24 February 2022

Link roundup for February 2022

Cartoon of poster session.

 Bruce Kirchoff, author of Presenting Science Concisely, has a nice blog post on strong scientific posters. You know, the ones that can bench press twice their weight. He discusses two misconceptions:

  1. People read posters like papers.
  2. A title just introduces the work.

John Wagner’s illustrations from the book (above) are also lovely!

• • • • •

Very helpful design post by Lisa Muth on how to cut down on too many colours. Nine of her ten suggestions can apply to posters:

  1. Simply don’t show different colors
  2. Show shades, not hues
  3. Emphasize
  4. Label directly
  5. Merge categories
  6. Group categories, but keep showing them
  7. Change the chart type
  8. “Small multiply” it
  9. Add other indicators

 I particularly think points 2, 4, and 8 are underused.

• • • • •

The Cognitive Style of PowerPoint
PowerPoint probably remains the main tool that people use to make posters. This Twitter thread by C. Thi Nguyen style shows that arguments by Edward Tufte against PowerPoint are new and powerful insights for many.

(J)ust had a peak teaching experience teaching Tufte's classic mega-rant, The Cognitive Style of Powerpoint, in my Tech & Design Ethics class. It went so good and I kinda want to teach it in my intro ethics class now.

Tufte says that PPT tends to get us to squeeze the really important details into small-font indented bullet points at the bottom, which de-emphasizes their importance. And I believe this from observation, but I was like: Why does it push that way? The students convinced me that it was really important that you didn't want the large-font lines in PowerPoint to wrap, and so you ended up writing very simple short things into the large-font headers, and saving the details for the tiny-font sub-points.

The default structure of PowerPoint templates asks you to constantly summarize. Bullet point structure asks you to have quick summaries of details. Every slide asks you to have a title-summary of all the bullet points. This mimics the management structure. Engineers have to give the “bullet-point” summary to middle-managers. Middle managers have to give the very compressed “action-summary” to upper administration. And PowerPoint is constantly looking over your shoulder, asking you to provide those summaries, and super-summaries. It wants your information already formatted into the right shape to be sent up the chain. Pre-digested.

Another way to put it: a lot of science, and other forms of reasoning, the details matter the most. In paragraph-narrative form, you can dwell on the details without summarizing until you choose to. But PowerPoint is constantly asking for the action summary of each slide. PowerPoint embeds the voice of management, constantly at your side, constantly demanding "the upshot, and make it snappy." (T)his also had the ultra-peak moment of us all wondering why people tended towards bullet points on PowerPoint, my firing up PowerPoint onscreen, picking the first template, and finding bullet-points as the default, and the whole class screaming “OH SHIT IT’S THE DEFAAAULT!”

(Lightly edited.)

Far from being a neutral container for information, PowerPoint (and other slideware) affects that information in specific ways.

03 February 2022

Online conferences offer many benefits, but online poster sessions still suck

Since 2020, most academic conferences have been online. A recent paper on conferences found many positive effects from online conferences. In short:

  • More people overall.
  • People from more places.
  • More women.

The paper argues that online conferences also reduced carbon footprint, but do this by calculating carbon emissions of air travel only. They do not calculate the carbon emissions from hosted server space, energy use of uploads and downloads, and so on. As blockchain technology has proved, “digital” does not mean “carbon neutral.”

The paper highlights the problems with online poster sessions in the abstract: “(F)urther development of virtual networking features and poster sessions is necessary to achieve widespread adoption and acceptance of this new format.”

A dive into the main text shows a more complex picture.

Skiles and company go into more details in this excerpt, lightly edited (acronyms removed):

Analysed virtual conferencess had poster authors publish their posters via Twitter, using a web-based iPoster sharing platform, or by uploading a 5-minute prerecorded presentation to the conference website. The poster presentations had high view counts (NAMS iPosters had on average 142 views) but presenters could not tell how many attendees were viewing their posters and features for communicating with poster viewers were not effective. In contrast, Twitter-based poster sessions are increasing in frequency and allow asynchronous communication. However, Twitter is not available in every country, limiting access. Consequently, virtual posters were less popular, with 85% of North American Membrane Society survey respondents and 43% of Photonics Online Meetups 2 survey respondents indicating that they preferred in-person poster sessions to virtual poster sessions.

So they present a sample size of two, with an even split on preference. One crowd preferred face-to-face poster sessions, the other preferred online poster sessions. The “preferred” margin for online poster sessions is smaller in one meeting than the other, but still. I’m amazed that one conference had more than half of respondents favouring online poster sessions.


Skiles M, Yang E, Reshef O, Muñoz DR, Cintron D, Lind ML, Rush A, Calleja PP, Nerenberg R, Armani A, M. Faust K, Kumar M. 2021. Conference demographics and footprint changed by virtual platforms. Nature Sustainability. https://doi.org/10.1038/s41893-021-00823-2