Senin, 01 Juni 2015

Kamis, 28 Mei 2015

Latest modern science | Link roundup for May 2015 - Si Bejo Science

I love this deep meditation on pixel-based art from video games. Even though it�s slightly off-topic for the blog, this is my �must read� of the month. It�s revelatory to read someone who know details of animation and art show the pros and cons of using pixels. The comparison of two takes on the Street Fighter character Chun-Li is wonderful:


When they see SFIII or KOFXIII, they don�t see the unbelievable craft that went into it, or if they do, they have to first reconcile what they see first, which is the magnified image above. They have to pay the pixel tax.

Here�s the rub, and a lesson that applies to conference posters:

Nobody owes us their time or attention. As such, when someone gives us their time, an implicit agreement is made and we are now in debt to that person. We owe it to them to deliver value for their time, and to deliver it efficiently. ... Speak in a language people can understand so that they can actually see what makes your work great without a tax.

Hat tip to Jeff Alexander.

A fairly good one sheet from Elsevier. Hat tip to Mike Taylor:


StressMarq Biosciences has a twelve point guide to making a poster. I agree with about 10 of those points. Their template is too busy, bullet points are rarely better than short paragraphs, and I don�t know why they recommend the PNG format for pictures. It�s still a pixel-based image; vector images are always better.

What can scientists learn from designers? Quite a bit:

Scientists need to remember that they are deliberately designing a product for an end-user.  This focus on an audience may seem obvious, but it is surprising how many scientists forget about their target audience.  We say �it is time to submit the paper to the journal,� or �we need to make a poster for the conference,� and we often forget that we are really creating products for people.

Jonathan Owen wants to help you decide when to use quotation marks. Hat tip to Mike Taylor.



�Once you understand the design of flags... you can understand the design of almost anything.� This nice TED talk makes a reasonably convincing case for that thesis, and there are lots of lessons for poster design, too. Hat tip to B. Haas. This gives me an excuse to show this flag, because it�s well designed, celebrating its fiftieth anniversary, and is my nation�s flag:


Compare to the flag of Milwaukee:


Far too many posters look like the Milwaukee flag.

Laura Bergalls talks about what a walk in the wood taught her about getting attention.


You may have heard that making something hard to read makes it more likely you will understand it. One fancy way of saying this is �cognitive disfluency.� Turns out... not to be the case. Hat tip to Emily Willingham , Janet Stemwedel, and Aatish Bhatia.

Mad Max: Fury Road and Captain America serve as reminders: some things are visual media. John Wick explains (original emphasis):

Watching (Mad Max: Fury Road) made me think of that meme going around with Captain America lecturing Spider-Man. It�s nearly three pages long and it�s just Cap quoting from a book. Quoting from a book.


Not only did this bore me to tears, but it also stunk like a burned out writer looking to fill page count. Now look, I�m a huge Alan Moore fan, so I�m used to verbosity in comics, but Moore understands that comics is a visual medium. This kind of exposition doesn�t belong in a visual art like comics or films. Moore gets that. So does George Miller. Everything in this movie communicates in such a powerful way that dialogue is almost unnecessary. Cap is a man of action, not a man of lecture.

Conference posters are also a visual medium.

I often use other people�s images in my posters. A new source of of high resolution public domain images can be found at the State Library Victoria. Many of these are old vintage black and white photos, which can give them a lot of visual interest. This pic of Wendy the Wombat is proof. This post is better because it has a picture of a wombat. Hat tip to my mate Ely Wallis.


Designing a new typeface is a challenge, and Japanese particularly so. Here�s a peek into a Japanese type foundry, where they are still designing each character by hand. (Original article, with images but paywalled text, is here.) Hat tip to Garr Reynolds and John Meada. http://t.co/QIiNl2N4Cm


Devony Looser on the joys of academic conferences. It includes tips:

Do not write or revise your paper or poster at the conference. I�ve seen junior and senior colleagues make this tactical error all the time. You must have your paper finished before you come to the conference... You do not earn any points with anyone by saying, �I can�t go because I have to go to my room and finish my paper.�

On the flip side, we have Christy Wampole, who is tired of conferences.

Academic conferences are a habit from the past, embraced by the administrativersity as a way to showcase knowledge and to increase productivity in the form of published conference proceedings. We have been complicit.

But... counterpoint! David Perry replies to Wampole and argues we should save conferences:

Everything I have ever published has direct origins in one or more conferences, a lineage I can trace through my CV, mapping the formal and informal ways that academic gatherings have shaped my work. And I know I�m not alone.

The book How Posters Work by Ellen Lupton is coming out soon (and you better believe I�ll be reviewing it!). Here�s an article about the exhibition the accompanies the book:

�Posters are the only genre of graphic design that is explicitly created to be stuck on a wall,� Lupton told me in an email. �Many people are more comfortable displaying posters in their own homes or work spaces than they are with more formal or serious works of art. Posters are part of everyday life, so they feel approachable and real.�

Here are 27 jokes for graphic designers. Hat tip to Garr Reynolds.

An even better joke for graphic designers is the #DrunkTufte hashtag on Twitter.

In last month�s link roundup, I pointed to an article about the perils of bar graphs. This month, I�m pointing to a round-up of the reaction to it.

Quote of the month from Lindsay Waldrop:

That thing where you said you�d do a poster and then completely forget about it until the day before you leave. o.O

Kamis, 21 Mei 2015

Latest modern science | Critique: The final four - Si Bejo Science

It�s long past March, but poster design knows no season! This week�s poster comes from author Cameron Fuqua, and is used with permission. Click to enlarge!


In a reversal of the norm, the tables � usually one of the things I like least on a poster � are one of the best things on this one. The tables avoid the �data prison� problem: that is, too many lines encasing each cell. The colour banding to distinguish rows is subtle. The gradient fills provide a little emphasis for each cell, but are not distracting. The silver gradient fills for the text boxes are also well done, providing some visual interest that is not distracting.

I wish that aesthetic extended to the rest of the poster.

The rest of the poster is confined within thick, heavy-lined boxes. The poster would probably be significantly improved by thinning or removing the lines entirely.

Worse, they�re boxes with rounded corners. The rounded corners brings the lines of the box closer to the text, which is already a problem. There is no space between the text and the lines, particularly in the upper left boxes. Further, the rounding isn�t consistent. Some corners are quite curved, others are closer to right angles.


I like distinctive typefaces for titles and section headings, but this one (something in the Eurostile family, I think) sacrifices too much legibility for decoration. From a distance, common letters like �a� and �e� are hard to tell apart. There is some variation in the heading weight: some things are in bold (which is contributing to making the letters hard to read), some not.

Underlining is used for emphasis, which also make it more difficult to read the text.

Throughout the poster, there are dozens of cases with things being poorly aligned or placed. Some of the mathematical equations have a space surplus on one side or another.

Finally, predicting the outcome of a competition is something that many people should be able to understand and relate to, regardless of the complexities of the mathematical equations behind the predictions. I�m surprised that when I look down in the lower right, where I expect to see an answer to the question, �Can I use this method to get a better bracket?�, I can�t see any answer. I would love a single sentence like, �This new model�s performance is better / as good as / worse than previous ones.�

Kamis, 14 Mei 2015

Latest modern science | Bullets versus sentences - Si Bejo Science

Some other resources on poster design recommend that people use bullet lists extensively for their posters. I advise against it, most of the time.

The pros of bullet lists is that by their nature, people tend to write less text. That concision is very useful on a poster, I admit.

But I want to argue there are more negatives to using bullets that positives.

First, my experience with looking at PowerPoint slides is that people are inconsistent in how they type bullet lists. For example, people often punctuate some bullet points with a period, but leave others without a period. When people write sentences in paragraphs, they will put a period at the end of every sentence.

Second, the size and spacing of bullet points is often badly done in software. Even PowerPoint, the culprit that made bullets ubiquitous, doesn�t scale well when you move outside of the standard slide sizes. Here�s a quick mock-up for a four foot wide poster with a bulleted list (click to enlarge):



Under this default scheme, the bullets are too far from the text. The spacing between lines and points is also a little dodgy. Microsoft Publisher, which I use a lot for posters, handles bullets even more poorly.

Third, bullets destroy narrative. Edward Tufte has made a thorough analysis (from The Cognitive Style of PowerPoint, excerpt quoted here):

Lists can communicate three logical relationships: sequence (first to last in time); priority (least to most important or vice versa); or simple membership in a set (these items relate to one another in some way, but the nature of that relationship remains unstated). And a list can show only one of those relationships at a time.

Bullet lists may be more concise, but they are impoverished compared to sentences in paragraphs. Sentences can express many more relationships.

Fourth, readers are trained to read sentences in paragraphs. It is the most common thing we read, and is how we expect to absorb complicated ideas.

This is not to say that bulleted lists are useless. They are completely appropriate for short lists. A poster, though, should be more than just short lists. For example, I feel okay about using a bulleted list for a quick summary of my case again bullet points:

  • Bullets are used inconsistently
  • Bullets are poorly typeset
  • Bullets show relationships poorly
  • Readers are used to sentences

I don�t think I would convince anyone of my argument if that list were all I posted.

External links

The Zen of Presentations, Part 41: Consistency

Photo by David Stillman on Flickr; used under a Creative Commons license.

Selasa, 12 Mei 2015

Latest modern science | Register now for #SciFund poster making class! - Si Bejo Science

The poster design class I mentioned last week is now ready for you to sign up! Click here to register for the class!

The official announcement on the #SciFund blog has a few more nitty gritty details.

This is the first time we�re charging for a #SciFund class (US$50), so I�d like to address is a question I�m sure many people will ask:

�What am I getting for my money?�

Valid question. After all, I�m one of the instructors, and this blog exists to spread ideas I have about poster design around for free. If someone read all the entries in this blog, you would probably have a very good idea of some of the things I�d be talking about in the class.

One benefit of the class is that you�ll be able to have a lot of interaction with the instructors. It isn�t just me leading the class; I�ll be joined by Anthony and Jai. And while a blog is static, working in the class will not be.

Perhaps more importantly, you will get to have a lot of interaction with other students in the class. In a class like this, the opportunity to get the ideas and feedback of others can be invaluable, because different people bring such different ideas about style.

Another thing that you get from being in the class is that we do have plans to offer a certificate of completion for those who make it all the way through. This could be useful in demonstrating that you�ve been engaged in professional development activities.

Once again, you can click here to register for the class! I hope you will join us!

Kamis, 07 Mei 2015

Latest modern science | Announcing the #SciFund poster class! - Si Bejo Science

I�m very excited to announce a new poster making class, sponsored through the #SciFund Challenge!

#SciFund started out as an experiment in science crowdfunding, but has expanded its mission to include science communication and professional development.

In this class, you�ll learn basic design principles, be instructed in how to use Adobe Illustrator (a powerful, vector-based graphics kit), and build your communication skills. And yes, you will make a poster!

Because we want class participants to make something that is useful to them, we ask that you have a research project with data or a research proposal. This might be a project you are presenting at conference this summer, or, if you�re an early career academic, might be a proposal for a thesis or dissertation. We also ask that you have access to Adobe Illustrator.

The class runs five weeks, starting Sunday, 7 June 2015 and running through Saturday, 11 July 2015.

Unlike some online classes, where it�s just you and the computer, this one has lots of meeting time with moderators and other class participants. The main moderators will be Anthony Salvagno and me (Zen Faulkes). We expect participants will put in about 5 hours a week for their assignments. We will also have hangouts (group therapy for poster design) and some group work for review and feedback.

Participants should be generally available between 10:00 am and 10:00 pm Eastern time to be in class hangouts and other events. (Multiple time slots will be available to meet.)

People who successfully complete the course will be given a certificate of completion.

The cost will be $50, and registration will begin soon. The last #SciFund class on video making filled up, so watch this space, follow the #SciFund hashtag on Twitter, and the main #SciFund page for more details.

Update, 12 May 2015: You can now register here! More details here!

Kamis, 30 April 2015

Latest modern science | Link roundup for April 2015 - Si Bejo Science

While this blog is mainly about poster sessions, poster sessions happen in the larger context of academic conferences. I love conferences, and part of the reason I write this blog is so that people can have good experiences at those conference sessions. Those good experiences do not include harassment. That�s why this blog post by Timoth�e Poisot is this month�s must read:

Across all ecologists we surveyed, 37% witnessed harassment, and 24% experienced it, at least once, only taking into account what happens during scientific meetings. This� wow, this is a lot. ...

1 out of 3 people is not an epi-phenomenon.

The post also shows strong support for organizers to be much more active in dealing with harassment. If you�re involved in organizing a conference, there are steps you can take to make them better and more welcoming. Take them.

I�ve often lamented that most posters are designed by amateurs. I have rarely seen a case for using professionals as compelling as this ad, which was a full page in several American newspapers:


It�s an attractive and well designed ad. Except for one thing.

The brain is backwards.

Not being able to get a brain the right way round is not the signal you want to send when you are trying to announce a �new era of discovery in brain research.� There are professionals who do medical illustration stuff for a living. Hire one. (Hat tip to Mo Costandi.)

The bar graph is a standard way of presenting data. A new PLOS Biology paper argues that it�s a bad way regardless of its ubiquity. Hat tip to Gaetan Burgio and Michael Hendricks.


Nevertheless, the humble bar chart is likely to remain a major workhorse for data presentation for a long time. Here is a short list of good tips. Hat tip to Garr Reynolds.


I Want Hue bills itself as a tool for �data scientists.� Its claims:
Distributing colors evenly, in a perceptively coherent space, constrained by user-friendly settings, to generate high quality custom palettes.


Looks interesting. Not sure why the colours jiggle when you make palettes, though. Hat tip to Dean Malmgren and Justin Kiggins.

I�ve always been skeptical when I�ve heard mathematicians and others wax rhapsodic about the �golden ratio.� This article calls it �design�s biggest myth,� and I��m inclined to agree. But maybe that�s just my confirmation bias. Hat tip to Tommy Leung.

Peter Newbury asked:

Conf poster style question: do you use present tense, as in �results are calculated by...� instead of �results were calculated by�?

This isn�t just a conference poster question, but a general scientific writing question. In general, any methods are in past tense, because you�re describing something that already happened. Results are often in present tense, because the effect you�re describing should be generalizable to past, present, and future situations. To put it another way, we write �E is equal to mc squared,� because it�s always true. You might write �E was equal to mc squared� if it was only true once.

Graphic designer Ellen Lupton has a book coming out in June that was an instant pre-order for me: How Posters Work.


Expect a review as soon as it arrives and I devour it, as I surely will. There is an art exhibit to check out if you�re in the New York area.

Haas Unica is an old typeface that has been made new again. It�s the sort of sans serif workhorse that works well in posters. Hat tip to Timoth�e Poisot and Genegeek.


Jarrett Fuller ruminates on his love of all sorts of posters, not just academic ones.

Throughout history, you could group posters into three purposes: to inform, to persuade or encourage, and to commemorate. Sometimes it straddles the lines between each of these, but the poster�s purpose must always involve one of them.

Alex Holcombe wants you to know this.

Each word you put on your poster reduces conference-attendee approaches by 0.2%. People need to know my invented statistic.

Now they know, Alex. Now they know.