22 March 2018

Critique: The Capricorn Experiment, plus: Font families

Today’s poster is about the Capricorn Experiment, not to be confused with the 1970s conspiracy movie, Capricorn One:

The only conspiracy in the new poster, from Vidhi Bharti at Monash University in Melbourne, is the justification for “Capricorn”. It’s supposed to be an abbreviation for, “Clouds, Aerosols, Precipitation, Radiation, and Atmospheric Composition Over the Southern Ocean.” The experiment should really be “Capracoso.” I mean, you just don’t get to make abbreviations out of any letter somewhere in the word! It would be like abbreviating the University of Texas Rio Grande Valley as “EXONE”.

But I digress. Let’s look at the poster, which you can click to enlarge!

Vidhi wrote:

I work on boundary layer meteorology, which basically deals with a lot of mathematical equations and unattractive diagrams. Therefore, presenting it all in an attractive package is a big challenge.

Vidhi does a good job of rising to that challenge with this poster.

I like the way this poster tackled the two column layout. While I normally would prefer the two columns to be even in width, when there are only two, having the two columns differ in width is perhaps not so annoying as when there are three or more columns.

Everything could use more space around it. I would try shrinking a lot of elements, maybe 85-90%, to give each bit a little more breathing room. The main text of the poster is so readable that it can afford to be a little bit smaller, so that the overall appearance isn’t so crowded.

The poster could also use a stronger sense of visual hierarchy. In particular, the author and institution names are bigger than the text below them. This causes two problems.

  1. The bylines chew up space that this poster needs to reclaim.
  2. The size indicates those names are more important than what the poster is about. With respect to the team, who I have no reason to think are anything other than fab human beings and scientists, the person reading the poster is probably more interested in the text of the poster than who wrote it.

The bottom of the “Analysis” box is driving me a bit crazy, because the bottom margin is obviously thinner that the top or left. The text on the right of that box occasionally strays a little close to the edge.

I like the fonts, but I noticed there were two of them: one in the main text, and another in the lower left box. I asked Vidhi if there was any particular reason to switch to a different font in the “Capricorn Experiment” box. She replied:

I derived the layout inspiration from magazine articles where they usually divide the sections into different columns and keep one highlighted box. For my poster, I wanted “Capricorn experiment” to be that highlighted box.

I’ve used callout boxes myself (see the 2012 Neuroethology poster here), and I applaud the idea. The execution might be improved with some different font choices. The two fonts are, to my eye, too different from each other, and they clash a little bit. What Vidhi needed was a font family: a set of complementary fonts deliberately meant to work together. Usually, they are all designed by the same person or foundry.

Here’s an example of a font family in action on an infographic I made for CBC’s Quirks and Quarks radio show (the version below was tweaked slightly after I submitted to the show):

There are at four or five different typefaces on that image. But they are all part of the same font family, Adorn. Adorn is an excellent example of a font family. In the case of Adorn, none of the edges are perfectly smooth. Every typeface, whether slab or Roman or script, has a hand-drawn feel, like it was created with ink and paper.

For Windows computers, Arial is a font family that many would recognize. It comes in a narrow, black, and rounded fonts.Not as different from each other as Adorn, but the idea is the same.

MyFonts, Fontshop, and other online font stores usually have bundles, families, or similar things for sale. Sometimes the fonts are simply different weights or styles of the same font (bold, italic, book weight, thin, etc.; see FF Tisa for an example). Sometimes, particularly with display fonts, the mixes of styles is more dramatic and ambitious, like Adorn or Phonema.

While I do think imposing limitations on yourself can be useful, the system fonts are that come on most computers are sometimes too restrictive. Times New Roman was a default font for a long time, and many people still use it for much text. Yet it still seems limited on many computers to a single font. There aren’t light or heavier weights.

While there is an initial cost outlay to buying a font family, just a couple of font families in your toolkit opens a lot possibilities in design. 

15 March 2018

Critique: Solid state hydrogen

Today’s poster comes to us courtesy of Mi Tian. Click to enlarge!

The individual blocks (like “Background” and “Research goals”) are good. I like the colour choices and the “pins” by the headings as graphic elements.

The arrangement of the blocks on the page is not as good. The reading order is confusing. The little lines to the pins, plus the height on the page (i.e., closest to title), suggest I’m supposed to start with “Research goals”. But normal reading order would suggest I start with “Background.” I’d try flipping “Summary” and “Acknowledgements”, which would place those two blocks in positions that are more typical of where those are usually placed.

The poster feels very crowded. Tons of elements are almost touching each other.

  1. The “Summary” heading is almost touching the edge of the blue box its in.
  2. The pin by “Introduction” is almost touching the graph above.
  3. All the logos down in the corner are almost touching each other.
  4. The “Applications” heading pokes up higher than the text in the section above it (“>86 kg/m3”), messing with the clear division of sections.

Everything below the title bar would benefit from being shrunk a bit -- maybe 95-90%, at a guess -- to make more space between the elements.

In the “Applications” section, it’s not clear why “Polymer” and “Composite” are capitalized, when nothing else is at that text level. Similarly, if “goals” (in “Research goals”) is not capitalized, “Solid” in “Investigation of Solid H2” shouldn’t be, either.

The red and blue in the title image might be worth tweaking. Red touching blue can cause chromostereopsis, which a lot of people find distracting. It’s not bad, because the blue is dark, but still.

08 March 2018

#RSCposter 2018

The hashtag #RSCposter is short for, “Royal Society of Chemistry poster,” and it blew up on science Twitter this week. This was a seriously organized event, with rules as comprehensive as I’ve seen for some in person conferences.

Organizer Edward Randviir explains (lightly edited):

The goal of this is to provide a new innovative conferencing format that takes advantage of modern social media... We also wanted to gives presenters a free platform to present and discuss their work, and encourage particularly young researchers to participate in academic discourse to build their confidence. Twitter was the most appropriate social media platform. Many professionals across a range of sectors use Twitter for professional purposes, unlike Facebook or other social media outlets. Twitter limits the discussion to 280 characters, which challenges participants to be concise while communicating key messages from their work.

This was the fourth time the Society had done this, but it was the first time I’d noticed. Edward explained that the first two years (2015, 2016) had about 80 people contributing (using the hashtag #RSCAnalyticalPoster). It expanded in 2017 to none areas of chemistry, and participation jumped to about 220 posters. “Following on from that success,” Edward continued, “we brought in chemical engineering this year. With help from several Royal Society of Chemistry journals, we have seen participation increase again by around 12%. We hope to grow the event further in the future.”

Tweeting posters presents its own particular problems. Twitter is a mobile phone app at heart (as much as Twitter tries to make it the “everything machine”), and mobile phones are small screens, not big poster boards. I was viewing posters on big desktop computer. Even with a fairly high resolution computer screen, I worried about whether people would dump posters meant to be printed 2 meters across into a tweet and that it would be too small to see.

Lucie Nurdin noticed one workaround:

Opening the poster into a new tab allows to zoom on it and have a high resolution image. Glad I figured that out!

To my surprise, most posters were readable. But alas, not all were. This poster by Jinchuan Yang, fell into the trap of not making the text big enough for a Tweet. Click to enlarge (or any subsequent poster).

Progyata Chakma mostly did okay on the right and middle columns, but some of the left hand text is too small to read.

This, from GKalqurashi, is another example of a poster that wasn’t readable on my desktop.

Most posters were readable on my desktop, although some were often barely so.

Another problem with tweeting a poster is that when you post an image on Twitter, it creates a preview image that is resized and cropped down. It used to be 440 × 220 pixels (a 2:1 aspect ratio) in landscape format (wider than tall). I’m not sure that’s still true, because I saw a lot of square preview images. And many people use clients other than Twitter.

Regardless, most posters I saw were not optimized for preview images. I saw lots of posters in portrait format (taller than wide), which no app I know uses for Twitter previews.

Because of the cropped previews, the poster’s title – the most important part of a poster – were often hidden. This problem was mitigated a little, because the tweet itself could serving the job the title usually does: to entice the passerby. (Or scrollerby, in this case).

Luke Wilkinson’s poster caught my eye by placing a cute robot right in the middle, where it will be seen despite how Twitter crops rectangular images. Placing it in a circle also helps break up the rectangle monotony that you get when faces with scrolling through lots of posters.

Yuanning Feng took advantage of the format to make an animated poster. This does not look as good here on the blog as the original tweet, because of the hoops I have to jump through to convert a *.gif posted to Twitter – which Twitter converts to a movie – back into a *.gif.

Feng’s animation seems to be getting him about three times as many “likes” as most posters.

But as of now, it seems one of the most popular posters was by Jo-Han Ng. (And once you visit that, check Errant Science’s riff on Ng’s poster!)

As I scrolled through #RSCposter, my overall impression was, “Oh, there are all the problems that I usually see on academic posters. Too much to read. Too many boxes, not enough white space. Photo backgrounds that make the main stuff hard to read. Colour overlead.”

“New bottles for old wine,” as the saying goes.

External links

Take part in a truly global scientific conference
RSC Twitter Poster Conference 2018

01 March 2018

Critique: RNA capping

Today’s contribution comes from Melvin Noé González. It was presented at an RNA meeting at Cold Spring Harbor Laboratories. Click to enlarge!

He writes:

Through the years I experimented with various templates for poster presentation, and I’m proud to say I’m really happy with how this one turned out. As you will find, I used a piece of advice you mentioned in one of your posts regarding a short summary section — and people loved it! I was approached by several people just because they thought the layout was cool, even though I wasn’t related to their research.

I’m always glad to have feedback that advice works!

The title bar works well, by presenting everything cleanly. The logo is sensibly over to one side, and blends into the background. The authors names are prominent, with institution and contact information legible, but low key.

This poster is well organized, which helps walk you though what is maybe a little too much material. The numbers by each heading ensure you don’t get lost.

Some of the layout would benefit from a little more tweaking. The spacing between the boxes is inconsistent. The margin above the “Graphical summary” are wider than the margins between the “Background” boxes and the data boxes on the right.

There’s one place where this poster goes off the rails. Fortunately, it’s down in the fine print section, in the acknowledgements and references. While I appreciate how beautiful that three-dimensional molecular structure is, and how much it adds visually to the poster, it does terrible things to the text around it.

It’s tearing that text apart.

When we read, we expect related text to be close together. When I look at the “Acknowledgements,” I see two blocks of text that I want to read separately.

But how you are supposed to read the acknowledgements is far more complicated. What I thought was the first sentence of the first text block is the third fragment of the entire acknowledgements section.

Just when I think I have gotten used to the lines broken into two pieces, the second to last line gets split into three pieces.

The same thing happens in the references, with a DOI number danging far from the “doi:” text identifying it.

Wrapping text around an object can look graceful and elegant. But you cannot just “set and forget” a setting in your layout software. You have to be willing to go in and adjust things by hand to avoid these kinds of problems.

Nine is fine!

Happy blogiversary to me!

It is a little bit crazy for me to think that this blog has been running for nine years straight. And still going (reasonably) strong!

It is mostly thanks to my readers and contributors – which is to say, you. I appreciate your attention, and hope this resource continues to help you.

Thank you for stopping by.

Picture from here.

22 February 2018

Lab posters are not conference posters

When I wander through department hallways and professor offices, I often see posters like this, from Rottner and colleagues (2017; tweeted by journal here). Click to enlarge!

These sort of posters often feature cellular processes or biochemical pathways. They are often professionally done, attractive, and valuable teaching tools. But they are not good examples that conference poster makers should be trying to imitate.

A poster like this is meant for experts, so presumes a high level of knowledge. It is intended to be something you can look at for days, weeks, months, sometimes even years. They can show lots of fiddly little details that you can discover over that long period of time.

In a conference poster session, you have a few minutes for someone to absorb the work, not months. You can’t stuff in the same level of detail in conference poster that you can in a lab poster.

Hat tip to Prachee Avasthi.


Rottner K, Faix J, Bogdan S, Linder S, Kerkhoff E. 2017.

Link roundup for February 2018

Neuroskeptic asks whether conferences are hostile environments.

I have never been the target of a harsh question at a conference but one of my colleagues was, a couple of years ago.