venues. - Financial: An internal CMU account was opened for NAACL-2001, allowing all related expenses to be centrally charged and payed off after the conference. - Preferred Rates: Since all financial commitments were done via CMU, we got to enjoy preferred rates offered to the university and tax free status. - On Campus Housing: Lorrie's office is responsible for arranging all on-campus housing for outside events, and this service was thus naturally added to their responsibilities. We next organized a local two-day site visit in mid July-2000 to review the main conference plans. Participants included: Diane Litman (NAACL President), Priscilla Rasmussen (ACL Business Manager), Lorrie Safar (CMU Conference Services), Lori Levin (conference General Chair) and Alon Lavie (conference LA Chair). We toured the meeting location site at the UC, reviewed tentative meeting rooms and event locations, and toured the proposed banquet location at the Carnegie Museum of Natural History. Priscilla and Diane stayed at the proposed conference hotel - the Holiday Inn University Center. In early fall, we assembled a local team of area coordinators, and divided the main responsibilities: - NAACL-2001 Website: Bob Frederking - Publicity: Ralf Brown - Email and Networking: Alan Black - CD-Rom Production: Eric Nyberg We also assembled a team of local student volunteers to assist with pre-conference work in the various areas, and began holding regular local arrangement meetings throughout the fall and winter. Main student area coordinators: - AV coordinator: Greg Aist - On-site Conference Student Volunteer coordinator: Kenji Sagae - T-shirt and Music Band: Ariadna Font-Llitjos - Local Information: Chad Langley - CD-Rom: Stan Jou - Signs: Stefanie Shriver - Misc Projects: Benjamin Han This team continued to meet regularly during the spring months leading to the conference (weekly in the two months before the conference). Coordination and contracting with all internal and external vendors was handled for the most part via Lorrie Safar, including: - UC space and facilities - Hotel Accommodations at the Holiday Inn - On Campus Housing Accomodations - CMU IT for all AV equipment - CMU Catering for all on-site food - Banquet hall rental and catering at Carnegie Museum - Vendors for Bags, Badges, T-shirts Late in the planning stages (April-2001), the School of Computer Science Development office came to our assistance in organizing the conference reception, some help with PR and with an offer for financial backing, in case of possible budget deficits. With their help, we moved the conference reception to the School of Computer Science, and put together a nice and impressive reception combined with an open-house of the Language Technologies Institute. Main Problems and Lessons Learned: (1) Some tasks should have been assigned to local coordinators earlier. One obvious example is the setup of online registration. Locally this fell on the shoulders of Bob Frederking, the web maintainer. We should have assigned this earlier on to someone with more time. Problems with getting online registration up in time were not just local in nature - this *really* requires some central streamlining from the ACL. See also Bob's report. (2) We relied a bit too much on the services of CMU Conference Services. Lorrie Safar did a great job overall, but was over-committed and was not available as much as I would have liked in the critical last couple of weeks before the conference. (3) Difficulties in accurately estimating the number of projected attendees. We estimated much too conservatively, and when we lost most of our tentative sponsorship money, we were forced to raise registration fees. This turned out to be completely unnecessary at the end. See details in the financial report. (4) Overall, I feel that we did a good job of structuring and distributing the huge work load of preparing for the conference. The clear division of responsibilities among the various team coordinators and frequent timely coordination meetings kept the entire process both manageable and effective. I worked with a collection of competent and very dedicated people to which I owe many many thanks. I could have never done this job without them! NAACL-2001 Financial Report Alon Lavie, Local Arrangements Chair The financial picture for NAACL-2001 was particularly difficult to predict and has changed significantly during the planning process and the months leading to the conference. We went from a feared deficit of $15K in March to our current estimate of a surplus of $28K. While our cost estimates were predicted with a fair degree of accuracy from early on in the process, income predictions changed dramatically due to two main factors: (1) Tentative sponsorship money in the order of $20K evaporated in the Spring months due to the collapse of the high-tech industrial market. (2) Projected attendance levels were initially low (~300 people), but proved to be extremely below actual registration levels -- pre-registration ended with ~450 registered participants. This completely changed the income picture, but late in the game. Registration fees had been fixed based on the predicted low numbers. The way this played out was the following: We had initially planned a balanced draft budget, based on the predicted low number of participants, and a tentative level of $35K in sponsorship contributions. We calculated proposed registration fees based on these estimates (see details below). The loss of $25K in pledged sponsorship money put this budget into an alarming predicted deficit of $15K. After consultations with members of the NAACL Exec, we revised proposed registration rates by raising them approximately $25 across the board. This resulted in a revised budget with still a deficit of $9K. Pre-registration numbers then exceeded all our expectations. By the end of pre-registration, we had ~450 participants registered. Revising both income and expenses based on these new figures results in a budget surplus projection of up to $28K. Some expense categories are still based on estimated figures, and there were some additional unexpected expenses in the last few days of preparation for the conference. I believe we will end up with a final budget showing a surplus of about $20K. Both Barbara Di Eugenio and Kathy McCoy reviewed our draft budgets throughout the planning process and, assisted us with some suggestions, participated in the decision making process, and approved the draft budgets. For comparison purposes, here are the initial proposed registration fees and the registration fees that were eventually adopted: Initial proposed fees: - Regular ACL member $250 (early) $300 (late, on location) - Regular Non-ACL member $310 (early) $370 (late) - includes ACL memb dues - Student ACL member $100 (early) $125 (late) - Student Non-ACL member $130 (early) $155 (late) - includes ACL memb dues - Workshops (each, 1 day) $50 (early) $75 (late) - EMNLP (2 days) $100 (early) $150 (late) - Tutorials (each, 1/2 day) $100 (early) $125 (late) student $75 (early) $100 (late) - Banquet tickets $65 (regular) $40 (student) Final registration fees: - Regular ACL member $275 (early) $325 (late) $375 (on site) - Regular Non-ACL member $335 (early) $385 (late) $435 (on site) - Reg Student ACL member $120 (early) $145 (late) $170 (on site) - Reg Student Non-ACL member $150 (early) $175 (late) $200 (on site) - CD-Only Student ACL member $100 (early) $125 (late) $150 (on site) - CD-Only Student Non-ACL member $130 (early) $155 (late) $180 (on site) - 1-day Workshops Reg Mem $75 (early) $100 (late) $125 (on site) - 2-day Workshops Reg Mem $125 (early) $150 (late) $175 (on site) - EMNLP (2 days) Reg Mem $125 (early) $150 (late) $175 (on site) - 1-day Workshops Student Mem $50 (early) $75 (late) $100 (on site) - 2-day Workshops Student Mem $100 (early) $125 (late) $150 (on site) - EMNLP (2 days) Student Mem $100 (early) $125 (late) $150 (on site) - Tutorials (each, 1/2 day) $125 (early) $150 (late) $175 (on site) Tutorial Student $75 (early) $100 (late) $125 (on site) - Banquet tickets $65 (regular) $40 (student) A copy of the current conference budget is available upon request.