Link roundup for August 2023

I wouldn’t have pegged the British Association of Drama Therapists as being into conference posters, but here we are.

Unfortunately, I do think this presentation’s suggestions lead to some very common traps: emphasis on trying to put in a complete story over something that will start conversations.

• • • • •

Speaking of videos, here is another from Mattias Rillig:

I mostly agree with the advice here, apart from the recommendation to make all the text into bullet points. Short text? Yes. But save bullets for true lists. See my most recent blog post on bullet points.

• • • • •

Inkscape has two advantages for making conference posters.

1. Vector graphics.
2. Free.

Here’s a tutorial that shows the basics of Inkspace in the context of creating a poster. 

Same author delves into writing abstracts for conferences.

I reviewed an earlier version of Inkscape some time ago. I found it super fiddly at the time, but that was many updates ago.

• • • • •

Font seller Monotype delves into the subtleties of legibility of type. Excerpt:

 What makes a font legible, and how do I find one?

Like many design questions, the recipe for this answer is a blend of art and science and is governed less by rules than by best practices.

• • • • •

Lisa Muth explores the power of gray in graphs over at the Datawrapper blog.  Excerpt:•

(G)ray is special and should be treated as such.

Very good article that summarizes a lot of what I have been thinking about in graphing lately. Recommended.

• • • • •

A couple of general papers on communication

A paper on science communication in The Journal of Experimental Biology  and a preprint on communicating to the general public in PsyArXiv.

• • • • •

How to convince your boss that a conference isn’t a vacation in disguise. Excerpt:

I understood the need to justify the expense. ... But (the CEO) implied that suspicion should be the default. These sneaky scientists. Always pretending they need to go to conferences for intellectual betterment, when they’re really just looking for the next sweet lanyard to hang from the corner of their bookshelf.

Hat tip to Hiroshi Ishkawa.

 

• • • • •

That’s the roundup! Thanks for joining me!

“Nobody will stop you”: The viral machine learning poster

You’ve probably already seen this poster:

“You can just put up a poster at ICML (International Conference on Machine Learning - ZF) and nobody will stop you.” 

Hail Eris!

Daniel Severo’s tweet about this poster has a couple of million views, and was also tweeted out by Nature’s Twitter account.

What is not easy to see in a typical online photo size it what the paper to the right of the poster says. It’s a gmail address. So of course I had to see if it worked.

And you know what, it did!

I asked for the story behind the poster:

The story’s basically what you might guess: a few sarcastic students at an international conference get to thinking that their field could use a little healthy lampooning and execute a quick turnaround on a satirical poster. The jokes are all ribbing on tropes of ML research.

There are two lessons from this poster.

The first is that there are no poster police. While conference organizers often direct presenters to make posters a particular way on their instructions, or tell people not to do things in the poster session (like the “No photos!” rule at Neuroscience meetings), the reality on the ground is, shall we say, more laissez-faire.

The second is that the unexpected is a strong predictor something going viral online. Just doing something different gives you an edge for attention.

What poster would you make if you weren’t worried about someone else telling you, “No”? (Common sense and good taste should still apply!)

Related posts

Let anarchy reign!