Useful checklist of things to do (from a past student chair)

From Admin Wiki
Jump to: navigation, search

Checklist for Planning and Running a student Session

Conference Handbook - Student Session Checklist

The following document provides a useful set of suggestions for organizing student research workshops from past organizers.

The outline below contains a checklist for activities occuring in each phase. You should probably check with the ACL or NAACL Exec and find out which items they want to have final approval over (it may be different than what is listed here). In general, you can do things as far ahead as you want, unless the chart indicates that you have to wait for something. Issues and decision points relating to some phases are on their own webpages (links are in the chart).


Getting Started:Things to do as soon as you're selected

  • Start a list (and keep it up to date) of things you did and when, as well as any comments you might have about how they could be done differently/better -- this will come in very handy when you're writing things up later!

  • Set up a website (start by copying a previous year's site).

  • Make sure the main conference website has a link to your website and prominently advertises the SRW.

  • Make sure the ACL web site links to your website.

  • Decide on the format to be used (poster session versus talks, integrated into main conference or workshop days), the timeline and reviewing procedures. Deadlines should line up with the main conference to the extent possible and should not fall on religious holidays. The general chair (and possibly workshop chairs) will make the final decision about the SRW format, and will probably need to approve the proposed timeline, so you should communicate with them about these decisions.

  • Start thinking about sources of student funding. This is an ongoing process from now until the conference happens. Student funding comes from many sources and you should be knowledgeable about it even if you're not managing it because students/potential authors will ask you questions. Work with the Exec and PC to decide whether there will be volunteer opportunities for students at the conference. Volunteers typically receive technical session registration fee waivers and 1-yr membership to ACL. They don't receive complimentary registration to tutorials and workshops; however, an attempt is made to assign volunteers who indicate interest in particular tutorials or workshops to those sessions as "monitors". Work with the Exec/Treasurer to see if travel grants will be available from the organization (like Walker grants at ACL events). There should be a person on the main conference planning committee responsible for soliciting corporate sponsors for the conference. Make sure funding student events and/or underwriting student travel are included in the categories of funding he/she is suggesting to people. Keep track of information from your faculty advisors who will write and submit a grant proposal for government funding if desired (eg. NSF or European Commission).

  • Set up an opt-in email list for students to get news and communicate with each other about the session. NAACL/ANLP 2000 used egroups.com Put a link on your site so people can enroll.

  • Make sure you are aware of any official policies of the organization regarding planning the student session. For example, the ACL Exec has adopted a policy that student papers should not be bound with the main conference papers.

  • Set up a softconf page to handle submissions and reviews. Setup parameters like conference name and submission deadlines.



Writing and Distributing the Call For Papers

The CFP should include information you've pinned down as definite, and should be vague on other things that are not yet determined. As a minimum, it should have an overview of the intent of the session, the intended topic areas, and submission instructions and deadlines. It should be created as a stand-alone document with all contact information fully specified (i.e. don't just point people to the web page for details).

  • If a joint conference, decide if there is separate review/submission.

  • On choosing the general schedule, i.e deadlines for submission, reviewing, author notification, try to work with other events and conferences, so that reviewers are available and not taken by other engagements.

  • Decide a policy on joint submission to other conferences. Joint submission to student and main session should be prohibited based on the different intent of the sessions.

  • Decide if there are different types of submissions (research papers, thesis proposals ..), and the associated constraints.

  • Decide what restrictions you will place on student authors, and how this might relate to submission types/tracks: e.g., all authors must be

    • (a) full-time or not

    • (b) PhD only, Master's and PhD, all university students, any student

    • (c) pre-proposal, post-proposal

  • Decide whether you will ask for proof of student status with the submission.

  • Indicate the softconf page as the way to submit papers in the CFP The CFP must usually be vague about grant support for students, since you probably won't know about that yet when the CFP needs to go out.

  • Ask for files to be sent in a common format that is adequate to you (usually, .pdf).

  • Write the CFP, post to your website in HTML and ascii format (an 'easy-print' version)

    • The CFP must be approved by the conference program chair.

    • Wait for main conference CFP to be finished before finalizing yours. You can get topic areas and general information from it.

  • You can use a list of categories or keywords for papers (the conference topic areas may suffice; and include an "other" category); ask authors to give one or more categories for their paper; the same list will be used when asking potential reviewers for their areas of interest. Alternatively, you can use the softconf tool for reviewers to bid on papers to review.

  • The ACL business manager (currently Priscilla Rasmussen) will send out the CFP and post it to the ACL website, SIGs, and bboards. You should post to additional lists that you may know of (e.g., LINGUIST List, MT-list, Ln, comp.ai.nat-lang, Corpora, SigIR, FIRE-List).

  • After the call goes out, touch base with the person assigned to solicit corporate funding for the conference to make sure 'funding student events' is one of the categories of funding being suggested to potential sponsors.



Receiving and Reviewing papers

  • Start working on recruiting reviewers for the program committee immediately after sending out CFP.

    • The program committee as a whole needs to be somewhat balanced with respect to industry/academia, geographic (i.e. by continent), areas of interest, student/non-student, gender, etc.

    • You have to make a guess at how many papers you'll get and make sure you have enough reviewers. Try to have as many reviewers as possible, so that they can only review 2 or 3 papers (which can be a huge factor in them agreeing to be part of the committee).

    • The program committee (especially non-student members) must be approved by the Exec.

    • How to get names of people to ask? It is useful to look for people who served for previous conferences, as well as people who volunteered at the previous event, and it may be possible to get a membership list from your organization, personal contacts, etc.

  • If you plan on doing pre-submission mentoring, recruit mentors as soon as possible after recruiting the committee, so that there is enough time between students receiving feedback and the submission deadline.

  • You will need to create a review form for reviewers to use. Softconf has an interface for this, which is what you should use if reviewing is being done through softconf.

  • Be prepared to handle a lot of email traffic the week that paper submissions are due (not only submissions but you get a lot of questions/requests for information).

  • As submitted papers come in

    • Make sure papers can be printed/viewed. In some cases one co-chair can print a paper that the other can't. Make sure that papers respect the requirements and disqualify those who don’t.

    • Assign reviewers to papers. Allow yourself 1-2 weeks for this, it's pretty tricky to get all the papers to a reviewer that works in the correct area. You may not be able to depend on the keywords provided by the authors, but instead need to look over the abstract/paper and assign your own keywords. Alternatively, you can use the bidding tools for reviewers to choose their papers (a bidding interface is set up in softconf), but it does take time for reviewers to make their bids and for you to then make final assignments. 

    • Must assign student and non-student reviewers to each paper.

  • Send papers to reviewers

    • Make sure submission documents are anonymous if reviewing is blind. 

    • Softconf also has a parameter that needs to be set to ensure that reviewing is blind within the system.

    • If you make your assignments within softconf, you can just direct the reviewers to the committee page in their softconf account, which will contain their assignments and links to the review form you created in softconf. The softconf MailTool is useful for communicating with reviewers and authors, and should have a template for informing reviewers of the start of the review process. 

  • Send reminder notice to reviewers approximately one week (or a bit less) before reviews are due.

  • Receive and collate reviews. A spreadsheet is typically helpful, and you should make sure to read all reviewer comments and create meta-reviews to assist in summarizing and making decisions.


Accepting Papers

  • Before deciding how many papers to accept, you may need to negotiate the page count allowed for student papers in the proceedings. 

  • Consider whether to accept an alternate paper, in case the author of an accepted paper is unable to attend at the last minute.

  • Decide how to weigh the various reviewing categories, whether non-student reviews carry more weight, etc. Making acceptance decisions takes about 2 weeks.

  • Typically there are about 1/3 clear rejects and 1/3 clear accepts just from the numerical parts of the reviews. For the remainder, you have to read the reviews carefully, take the written comments into consideration, and read the papers if necessary.

  • One thing you may want to take into account is if the student is working at an institution where it may be difficult for him/her to get good feedback, i.e. because they don't have a strong CL program or the school has no money for students to travel to conferences.

  • Send acceptance/rejection notification to the designated contact authors. This can also be done automatically through softconf, via the MailTool. (If necessary/possible with the interface you use, sanitize reviews before emailing to authors -- remove reviewer identifying information and any comments that were specifically addressed to the co-chairs.

  • Post author instructions for camera-ready submissions.


Miscellaneous stuff to work on after the accepted papers are chosen

  • Finalize the program and post on web site. If applicable, try to keep breaks lined up with other events. The schedule along with other details will need to be sent to the handbook chair for the handbook.

  • If you plan on doing post-acceptance mentoring, try to have an idea of how big a part of the committee will be present to the conference - you may need to recruit mentors outside of the committee.


Travel Funding Recommendations and help to student authors

  • Find out how much money is available for travel grants (corporate gifts, Walker fund, NSF, etc.). Decide on the criteria for awarding grants, exactly what categories of expenses the grant will cover, will registration fees be waived, etc. (Registration fees have generally been waived in the past for volunteers, and refunded as part of the grant for travel grant recipients.)

  • Announce availability of travel grants and post applications on the website.

  • If you are in charge of allocating funding for non-SRW funding applicants, publicize the relevant funding application more broadly, e.g., on the main conference website.

  • Make sure international authors start working immediately on getting visitor visas to the country hosting the conference, and produce any supporting documents they need.

  • If you're running it as a separate 'workshop' event, make sure there will not be a charge. Make sure all versions of the registration brochure specify that the workshop is included when you pay for the main conference.



Recruiting/Assigning Panelists (if you use that format)

  • Contact people in same areas as papers. It's nice to get a mix of industry and academic people. You can do some preliminary work to get a list of potential panelists together, but don't contact people to ask them to be a panelist until the list of accepted papers is finalized and the final program is posted (because potential panelists need to know whether they have a conflict with the time).

  • How to get suggestions of people to ask? Look over the main conference program to find established researchers who are planning to attend the conference. Ask the NAACL officers and conference program committee. Ask your faculty advisor and thesis advisor for suggestions.

  • Assign panelists to papers. Two panelists per paper is sufficient, with approximately 1-2 papers per panelist. .

  • Decide whether to allow authors to submit additional information "for panelists' eyes only" (e.g., extended papers, links to websites, etc.) -- especially if proceeding versions of papers are short (e.g., 3 pages); decide on format for submission of this information, let authors and panelists know about it.


Compiling proceedings

  • If you are compiling proceedings through softconf, see the detailed instructions by Meg Mitchell here: https://github.com/acl-org/acl-pub/blob/gh-pages/002.book.md

  • Make sure the manuscripts are legible and there are no formatting problems with the papers and they adhere to your formatting guidelines.

  • Make sure you get the signed copyright release form for all papers. In softconf this just involves checking that they included their electronic signature when submitting the camera-ready. It's ok for only the first author to sign, but preferable if all authors sign (according to Priscilla).

  • Make sure at least one author of each paper has registered for the conference before submitting the proceedings.

  • Write the preface to be included in the proceedings. This should thank everyone who contributed (faculty advisor, reviewers, panelists, etc.) and describe the reviewing process and the number of submitted and accepted papers, etc.

  • Resolve binding issue: will the student papers be bound with main conference papers? In a separate workshop proceedings? Some other way? The current ACL policy (as of 10/2000) is to bind them separately.

  • If you're using panelists, send papers to the panelists, or make them available on a website.

  • (The rest of the bulletpoints of this section likely are not relevant if you use the above softconf instructions, but may be relevant otherwise.)

  • Make a table of contents and whatever indexes you want (index by author last name, for example) to send to the printer (they don't make them for you). Leave page numbers off.

  • Send final versions of papers to printer. The printer can fix things like ink that is a little too light or papers printed on A4-size paper.



Pre-conference preparation

  • Write a summary to present at the business meeting, find out what meetings you need to attend at the conference.

  • If applicable, set up the social program, find out if you have money for the social program, try to get money contributed/allocated for social program.

  • Send instructions to authors on preparing their presentations or posters (can use softconf MailTool).

  • If applicable, send instructions to panelists (i.e. what is their commitment, what do you want them to do).

  • If applicable, assign session chairs to introduce and time paper presentations. Make yourself some 'you have 5 minutes', etc. signs. It's important to show a sign at the halfway point.

  • Double-check with the local arrangements people that your room is reserved

  • Organize volunteers or recruit someone else to manage that (typically volunteers get registration waived).

  • Make feedback forms to take to the session.

  • Decide how you will recruit reviewers for the next year, either by posting a sign-up sheet, circulating sign-up sheets at the sessions, etc.


Running the session

  • Plan a way to publicize the event during the conference. We put flyers on the registration table.

  • Post a sheet on the notice board for people to sign up to be reviewers for the next year.

  • Distribute/collect feedback forms.

  • Take a headcount or make a list of audience members that attend the event. This is necessary for future planning, to determine whether the format was a success, and to include in post-mortem reports.

  • Make sure you have a pointer, blank slides, markers, and water for the speakers. Also, make sure you have an extra bulb for the overhead projector (it may be inside the projector). Make sure you know who to contact for support if you have technical difficulties.

  • Plan social activities, find a place to publicize social events Also, plan a way for students to contact each other for informal social activities. Maybe put a sign-up sheet on the notice board.

  • Make time for the students to have practice talks? The 1999 ACL student co-chairs did this.

  • Present summary at the general business meeting.

  • Hold a student business meeting. Probably hold it on the next-to-last day so people have a chance to find out about it. Use it to discuss student involvement in ACL in general, and issues relating to student research at the conference, how they liked this year's format, etc.

  • Recruit new co-chairs. Talk to people during breaks and at the student meeting, take down names to give to the Exec.


Post-conference Fallout: Do ASAP after the event

  • Help students if they need help submitting expenses and getting reimbursements (if you had a grant).

  • Make a summary to send to granting agency. Wait until all reimbursements have been sent out.

  • Write up your lessons learned and summarize your feedback from the session. Do this immediately after the conference before you forget!

  • Propose co-chairs for next year to Exec. They should be from different geographic areas if possible and different areas of CL.

  • Hand over documents relating to the session and spreadsheets on the submissions, acceptances and recommendations for funding to the ACL secretary.