Paper Chair
Main role: Be responsible for the scientific program during the conference. This includes:
Direct the paper reviewing and selection process
Make the presentation/poster/demo program
Run the paper presentations (oral and poster) during the conference
Prepare the proceedings archive
The Paper Chair will work closely with the Paper Archive Officer.
Before the conference
Prepare Call for participation (September)
The CFP is usually adjusted slightly from year to year, but not be drastically different. Please refer to previous calls.
Set up the Conference Management System (October)
NIME used Easyconf as the Conference Management System (CMS) for many years. After some bad experiences with server crashes, lack of support, and lack of multi-track functionality, we decided to move away. From 2012 NIME has been using Precision Conference as CMS. Chairs have reported this to be a solid system, although with some of a learning curve. It is also quite costly.
In 2019 the SC has decided to switch to Microsoft Conference Management Toolkit. This CMS is used by ISMIR, WAC (and many others). The other option was EasyChair, which is used by SMC, MOCO (and many others). The reason for going for the Microsoft solution is the cost (free versus a £1000 cost for EasyChair) and a more modern GUI.
Recruit reviewers (December)
Master list of NIME reviewers: The list of reviewers are passed on from one year to another. You should add new people as you see fit, and the previous chair(s) should also remove reviewers that did not work well. It is common to add some local people each year, and some of these should probably be removed if they are not regular NIMErs.
Gender parity (and also diversity): It should be addressed at all parts of the organization, and also at all times. It is important to encourage the community to reflect on the importance of these matters in order to deepen the awareness that simply reducing it to numbers (e.g. striving to include as many women as men) is worthless if not accompanied by the acknowledgment of the importance of each individual's contribution. It is also important to promote actions that strive to create equal opportunities to all participants by compensating unforeseen biases.
Be careful about accepting to take on reviewers that start out saying they have little time, and can do the reviews a bit late. If people first start being late, they may be vary late, or not deliver at all. It is a very tight timeline for the review process, so we cannot really have much delay.
Be very strict about deadlines, and plan well ahead. The reviewer pool should be ready way before the deadline, preferably before Christmas. It is better to get too many than too few reviewers, so that you can spread the task on more people. Maximum number of papers should be 5, but 2-3 papers per person is better.
Prepare reviewers and meta-reviewers about what is coming when. Send out e-mails a week before every deadline. It is also important to send out information in advance to tell when papers are being distributed. That way people can plan time to do their review. This will save you time in the end.
Peer review (February)
This is the most critical phase for the paper chair(s), since time is limited and the pressure is high. It helps the process a lot if reviewers have been recruited early on. Please also be very precise in your communication, with clear deadlines and expectancy for everyone involved.
The paper chair(s) assign papers to reviewers (3-4 reviewers per paper to ensure that all papers are reviewed by at least 3 people) and follow up the deadlines
A group of meta-reviewers write summaries for each paper based on the individual reviews, and suggest a decision
The paper chair(s) do the final selection
Distribution of reviews
Centralized distribution: In the interest of running the process as smoothly as possible, the paper chair(s) should do the review assignment and follow-up. This is how it has always been (we only started with meta-reviewers in 2014). The CMS helps out a lot by coming up with suggestions of reviewers based on expertise. Having a centralized reviewer assignment also helps in avoiding individuals to be contacted by multiple meta-reviewers.
Uneven distribution: It is important to avoid an uneven distribution of reviews (some reviewers with many reviews, and some reviewers with none). You should aim for a better distribution of papers to the community of reviewers by distributing at least one paper for each reviewer. This also involves the community more broadly, and helps get new and young people on board early in their careers.
Ask reviewers to complete their profile: To ensure better fit with reviewers, you should promote adding sample papers to the CMS profile, and also open for bidding for papers.
Tentative reviewers: We had a number of reviewers who did not reply / were tentative. With the aim of getting 3 reviews per submission and helping the meta-reviewers, we send out more invitations, and once we achieved the quota, we deleted the tentative reviewers. It turned that some of the reviewers were having an independent conversation with the meta-reviewer, and perhaps agreed using another medium (e.g. email). We recommend that it is OK to leave tentative reviewers hanging in the system even after inviting new reviewers. This may have a double advantage: on the one hand we can't really know in advance if tentative reviewers will eventually respond nor if accepted reviewers will in fact comply with their promises, so the worst-case scenario for having 5 or 6 external reviewers with some "pending" is to end up with a few extra reviews for a few papers; on the other hand, by leaving them hanging in the system, by the end of the reviewing period we would have a clear picture of the "unreliable reviewers", maybe to serve as a heads-up for the next chairs (sort of a blacklist).
Removing reviewers: For each submission, the system is unclear with reviewers who are removed for any particular reason (reassignment due to lack of anonymisation, unresponsiveness, busy, etc). We recommend to have some kind of history and be able to see for each submission 1) who has been already invited to be a reviewer and removed (and the reason why, e.g. lack of anonymity, unresponsiveness, busy), and 2) who has declined the invitation already. This would not apply if we keep the tentative reviewers as suggested in the previous point.
Meta-reviewers
NIME has been using meta-reviewers since 2014. This has helped to improve the general quality of reviews. Meta-reviewers have typically been responsible for 5-15 papers each. Obviously, the quality of the work goes up with fewer papers, but having more meta-reviewers also makes the whole process more vulnerable if they do not deliver. It is important to set up a reliable group of meta-reviewers, and they should ideally know the conference fairly well.
Meta-reviewers read the individual reviews and write up a meta-review for each paper. The meta-reviewer should base his/her argument on the individual reviews, but could/should also read the paper in question to make up an opinion. This is particularly important when the individual reviewers disagree.
The meta-reviewers should be encouraged to make a clear suggestion as for whether to accept or not (since many individual reviewers tend to gravitate towards borderline). Decisions on acceptance/rejection should always be based on a qualitative judgment. The scores are useful to help in sorting papers, but should not be used as the basis for the final decision.
For many papers, making a decision is quite straight forward. For some papers there may be a lot of disagreements between individual reviewers. Then the meta-reviewer could ask the individual reviewers about their arguments before making their recommendation.
The final decision of acceptance/rejection is made by the paper chair(s), and if they do not reach an agreement, it is ultimately the general chair that will have to make a decision.
Double-blindedness
This is how double-blind policy is implemented in the PCS System used at NIME:
authors do not know external reviewer nor meta-reviewer names
external reviewers do not know author names
meta-reviewers know both author and external reviewer names
external reviewers do not know other external reviewer names
external reviewers know meta-reviewer names
The only strictly double-blind part is that between external reviewers and authors.
A few number of poorly anonymized submissions that were detected were "rescued" by asking the authors to submit a corrected manuscript and redistribute it to new external reviewers, which was a hard task (meta-reviewers could be kept because they are able to see the author names of the submissions). We recommend to check if papers are properly anonymized before distributing them among reviewers.
Anonymization: The templates are anonymized with seemingly real names. It is not the first time that this creates confusion among reviewers and even meta-reviewers. Often times, the anonymization is so strict that it is very difficult to get the full picture. It is very important to take the anonymization policy seriously otherwise some authors will be better positioned than others. We recommend offering solutions / options to the authors so that they can still upload videos, datasets, links, etc to improved the reviewing experience.
Conflicts of interest
It is very important to take conflicts of interest seriously, and to follow the NIME statements published online. Conflicts of interests are self-reported, but there are cases where reviewers have not reported possible conflicts. So it is important that the chair(s) do an extra check of this. You should remove any reviews that appear to be based on conflicts of interest.
Selection of papers
The acceptance of papers should be based on the quality of the papers. We do not use a specific percentage for acceptance, it should solely be based on the quality reported by reviewers.
While there is usually a good correlation between the review scores and quality, there may be cases, particularly with disagreements between reviewers, where also papers with borderline or even negative scores may be accepted. These, and other borderline cases, are usually discussed between meta-reviewers and chair(s).
Programming
Programming of papers should be based on a combination of the quality of the paper and the chairs' idea about how well suited the paper is for the different presentation formats.
Relatively few papers can be accepted for oral presentation, typically 20-30 dependent on the artistic program. This means that it is very competitive to get an oral presentation slot. Please keep in mind that the program is curated, and there should not be a one-to-one relationship between reviewer scores and oral presentation. It is more important to think about creating balanced and relevant sessions, and to ensure a diverse oral program. This diversity should include a good balance between different fields, artistic vs scientific content, institutional/national distribution, gender diversity, etc.
If the chair(s) think it works best, it may be worth "upgrading" a poster submission to an oral presentation. This should be checked with the authors first. Some people deliberately submit for a poster for various reasons, and should not be forced to present orally if they do not want to. Other times people are shy and don't dare to submit for oral presentation, and in these cases they may be honored by being "upgraded".
It is more common to "downgrade" an oral presentation to poster, or a poster to demo. It should be communicated clearly that at NIME, both posters and demos are of high value, and often also papers that receive very high review scores are being presented as poster. It is the responsibility of the chair(s) to decide whether a "downgrade" should also be combined with a reduction in the number of pages of the paper. This is a question about the content of the paper, and whether the reviewers think the paper has enough content to fill, say, 6 pages. If not, then the authors should be asked to reduce it to 4 pages.
Confirmation letter to send
After paper reviewing there are different kinds of letter to send:
submitted and accepted as paper
submitted as paper but downgraded to poster
submitted as poster and accepted as such
submitted as concert but performed as club
submitted as concert but downgraded to demo
Prepare proceedings
It is very important to thoroughly check that all camera-ready papers fulfill the requirements of the template.
Test print the proceedings
We are not printing the NIME proceedings for the conferences any longer. Please remember that many people still print individual papers, so it is advisable to test print all papers to check if any errors occur.
During the conference
Presentation duration
Previously there was a difference in presentation time between short and long papers. That has changed, and all presentations are now equally long.
There have been different duration of the talks, but they generally follow this pattern:
10-15 min presentation + 3 min questions + 2 min turnover
Remember to inform presenters about the duration of talks every time you have the chance, since people tend to forget.
Running paper sessions
Time is short, so it is important to have a structured approach to chairing the oral sessions. The paper chair needs to find session chairs to run each session. It is common to use experienced participants as session chairs, but please also consider giving new people a chance (and remember the diversity also of session chairs).
It is important that the paper chair clearly informs the session chairs about their duties:
Arrive early to the session, and welcome all presenters.
Ask presenters to test their setup before their session.
Start on time, every minute matters. Remember to close the doors.
Keep track of the time, and signal to presenters when there are 5, 2, 0 minutes left.
Ask next presenters to set up during the Q&A session.
Please also ensure that you have volunteers available to assist during the session. Ideally you will have one volunteer helping with the audiovisual setup, and at least one volunteer to run with microphones during Q&A sessions.
Remember that session chairs do a very important job during the conference! An important part of the job is to smile, breath, and help nervous presenters to relax. But it is equally important to keep the time, nobody likes to be delayed to a coffee break.
No-shows
We do not accept remote presentation, video presentation, or presentations by non-authors at NIME conferences. The reason for this is that the conference should not only be a place for publishing papers (then a journal is more appropriate), but be a meeting point of people. That is not possible if the authors are not present. No-shows happen, for various reasons, but it should be discouraged. The most efficient way of doing this, is to remove no-show papers from the proceedings.
This rule was created after seeing a number of cases in which people would submit their work, get it accepted, and not show up to present it at the conference (usually last minute or not even without notice). This is unfortunate for the participants, because it creates "holes" in the program, and it prevents other papers from being accepted for presentation. In addition it undermines the idea of the conference as a meeting point and a venue for active discussion.
We do realize that this rule may seem unnecessarily strict, but we also see that it would be very difficult to come up with another rule that does not have a number of unintended side-effects.
For people that have been accepted for one conference but that could not make it, we have allowed for the paper to be presented at the next conference without a new submission/reviewing process. Please do consider this option.
After the conference
Getting the proceedings on NIME.org
The nime.org archive is built on BibTeX files stored at Github. As paper chair it is your job to prepare a complete BibTeX file adhering to the NIME standards. The steps involved in this is as follows.
Remove papers that were not presented
There are always some accepted papers that are not presented for various reasons. No matter the reason, papers that are not presented at the conference should be removed from the proceedings. This is because the proceedings should reflect what actually happened at the conference. It is also very important to let all (future) authors know that NIME has a strict no-show procedure.
Fix errors in the metadata or papers
Due to the short time in preparing the proceedings before the conference, there may be occasional errors in metadata (author names, affiliations, etc.) or in the PDF files (most often wrong page layout, missing fonts, low-resolution images, etc.).
It is in NIME's interest to have as good metadata and PDFs as possible, so we do accept such error corrections and minor PDF modifications.
Export from PrecisionConf
When the basic "post-processing" of the proceedings is done, it is time to generate the BibTeX file. As of 2018 it should be possible to export BibTeX directly from Precision Conference. Prior to that we had to do some post-processing with the script Precision2BibTeX.
Massaging the BibTeX file
The exported BibTeX file need some "massaging" to match the other BibTeX files in the repository. Please check the files on Github to see how the file shold be formatted. Typically this includes:
Check the title of the proceedings. It should be the generic title "Proceedings of the international conference on new interfaces for musical expression". The year and place should be included in separate fields, not as part of the title. This is because the ISSN for the series is bound to the generic proceedings title.
Check that abstracts are included for all papers. Please also check that there are no line-breaks in the abstracts.
Check weird formatting (typically unicode-related issues). Can be done automatically using for example JabRef or bibclean
Insert ISSN number (and possibly ISBN) in each post (query-replace usually works fine)
Insert right link to PDFs. Should be something like nime20XX_ID.pdf.
Bibtex with Pubpub
Here is a script that scrapes Bibtex information from NIME's Pubpub community: scrapeBibtexFromPubpub.py. Instructions can be found at the top of the file. Note that the author names collected this way is based on what the author supplied on pubpub, and may differ from registration data (e.g. on CMT).
Here is a script that converts the scraped Bibtex into NIME-compliant formats: writeBibtexToBibFile.py. Carefully check the output for errors.
If you are using your own scripts, remember to replace line breaks with whitespaces in the "abstract" field of Bibtex. Beware of unusual line breaks such as LS (Unicode character 2028) and PS (Unicode character 2029).
Send the material to the paper proceedings officer
The NIME paper proceedings officer will help with the above, and will also do the final upload to nime.org.
Last updated