Author: Cynthia C. S. Liem
Affiliation: TU Delft
Editors: Cynthia C. S. Liem and Jochen Huber
A little over 10 years ago, I chose to pursue a PhD. This meant I chose a professional life in which research publications and their uptake would be seen as major evidence of achievement. For those working in computer science, the major dissemination platforms for such publications are conferences.
Given my dual background in music and computer science, it was logical that my main interests were in topics that connected these both worlds. As a consequence, I hoped to become part of the Music Information Retrieval community. The International Society for Music Information Retrieval (ISMIR) therefore seemed the professional community to target, and the annual ISMIR conference the most logical place to present my work at.
In terms of its education and research, my department at TU Delft had track records and agendas in visual and social multimedia content analysis, but not particularly in music. Considering methodology and philosophy, I did think a lot of the work at the department was compatible with what I tried to do in music. Furthermore, as I still was in training in a selective major at the conservatoire, I was not in a good position to geographically move to any other institute that would have a more established Music Information Retrieval track record. So I inquired whether I could stay in Delft for pursuing my PhD.
The answer was somewhat complicated. There was no funding for a PhD position in Music Information Retrieval, and there were no strategic plans to change that. At the same time, the people who had supervised me as a student (in particular, my thesis supervisor Alan Hanjalic) saw promise in me, and would like to keep working with me. Ultimately, I got a one-year contract in which my main task was to try acquiring funding and international community backing to pursue a Music Information Retrieval PhD in a multimedia group.
At the start of that year, I got to attend my first ISMIR conference, where I presented a paper based on my master’s thesis. In a previous column for the SIGMM records, I already discussed my experiences at that moment: how debuting alone at a conference was intimidating, but how I was lucky that senior members of the community pro-actively took care I got introduced to other attendees. Frans Wiering, the senior member who looked after me in particular at that moment, was general chair of the upcoming ISMIR, which would take place in Utrecht, so in my home country. Frans was quick to invite me to serve as a student volunteer, which was very good news for me. As my year would be filled with grant-writing, I did not yet have a sufficiently stable infrastructure around me to be able to truly do research, so submitting to the next ISMIR was out of reach. But this way, I could still attend the conference, and even would have an excuse to keep mingling with all the attendees, as we as volunteers would be the first people to answer any participant questions regarding logistics.
Getting funding turned out a true challenge. In 2009, digital music consumption was not as large yet as it is today, and many potential data-providing partners were reluctant to collaborate. Of course, it also did not help my cause that I still was a complete nobody. Finally, when working on music, one faces an interesting paradox. On the one hand, many people, regardless of their backgrounds, identify with music, up to the point that they personally deeply care about it. As such, working on music makes for a good conversation starter, in which people are always happy to share their personal experiences. On the other hand, this makes music a commonplace topic, which risks it being shoved aside as ‘less serious’. Even though technically, the problems we are working on are framed in very similar ways as they may be in neighboring domains such as vision (and the research challenges are at least as hard, if not harder, due to subjective human factors being an integral part of the problem), common criticisms we receive are that music is fun but does not save lives, and does not deal with areas of major economical impact, nor easily measurable societal impact. So while we never have any problems legitimizing our work in public outreach, in grant-writing, we always need to justify extra why our work is more than a fun hobby, and sufficiently relevant to justify serious funding.
After several collaboration rejections, and the one proposal I did manage setting up getting rejected despite good review scores, I was very lucky that at the very end of my grant-writing year, I managed securing PhD funding through a Google Doctoral Fellowship (now PhD Fellowship). For this, I needed to get a research mentor, although my Google contacts weren’t so sure who would be appropriate for this role, as they were not aware of anyone working in music in the company at that stage.
Several weeks later, I was volunteering at ISMIR in Utrecht. That was where I found out that Douglas Eck had just moved from academia to industry, to work on music research at Google. And that was how I got my research mentor, with several extremely useful interning experiences at the company as a consequence.
When Emilia Gómez, the 2018-2019 president of the ISMIR society invited me to become general co-chair to ISMIR’s 20th anniversary edition with her, and host the event in Delft, this was my chance to give back. Now I had general chair powers, and as the society was quite open to discussing any innovations, I could try realizing the conference of my dreams.
As described in my previous column, the inclusive spirit of ISMIR has always been quite elaborate, including mentoring programs spearheaded by our Women in MIR movement, an explicit focus on multidisciplinarity over exclusivity, and on being medium-sized but single-track. Since two years, all our accepted papers are presented in a 4-minute presentation and a poster, such that all the works get equal visibility. This year, we chose to not do themed sessions but to randomize the paper order, such that authors on related topics would not be presenting their posters at the same time. As a side-effect, this also would nudge attendees towards learning about everything that got accepted, beyond the topics of their specializations. This is something I have seen the ISMIR community always being enthusiastic about, while I had very different experiences at (more prestigious) larger-sized conferences. In many cases, their larger size led to many parallel tracks with fragmented audiences, while any plenary program elements were so massive that it was hard to engage with anyone you did not happen to know already, or incidentally happened to stand or sit next to.
We made sure we offered more than paper presentations. For the keynotes, we invited speakers from neighboring fields and disciplines, and encouraged them to give some critical perspectives on our field. We engaged with a local school in an outreach program. Before the conference, we held workshops, including the Women in MIR prototyping workshop, so people would already get to know one another; we had a dedicated Newcomer Initiatives chair to make sure no one felt lost, and the socials were set up such that people could really mingle. With many people in music also happening to be active music players, we offered both formal and informal options to jam together, so that week, several cafes in Delft faced more live music than we would normally see.
But while I was preparing for this conference, one of my strongest experiences was that I kept being haunted by these memories of the past: that being able to join this community (and an academic career at all) had been a really close call, that really was catalyzed by me having been able to join the conferences, and having met supportive seniors, while I was still an early-stage student without a full research embedding.
So one of the ISMIR 2019 achievements I am most proud of, was that we extended our financial support programs, enabled by the ISMIR board and sponsorship funds. Beyond the existing grants for student authors and female participants, we added a third ‘community grant’ category, meant for individuals who would like to attend ISMIR, but who had not been in the capacity to actively participate to the conference at this stage. Reading through the motivation letters for this grant made me realize that my experiences not as much of a freak case, and that colleagues have been facing similar challenges.
I am deeply grateful that these grants enabled for us to get more people over to ISMIR. Young professionals in between positions, students in other disciplines seeking to collaborate more closely on music topics; students that have found themselves as sole people in their labs working on music, as the labs faced other strategic priorities; but also, seniors who used to be members of our field, but who had gradually been drifting out, when entering a vicious circle of not getting music projects funded, then having to do more teaching in other topics, and then taking hits on their research output and profile. It was a wonderful experience seeing all of them actively mingling with the community, and hearing how being at ISMIR indeed had been personally impactful for them.
For my student volunteers, I especially targeted local and national students who were not yet at the PhD level, such that they could experience our academic atmosphere. Here as well, I saw the positive impact of the ISMIR spirit; several of these students (of whom I am not even the thesis supervisor…) made friends with international colleagues, and are even trying to collaborate on music information research with them in their free time today.
Hopefully, this story can help inspiring colleagues who are seeking to make their conference cultures more inclusive and impactful. With this, I do want to add a warning that endeavors like this will not come for free, but demand considerable extra work and advocacy. Much of our proposed innovations initially faced pushback in some form, as these were not how things normally were done, and they required financial and human resources that would not be normally accounted for. But I am very grateful that we followed through, and extremely proud of what we achieved in the end. My great thanks go to the ISMIR society, my fellow ISMIR 2019 organizers and our sponsors for their trust and support.
All ISMIR 2019 presentations have been recorded, and are available through this link. The accepted (open access) papers with supplementary material are available via this page. Photos of the socials are available here.
About the Column
The Multidisciplinary Column is edited by Cynthia C. S. Liem and Jochen Huber. Every other edition, we will feature an interview with a researcher performing multidisciplinary work, or a column of our own hand. For this edition, we feature a column by Cynthia C. S. Liem.
Dr. Cynthia C. S. Liem is an Assistant Professor in the Multimedia Computing Group of Delft University of Technology, The Netherlands, and pianist of the Magma Duo. Her research interests consider search and recommendation for music and multimedia, with special interest in making people discover new interests, as well as questions of interpretability and validity. She initiated, co-coordinated and participated in various (inter)national collaborative research projects on the accessibility of content which would not trivially be retrieved, both in the music/cultural heritage world, as well as in social sciences applications, e.g. collaborating with organizational psychologists. Beyond her academic activities, Cynthia gained industrial experience at Bell Labs Netherlands, Philips Research and Google. She was a recipient of the Lucent Global Science and Google Anita Borg Europe Memorial scholarships, the Google European Doctoral Fellowship 2010 in Multimedia, and a finalist of the New Scientist Science Talent Award 2016 for young scientists committed to public outreach. In 2018, she was Researcher-in-Residence at the National Library of The Netherlands, and in 2019, she served as general co-chair of the ISMIR conference.
Dr. Jochen Huber is a Senior User Experience Researcher at Synaptics. Previously, he was an SUTD-MIT postdoctoral fellow in the Fluid Interfaces Group at MIT Media Lab and the Augmented Human Lab at Singapore University of Technology and Design. He holds a Ph.D. in Computer Science and degrees in both Mathematics (Dipl.-Math.) and Computer Science (Dipl.-Inform.), all from Technische Universität Darmstadt, Germany. Jochen’s work is situated at the intersection of Human-Computer Interaction and Human Augmentation. He designs, implements and studies novel input technology in the areas of mobile, tangible & non-visual interaction, automotive UX and assistive augmentation. He has co-authored over 60 academic publications and regularly serves as program committee member in premier HCI and multimedia conferences. He was program co-chair of ACM TVX 2016 and Augmented Human 2015 and chaired tracks of ACM Multimedia, ACM Creativity and Cognition and ACM International Conference on Interface Surfaces and Spaces, as well as numerous workshops at ACM CHI and IUI. Further information can be found on his personal homepage: http://jochenhuber.com