Kamis, 06 November 2014

Latest modern science | Critique: The data flow dragonfly - Si Bejo Science

Today�s poster is from Marianna Rapaport, and is shown with her permission. Click to enlarge!


Marianna writes:

The poster presents my masters thesis, the general area is programming language research.The only illustrations in my thesis are graphs and math formulas. I wanted to add some graphics to the poster that would attract people (and not scare them away with math). My dad told me that my graph looked like the wings of a dragonfly, so that's where that big insect comes from.

I would also like to acknowledge the invaluable help of my friend Erica Dufour. She helped me to arrange the text boxes and came up with the idea of the gray background that helps the reader understand the order in which to read the material. She also helped me understand how to use Adobe Illustrator and InDesign.

Finally, I used the same font as in this poster on which you also have a critique on your blog

I like this a lot. It�s clean, and has a strong visual impact. The dragonfly is a nice design touch. The use of the contrast colours orange and teal to highlight is consistent, and subtle enough not to be garish or overwhelming.

The one thing I question is the reading order. The "Result" box is not where I would expect it. Based on headings, I would go:

  1. Summary
  2. Intro
  3. Goal
  4. Problem
  5. Method
  6. Result

But based on its position on the poster, �Result� would slot in at position number 4, not 6. Marianna replied:

I agree that the reading order is still unclear. But I don�t even know what could be done about that without changing the whole poster layout.

Regarding the �Result� box, I read somewhere (maybe even on your blog) that it�s a good idea to put the results right in the beginning; in a way, it�s a replacement for the abstract. I thought that in my case, the results are in the beginning and at the same time in the end. But maybe that doesn�t make sense because it�s impossible to understand the results without reading everything else, so you�re right there, too.

Having the results up top, as here, is not horrible. The approach I might have taken would be to think of that top row as the �take home� messages, and the second row as being �for the aficionado.� The trick then becomes distinguishing the two.

The gray band on the second row signals this a little, but it might have been stronger if there was a second cue to signal that the second row was less important. For example, a slightly smaller point size for the text might have helped.

Alternately, perhaps using different way to highlight the �Results� box, instead of the same gray as the row below, would have broken the connection between them, and emphasized that �Result� was meant to stand on its own, as a conclusion. 


Still, the overall effect is quite lovely.

Kamis, 30 Oktober 2014

Latest modern science | Link roundup for October 2014 - Si Bejo Science

I don�t think I�d seen this resource on Giving Poster Presentations before. It�s part of a larger online resource on �English communication for scientists.� I think I�d remember if I�s seen this Jorge Cham gem from the front page before:


Elizabeth �Inkfish� Preston covers a paper that examines how a simple graph significantly increases the persuasiveness of an argument. And when I say �simple,� I mean very simple:


Another primer on how to get the most out of a conference from Mandi Stewart, which wins points for citing We Bought a Zoo:


My partner and I talk about having �five seconds of professional courage� when networking at conferences. Conferences are a great time to meet people, and unless you put yourself out there and introduce yourself, you could miss out on some great conversations. I love the movie �We Bought A Zoo� which is where having five seconds of professional courage came from. �You know, sometimes all you need is 20 seconds of insane courage. Just literally 20 seconds of just embarrassing bravery. And I promise you, something great will come of it.� Try it. Five seconds of professional courage.

This article on the importance of comics has some analysis of reading flow after my own heart. Hat tip to Siobhan O�Dwyer.



You too can learn the difference between a soft crop, a split crop, and a stickout crop in this post at the different ways you can crop an image by John McWade.

I also like McWade�s short reflection on how design can make life better:

Design is about more than whether something �works.� Lots of things �work.� A theater marquee with chipped paint and missing letters �works.� If the local strip mall has what I need, you could say its ugly plastic sign �works.� Each identifies my destination well enough to get there.

What they don�t provide is delight, inspiration, fulfillment.

Wired has a lovely profile on book cover designer Peter Mendelsund.  Book covers have some goals that are similar to conference posters: attracting passers-by, for instance.

On one level, dust jackets are billboards. They�re meant to lure in potential readers. For a certain contingent of the publishing industry, this means playing it safe. �The path of least resistance when you�re designing a jacket is to give that particular demographic exactly what they want,� Mendelsund explains. �It�s a mystery novel, so you just splatter it in blood, and put the shadowy trench coat guy on it, and use the right typography.� Familiarity, the thinking goes, will always sell something.
Mendelsund does not subscribe to this view. He�s said that he prefers an ugly cover to a cliche one(.)

One of Mendelsund�s better known projects is The Girl With The Dragon Tattoo. Here are some rejected ideas:


I have not seen the movie Idiocracy, but this post on making fake corporate logos is interesting just the same. Hat tip to Alex Jones and Amanda Krauss.

The Current radio show on CBC has been running a series called, �By Design.� It�s going to be running all this season. This series is not about graphic design, but is a wide ranging exploration of how we make things.

I�m months behind in bringing you this blog post on redesigning maps for the modern age.

If you�re finally ready to learn how to use a higher end graphics package than PowerPoint, try Vector Tutorials for Adobe Illustrator. Hat tip to Anthony Salvagno for this resource.

And today in type crimes, or �Someone did not read their directions closely enough�:



From here.

Kamis, 23 Oktober 2014

Latest modern science | Stretching out your title - Si Bejo Science

People are used to tinkering with the vertical spacing of text; having to make a manuscript double spaced, for instance. But they are not as familiar with how to make text look good by adjusting the horizontal spacing.

John McWade reminds us of a useful tip about the spacing of type:


Text meant to be read at a distance � like the title of your poster � should be expanded a little!

Since most people are making posters in PowerPoint (despite my constant pestering for you to stop doing that), Let me tell you a couple of ways to do this in PowerPoint.

Select your text, right click it, select �Font,� and pick the �Character spacing� tab. That allows you quite precise control over the spacing:


There is also a �horizontal spacing� button in the �Font� ribbon. The drop down options for that one, however, are more general: �Loose� and �Very loose.�


Here�s a sample of how text looks expanded. �Loose� is a little more than 2 point spacing.


�Loose� might not be a bad setting to try for titles, and maybe headings, on posters.

After you�re done here, practice your horizontal letter placement skills by playing this kerning game.

External links

The unexpected typestyle of Ikea

Kamis, 16 Oktober 2014

Latest modern science | Critique: Astrophysics code - Si Bejo Science

Today�s poster is contributed by Alice Allen, and is used with her permission. Click to enlarge:



She wrote:


Under 100 words on this poster... or so I will claim since the screenshots are there to illustrate the points! (Not counting the authors' names, I think the count is 89 words.) This poster is for an online resource that people at the conference are familiar with, and is to inform people of recent changes to the website.


This poster wins points for simplicity. It can be read with a few glances, which is a definite win for any conference poster.

I�m always curious to see what improvements people make on their own. After she sent her first email but before I replied, Alice sent along another iteration.


I think the changes you made for the second version are good ones. Making the title more prominent, and getting rid of the outline around the �Over 900 codes!� were both good moves.

I can see why the screen grabs were rotated as a design element. I�d be tempted to tinker with the amount of rotation. I might try bringing the central screen grabs a little closer to horizontal (though not normal straight up and down).

Tiny little typesetting detail. In the right box, the field �See also� is in quotes, but the links, �Previous� and �Next� are not in quotes. There might be a case for putting �Previous� and �Next� in quotes, too, for consistently. I�m not sure what a style guide would recommend; perhaps there is some subtle stylistic difference between a field and a link.

Kamis, 09 Oktober 2014

Latest modern science | Critique: Affective feedback - Si Bejo Science

Today�s poster comes from Mary Ellen Foster, and is used with her permission. Click to enlarge...


I like the main body of the poster a lot. It�s clean, big, uses lots of graphics, and is well-organized. The one thing I would try would be to crop the middle photo, rather than having other pictures overlapping on top of it.

While I appreciate that there is very little text, this may have been pared down just a little too far. I can�t tell two important things:

  1. What�s the question?
  2. What�s the answer?

As a browser, I often want a take home message.

This isn�t helped by the weak title, which represents most of your communication effort. �Studying the effect of� in a title is bland and uninformative. Every academic thing is �studying the effect of� something. A question would be better, and an answer would be better still.

I�m always sort of surprised that people still try to incorporate institutional logos on their posters as often as they do, given how often they cause problems. This poster is a great example: every logo here weakens the poster.

The logo on the left causes problems because it is too close to text, and it messes up alignment of the authors with the title. That it�s a big dark block makes it draw too much attention away from the title and the authors. The logos on the right just look thrown together and messy.

Hiding among the logos is a QR code. This has a few problems, too. The QR code is high on the poster, which might make it inconvenient to scan, depending on how the poster is mounted. But more importantly: why should I scan it? What does the QR code lead to? It�s always good practice to tell people what they will get!

Related posts

Detective stories: �Whodunnit?� versus �How�s he gonna prove it?�
The epic logo post
Your title is 90% of your poster
Take me home tonight

External links

Kamis, 02 Oktober 2014

Latest modern science | Critique: Hard problems - Si Bejo Science

Today�s poster come from Ciaran McCreesh and is shown with permission. Click to enlarge!


My first reaction was, �Nice work!� I like the colours and the clear organization.

Personally, I would try removing the boxes around each section, maybe creating boxes around each column (so there are no horizontal bars).

I�d also want to fiddle with the lower right box to make the bottom edge align with the other two boxes. This poster does such a nice job of keeping things clean and aligned that little things like that stand out!

I like using bold to emphasize key points, but I wonder if there might be a little too much bold. The less you have, the more punch each instance has. It�s diluting some of the impact.

The very top box is a nice attempt to introduce the problem, with sort of a sub-headline. But it doesn�t have any other clear signals to its importance apart from its position. It comes across as a small sliver of text, and your eye hops over it to the first box in the top left. It might benefit by being made larger, or using something else to distinguish its place in the poster�s information hierarchy.

Looking at this from a distance, it feels off kilter, because of the asymmetries in layout. There are uneven columns, and the logo on the right is also breaking the symmetry. I suspect that the title is truly symmetrical when measured with a ruler, but it looks like it�s too far to the left. The normal expectation is that the title will line up with the central column, which is pushed right because the right column is narrow. I suggested making the title aligned to the left (perhaps enlarging a little), and putting the authors and institution on one line below that. Then it could be roughly the same height as the institutional logo.

Here is Ciaran�s revised version:


He wrote:

I followed your suggestions, except for removing the horizontal bars: I couldn't get that to look right. I ended up coming second place in the vote at CP 2014 (http://cp2014.a4cp.org/), which was a pleasant surprise.

Hooray!

Kamis, 25 September 2014

Latest modern science | Link roundup for September 2014 - Si Bejo Science

Andrew Maynard included this as an example of a �more radical� poster design in his post, Creating Poster Presentations that Tell Stories:


Andrew goes on to write:

To me, a poster presentation for me is an aid for story telling � to be used by an in-person narrator. The reality though is that sometimes the poster needs to be able to at least hint at that story without your in-person input.  This creates something of a design-conflict.

He also says:

Coward that I am I should note that the posters aren�t great, but hopefully illustrate process

Over at Southern Fried Science, Chris Parsons has penned Mr Darcy�s Guide to Conference Etiquette � Part 1, which includes thoughts on posters:

Posters give one a unique ability to talk directly to conference goers, often while they are well flown on a glass or two of wine, in a depth one cannot achieve with the audience at an oral presentation. A single good, well-designed poster is also very memorable, much more so than dozens of slides in an oral presentation.

This also prompted this Twitter exchange about poster sessions.

Keith Bradnam is a person after my own heart, doing his bit to improve conference posters. He has a nice post called The problem with posters at academic conferences. And the problem, according to Keith?

The problem here is not with the total amount of text � though that can sometimes be an issue � but with the width of the text.

A new paper in PLOS Computational Biology by Rougier and colleagues offers ten guidelines for better figures, which can be an important component of better posters. I would put their rule #5 much higher on the list...


  1. Know your audience
  2. Identify your message
  3. Adapt the figure to the support medium
  4. Captions are not optional
  5. Do not trust the defaults
  6. Use color effectively
  7. Do not mislead the reader
  8. Avoid �chartjunk�
  9. Message trumps beauty
  10. Get the right tool

While this poster leaves something to be desired graphically (too much stuff), I enjoy the title. Hat tip to Nick Loman and Mike the Mad Biologist.