CHI Conference Finances and Format Part Two: Shaping the Future of CHI

Published in

ACM SIGCHI

9 min readNov 24, 2023

Purpose of this blog post: In this post, we describe the future of CHI — both in terms of models for a financially self-sustaining CHI and models for a format that will allow for a better attendee experience with time for structured discussion and conferring.

In Conference Finances and Format — Part One: Prior and Current Context, we described the financial and format constraints of CHI. Please read it for details, but the takeaway message is that CHI has grown in full paper submissions much faster than in attendees. Further, no major changes have been made to the format or fee structure in a time in which paper submissions have more than quadrupled and the interest rate has risen by 43% in the USA. Each year, we make small changes by adding additional parallel tracks, reducing the length of talks, and squeezing more out of our budgets in the face of increasing expenses and stagnant revenues. Because the finances and format of CHI are intertwined, we consider these issues together.

The CHI Steering Committee (SC) has formed several working groups to consider how to address the scale, format, and finances of CHI. We have presented the situation regularly to the SIGCHI Executive Committee and have worked closely with the ACM to consider options moving forward.

Future Finances

Continuing to make budget cuts is not a viable plan for the future of the CHI conference. We have to raise revenue and decrease expenses by changing the conference format to make CHI financially self-sustaining in the future. There are some format and budget changes that we do not support (e.g., removing accessibility supports, choosing a singular inexpensive location in the USA); however, here we list some of the format changes that we could combine with a small registration hike to create a balanced budget:

We could remove the exhibitors. While this would change the face of CHI, it would also allow us to keep growing without increasing registration fees, because we would remove the exhibit hall, opening up more large hotels as site options, which can be cheaper than convention centres, and would avoid significant costs associated with exhibitor booth construction and access. However, many students and first-time attendees really value the exhibitors and attend primarily to meet potential employers from companies like Google, Microsoft, and Meta. The exhibit has provided CHI with a ‘wow factor’ that differentiates it from other academic conferences.

We could remove or reduce interactivity. Many CHI attendees state that this is their favourite part of CHI (on the post-conference surveys) and a big differentiator of CHI from other conferences. However, the program is very expensive to run in its present form. Expenses like drayage (paying people to carry boxes), shipping, electrical drops, lighting, and tables add up to a significant expense for the conference and most attendees do not even understand that CHI must pay for these fundamental supports to the interactivity showcase. Removing the track or reducing the track to a smaller form would eliminate these expenses; however, would fundamentally change what has been an integral aspect of CHI for a very long time.

We could charge a submission fee or presentation fee. This would allow us to separate out the costs of presentation and publication, while also keeping fees low. However this may privilege people from the global north or well-funded institutions.

We could move to a new format in which we met physically every other year as an international community and virtually in between. This would reduce the venue costs and allow us to earn revenue every other year, as entirely virtual conferences are significantly lower in expenses. However, it may have an impact on important networking opportunities for early career researchers and PhD students, who may only have the option to present physically at a single CHI conference during their degree. Hosting a virtual CHI by choice may also affect our ability to recruit willing volunteers into major organizational roles. Although some community members could organize regional events, this is not possible in less populated areas without a large HCI community and would privilege people in HCI-dense areas.

We could stop subsidizing student registrations. Currently, student registrations are subsidized by non-student registrations and set at no more than half the cost of a full registration, yet providing all of the conference supports to students is the same cost as for non-students. There is no rule that prevents us from changing this model. However, this may impact the ability of students to attend and grow their HCI networks, particularly if they are not from well-funded institutions or research groups.

We could cut the student volunteer program. The student volunteer program is as expensive for the conference as holding interactivity — we spend a significant amount of money (well over $200,000 per year) to keep the SV program going. However, the program provides students with valuable experience in community service and networking, and allows many to attend with reduced costs. Anecdotally, most of our conference leadership was at one point in the student volunteer program. On the other hand, the vast majority of student attendees cannot access the benefits of the SV program (due to the lottery system), so spreading the benefits more broadly around our student attendees may have more community benefits.

We could rotate CHI around 2–4 known locations. This would save the conference logistic costs and may allow us to negotiate better prices in longer-term contracts. However the choice of these locations would probably reflect past structural biases to the global north and we lose the opportunity to move the conference to new places to aid access for different parts of our community across the world.

We could expend more effort on securing industry sponsors. Some conferences gain revenue through sponsorship from major companies. CHI’s sponsorship revenue has traditionally been very small. However, increasing industry sponsorship would yield space on the exhibit floor and time in the program to sponsors in return. Further, in the post-conference survey, members of our community have expressed concerns with vetting the ethical practices of companies who wish to sponsor CHI, creating additional labour for our volunteers who coordinate sponsorships.

There are many options for the future of the CHI conference to host a valuable and meaningful event without increasing costs, but it will require new ways of thinking about why we confer and what we expect from our annual meeting. The Steering Committee is actively considering all these options in collaboration with past, present, and future CHI organizing committees. In January 2023, the CHI SC had a design session to generate a short list of possible versions of CHI that are financially self-sustaining.

The question for the community is: what aspect of CHI are you willing to change to see CHI become financially self-sustaining?

Future Format

Beyond the financial issues, the format of CHI keeps scaling to accommodate our growth, without any major redesigns. When we have too many papers, we simply add another parallel papers track (current modeling suggests this needs to happen every other year). When we have too many posters, we contract larger exhibit halls, further limiting our site selection options. This has resulted in a bloated, intense program in which attendees struggle to plan a daily schedule that allows them to see what they wish to see (according to the feedback on our post-conference surveys). Further, there is little structured networking time or time for discussion of content or ideas: we have lost the ‘confer’ part of our conference.

Based on the CHI 2030 visioning exercise, the CHI SC formed a working group in 2021 to address the scale and format of CHI. We came up with the following guiding principles for a re-imagining of the CHI conference program:

CHI provides a range of opportunities to participate and engage around content and ideas
- There are diverse ways to attend and engage in meaningful ways (i.e., range of content maturity)
- There are many ways to fund the ticket for attending CHI (i.e., not just through full papers)
- There are ways of engaging people at all stages (i.e., students through senior leaders) and from multiple perspectives (e.g., practitioners, researchers, teachers)
CHI provides structured mechanisms for connecting people
- The post-CHI surveys show that we continually fall short on focusing on structured mechanisms to provide networking and connection opportunities that are desired by attendees and this needs to change
Papers are the key element that needs to be addressed in a CHI re-design, as it’s such a huge number and one of the main constraints on venue space and program time
- Too many parallel talks
- Takes space away from other interactive engagement
- Too much content absorption and not enough time for discussion
- But papers have traditionally been the core of the CHI experience
- Our program is too large to ‘curate’ sessions; we need to rely on tools for session formation

We came up with several solutions to the scale of CHI, but all of them involved not presenting all of the papers as talks. Our favoured solution involved spending half of the day learning about the state of the art in HCI through paper presentations of peer reviewed content. In the other half of the day, we construct the future together through interactive content (e.g., workshops, SIGs, panels, roundtables, provocations, mentoring events, student competition venues, other discussion formats, and posters) — venues that would be juried or curated. This would have the benefits of increasing time available for conferring in structured formats and also increasing the time for discussion around each paper. However, we would necessarily decrease the time in the schedule devoted to paper talks. Examples of how we could reduce the number of talks include:

We could have all papers give lightning talks, followed by parallelized discussion time in a ‘scrum’ format. If we reduced all talks to lightning presentations, each accepted paper would be able to present to a larger audience through less parallelization. Then, attendees could discuss the papers asynchronously with authors in parallelized discussion scrum, allowing for greater depth in discussion. However, this makes it harder for newcomers to engage with content and does require social interaction for content access.

We could present only the top 5%–10% of papers as talks. The remaining full papers would be either presented asynchronously on YouTube or given a poster slot. However, we would need to come up with methods of determining which papers were presented as talks in an already overloaded review process. Further, we need to consider how we differentiate late-breaking work or work-in-progress posters from full paper posters.

We could accept only the top 5%–10% of papers. Other large conferences have very low acceptance rates. However, this might decrease the number of people willing to attend CHI and could affect the review process in negative ways.

We could increase parallelization of the talks. We currently have around 13 parallel tracks of paper talks at any given time. We could double this, essentially halving the audience of each paper session. However, this would make an already packed schedule even more challenging for attendees, and may make room booking impossible for technical program chairs.

We could limit the number of submissions. To return to the submission and acceptance numbers of a more manageable era, we could simply limit the number of papers that an author can submit. However, this would cause issues in large research groups in which a professor advises a number of PhD students. We could also institute a submission fee, to help authors think carefully about their submissions; however, as noted above, this might privilege individuals from well-funded institutions and research groups.

The question for the community is: what are you willing to change to increase the time available for discussion and conferring?

Continue the Conversation…

We invite you to consider this historical context, current situation, and future formats by joining us at one of two sessions in which we will work together to shape the future of CHI. We will listen to your values and concerns, and hear your ideas for the future of our conference. We will ask the hard questions about what you are willing to trade off for the parts of CHI that are most important to you. We will take what we hear in these sessions and propose several future models of CHI to the community at large, for further discussion and implementation as soon as 2026. These decisions will be in line with the CHI SC’s Vision and Values Statement.

Synchronous Sessions on Shaping the Future of CHI

You can join us at one (or both) of two zoom + miro sessions on shaping the future of CHI.

Session 1: Tuesday, December 5, 16:00 UTC
Vancouver: 08:00, New York: 11:00, Paris: 17:00, New Delhi: 21:30, Tokyo: Wed 01:00, Melbourne: Wed 03:00 (https://dateful.com/eventlink/3895171221)

Zoom link is being sent to SIGCHI members via email.

Session 2: Wednesday, Dec 6, 05:00 UTC
Vancouver: Tues 21:00, New York: Tues midnight, Paris: 06:00, New Delhi: 10:30, Tokyo: 14:00, Melbourne: 16:00 (https://dateful.com/eventlink/1334330769)