It is frankly ridiculous how much time is spent manually marking up LaTeX to show how a paper was revised in the CHI R&R process. Sure, some of this is about clear communication with reviewers. On the other hand, we know how to create automatic diffs of PDFs: why not just use this? #chi2026

VIS 2025 wrapped! Our OPC team introduced two experiments: student reviewers (20% uptake, successful mentorship model) and public peer reviews (16 papers, 52 reviews on OSF). Both continue next year. Thanks to our IPC and the incredible Petra Specht. #ieeevis 🎢 ieeevis.org/year/2025/bl...

Ever wondered why your advisor keeps suggesting the same ideas? I'm a deterministic state machine with no memory—and I suspect many professors are too! I just published a simple suggestion for Ph.D. students to manage their advisor's memory 🧠 niklaselmqvist.medium.com/the-state-ma...

We’re very pleased to present DashSpace, a web-first live and collaborative platform for immersive analytics using WebXR: github.com/Webstrates/D... Here is @maski89.bsky.social ready to give the talk at #ieeevis 2025 and @pwsbutcher.bsky.social on deck for a demo. PDF: pure.au.dk/ws/portalfil...

Here we have Johannes Ellemose giving his #ieeevis 2025 talk on “Eye of the Beholder 🧿: Towards Measuring Visualization Complexity”. Is there something intrinsic about a visualization that can be used to characterize its complexity? Johannes connected several studies to investigate this.

Sungbok Shin is now presenting Visualizationary (needs some practice to say; think “revolutionary”) as an automated LLM feedback tool for visualization design. The tool gives feedback for novices to improve their visualizations over iterative design processes. #ieeevis PDF: arxiv.org/abs/2409.13109

I just enlisted the help of Arnold from Austria 🇦🇹 to bring home my controversial message that immersive analytics has a Unity problem at the “DataVis in XR” panel. Our CG&A paper: pure.bangor.ac.uk/ws/portalfil... Thx @pwsbutcher.bsky.social, @ritsosp.bsky.social, and @maski89.bsky.social for help!

Wow—Sungbok Shin points out that our work on Drillboards was started in 2020 (deepest COVID). Drillboards is a dynamic dashboard with authoring and reader modes where you can create chart hierarchies for varying detail. #ieeevis ArXiv: arxiv.org/abs/2410.12744 Demo (laptop): drillboards.pages.dev

Important question in accessible visualization. Our interviews (doi.ieeecomputersociety.org/10.1109/TVCG...) tell us that basing an accessible representation on the visualization is good for (1) working with sighted people, and (2) for people not born BLV. Dylan Cashman and Areen Khalaila #ieeevis

It was our great honor to serve as #ieeevis 2025 overall papers chairs. The conference is now open and we can’t wait for you all to enjoy the program. Thanks to Petra Specht for her tireless work as assistant to the OPCs! (And good luck to Melanie, Alex Endert, and @tisenberg.bsky.social for 2026!)

Johanna Schmidt from VRVis and TU Wien opening #ieeevis 2025 in Vienna! More than 1,100 attendees on location.

Here’s @zcliu.bsky.social presenting “contextual dynamic explanations” (CoDEx) at the #VISxGenAI workshop at #ieeevis 2025. Go @hcil-umd.bsky.social and UMD! visxgenai.github.io

Brand new Ph.D. student Anton Wolter from my group at @csaudk.bsky.social presenting his team’s work on multi-agent data visualization and narrative generation at the #VISxGenAI workshop at #ieeevis; a runner-up in the VisAgent min-challenge at the workshop. Congratulations Anton and team!

@pwsbutcher.bsky.social and @ritsosp.bsky.social tag-teaming immersive data analysis at the DashSpace tutorial during the first day of #ieeevis 2025!

Even in Vienna it seems nobody is willing to take credit for what the Danes call “wienerbrød”. Circular reference ftw. #ieeevis

The #ieeevis 2025 OPCs, working under the direction of the VIS Steering Committee, has just released 52 anonymized peer reviews for 16 accepted papers to be published at VIS 2025. We hope each year will add to this repository. OSF link here: osf.io/s9j5b/ (download the spreadsheet directly)

New favorite use of LLMs: asking it to explain my own slides from last year back to me because I was too stupid/lazy to write proper speaker notes at the time.

The duality of man (peer review edition): - System paper: “where is the user study?” - Evaluation paper: “where is the system description?” 😑

Thinking of your academic paper as less your magnum opus and more as a line of dialogue in a greater conversation can take the pressure off scientific publishing. Here are my most recent thoughts on "Science as a Conversation": niklaselmqvist.medium.com/science-as-a...

Sorry to say it, but if you're a Ph.D. student, chances are that your code will not live long past your degree. Instead, you should strive for separating your research contribution from their implementation (which counts as validation). Read more: niklaselmqvist.medium.com/your-code-wi...

Only my second UIST paper ever, but please check it out! I can’t go to South Korea (too much travel this fall already), but @clemens.klokmose.net and @maski89.bsky.social will be there to show the flag!

In an act of true courage, @clemens.klokmose.net opens the JavaScript study lecture in our @csaudk.bsky.social HCI class by challenging the students to “ask me anything” about web development, and then proceeds to live code solutions to all contenders. Pictured after white-knuckling drag and drop.

Professor Corrie is on deck for the faculty meeting at @csaudk.bsky.social!

The way we prompt AI has been bothering me for a long time. Accidentally hitting Enter sends off prompts—but surely they should be carefully crafted, not fired off haphazardly? I wrote about why this frustrates me so much and what we should do instead. niklaselmqvist.medium.com/to-prompt-is...

This bowling alley has just rickrolled its entire clientele. Gotta respect the hustle.

After seeing one too many #ieeevis papers that were clearly resubmits with little or no revision from the previous iteration, we VIS 2025 OPCs just wrote this blog article (ingeniously named if you ask me): ieeevis.org/year/2025/bl...

"This is not vis." I've heard this criticism throughout my entire career. But when does upholding #ieeevis standards cross the line into harmful gatekeeping that stifles innovation? In my new blog post, I argue that our field's vitality depends on embracing new ideas, from accessibility to AI.

Since I’m finding myself in this curious position once more (my fifth time as papers co-chair for #ieeevis), I thought I would post my 2017 blog post on not trying to appealing paper reject decisions (it’s unlikely to work/peer review/not scalable): niklaselmqvist.medium.com/an-appeal-no...

Friendly PSA: When either happily rejoicing or angrily complaining about an accept or reject decision from #ieeevis, remember to remove your friendly neighborhood OPCs from the CC. We don't mind to see these emails (especially the happy ones), but it invariably causes embarrassment after. 😅

New w/ Vaishali Dhanoa, Anton Wolter, @gabymlp.bsky.social, Hans-Jörg Schulz on patterns for "agentic visualization" - AI agents aiding vis w/o losing human agency: - Roles (what agents do) - Communication (how agents share) - Coordination (how agents collaborate) Paper: arxiv.org/abs/2505.19101

It baffles me how some #ieeevis authors cynically resubmit their rejected papers to the conference without making the most basic of revisions. The level of disrespect towards prior reviewers is staggering. No, you can’t just pass off your own failures on “Reviewer 2”. Own them, and revise your work.

VIS and CHI both have two review rounds, but there is a significant difference: VIS summary reviews are supposed to list required changes for acceptance, whereas CHI meta reviews explicitly should NOT include them. #ieeevis PC members should keep this in mind: please DO list required changes. 1/

20 years after "Illuminating the Path" established visual analytics, @clemens.klokmose.net and I wrote "Automating the Path: An R&D Agenda for Human-Centered AI and Visualization" (soon @ieeecga.bsky.social). Read for how HCAI can benefit vis (and vice versa)! arXiv: arxiv.org/abs/2504.07529

Datamancer is a wearable system that lets you control multiple screens with hand gestures. Point to select displays, then use bimanual gestures to manipulate data. No more fumbling with mice—just the wave of a hand. To appear at #chi2025 in Yokohama! niklaselmqvist.medium.com/wave-of-hand... 1/2

I'm hiring 1-2 Ph.D. students for human-centered AI in visualization. Research: interpretability, creativity, inclusive access. Come design intelligent systems for complex data interaction w/ me. Backgrounds in HCI, vis, or AI welcome. Design/engineering skills a plus. phd.nat.au.dk/for-applican...

Puzzling change observed from OPCing for VIS 2024 vs. 2025: this year we are getting an onslaught of new people emailing us to volunteer for reviewing, signing up on the review system, and marking themselves as expert in all areas. Volunteering is commendable, but this is fishy...

Friendly PSA for #ieeevis 2025 authors: Yes, PCS is currently closed so you can't upload your supplemental material---don't panic! We're reconfiguring the submission form to allow for supplemental material to be submitted by April 7. Please bear with us!

The fact that the midnight AoE on March 31 deadline for #ieeevis actually means April 1 here is a serious mental health concern for my poor students and a source of infinite joy for me. Sometimes the Venn diagram of dad jokes and advisor jokes is a circle.

"The man in black fled across the desert and the gunslinger followed." One day I will I find a way to integrate this line into a paper, and that will be a great day.

Thrilled to have Dan Russell here at @csaudk.bsky.social giving his AUCS colloquium on “People, AI, and Online Research”!

Thanks to #ieeevis folks who have accepted our invitation to join the VIS 2025 PC. Unfortunately, out of 265 invitations, 62 have not yet responded (and 48 have declined; 23.6%). We need at least 200+ accepts. If you're still considering, please accept! (Going back to the well means new approvals.)

Congratulations to all of the awardees! It was an honor and a pleasure overseeing the selection process.

Very cool work! This is exactly the kind of research that Sungbok Shin and I hoped someone would do when we proposed our own very limited chart reading model in 2022.

Given the way things are going in the HCI/visualization field, I think it's time I change my lastname to "eLLMqvist." Yay or nay?

Today’s novel HCI term collected from a student mini-essay: “locust of control.” I, for one, welcome our future insectine overlords! 🦗🦗🦗

Pro tip: When giving an invited talk that you will give again in the future, record yourself so that you can transcribe your narrative and clean it up with an LLM. There's nothing quite like trying to remember the words that went with your pretty slides three months after the talk. 😑

I have decided that “congratulations, fellow Fellow” is my favorite genre of email… Big congrats to @jheer.org and Claudio Silva as the two other #ieeevis authors who were elevated to ACM Fellow this year! www.acm.org/media-center...

DashSpace, our platform for web-based collaborative immersive & ubiquitous analytics, is on GitHub: github.com/Webstrates/D... Paper w/ Marcel Borowski, @pwsbutcher.bsky.social, Janus Bager Kristensen, Jonas Oxenbøll Petersen, @ritsosp.bsky.social, @clemens.klokmose.net + me just accepted to TVCG!

I've increasingly found myself talking about the current wave of LLMs in HCI and vis research as a "gold rush", so I thought I should write something about it. I've been guilty of the "thin veneer over GPT" papers myself, but perhaps it's time we go beyond? niklaselmqvist.medium.com/all-that-gli...

‘It’s like replacing a precise surgical instrument with a Swiss Army knife. Yes, the knife is more versatile, but would you want your surgeon using one?’ Lol, yeah. The inferiority of causal language models in most domains is becoming very, very clear.