30 June 2011

Clean solutions to a visual mess

Clean Up Your Mess is a short, very helpful website about visual design by Daniel Higgenbotham. It talks about things close to my heart at this blog, like alignment, boxism, and hierarchies.

Photo by Wonderdawg777 on Flickr; used under a Creative Commons license.

23 June 2011

Printing posters

One reason that people like giving a talk more than a poster is that talk doesn’t have to be printed. While almost everyone has the computing power to make and display a PowerPoint talk, few have what they need to print a full-sized conference poster. Your options:

Own a large format plotter printer.

It’s not within reach of the typical conference goer to own their own personal big printer. This price list starts at $1,400 and ends at, “More than $4,700.” I’m too scared to see how much more.

Such a printer is not out of reach as a piece of shared equipment, though. Our department shares a plotter printer with another departments. The ink, paper, and so forth is paid for by various student fees and department budget. Students and faculty alike use it.

Our university library has a second plotter poster, which is smaller than our departmental printer. They also have a policy limiting people to one free poster, of a certain size (48 inches, I think) per semester.

If you are in an institution that does not have one, try to convince people that this is a useful resource.

Hire someone to print your poster.

Your first option is to go local. Check your local business directory and look either for “graphics” or “signs.”

Your second option is to use an online printing company. There are several that specialize in doing conference posters. Most are able to send you your poster by overnight courier the day after you submit it.

I’ve talked about some of the services offered by PosterSession. Their blog has inspired some posts of my own.

I’ve also recently learned about MakeSigns. Their website has a lot of resources, especially if you’re one of the many who still insist on using PowerPoint for posters. (I say again: There are better options!) In particular, they have a lot of PowerPoint templates at many different sizes.

Minor quibble: Their page on web graphics cautions against low resolution, which I have also done. But it doesn’t make it clear that an online graphic may be fine on a poster, regardless of the low dots per inch (DPI). The number of dots (pixels) of the picture when printed is what matters, not the screen resolution.

Example: An online picture that is 1,920 pixels wide at 96 DPI is 20 inches on screen. If you print that 5 inches wide on a poster, the resolution is 384 DPI, far more than you need. If you wanted to use that picture as a massive background 48 inches wide, however, you have 40 DPI and it would look terrible.

Print the poster in letter paper sized pieces.

This is favoured by people who just want to re-use their PowerPoint slides as posters. Print off each slide, glue them to some coloured paper, and there’s your poster.

These generally look quite poor. It is very difficult to get all the pieces glued down onto decent heavy paper, then tack up all the individual pieces so they are neatly aligned. Plus, most inkjet printers and standard letter paper don’t have the crispness of printing or the bright colours that you can get from a high end plotter printer.

Not recommended.

16 June 2011

The edge effect

At a conference last week, I had two posters next to each other. The next day, I scribbled this in my notes:

I think the difference in traffic was not due to any inherent difference in the poster, either scientifically or graphically, but was because of the room layout.

All the posters in the session were mounted on single poster boards. Every single mounting board had one had one edge along a wall (or, in the center, a curtained room divider). This meant that poster viewers couldn’t walk along a row of posters. They had to walk in towards the wall, surrounded by posters on either side, over and over again.

Remember that people read from left to right, so the starting point of a poster is on the left edge.

The poster marked “more traffic” above has its left edge away from the wall. As soon as people walked into that cul-de-sac, they first thing they will tend to read will be the “more traffic” poster.

The poster on the right (marked “less traffic”) has its left edge next to the wall. To get to the start of this poster, people have to make their way all to the end of the cul-de-sac. The path might be blocked with people reading the other poster, or people reading the end of the poster they want to start reading.

One poster had an unfair advantage in attracting attention over the other.

And the moral of the story is: Conference organizers, think about the patterns of foot traffic in the poster session! They influence what posters get viewed the most.

09 June 2011

Critique: Motor units

Today’s poster comes from Mike Pascoe, and is used with his permission. Click to enlarge:

There is a lot to love about this poster.

  1. Uncluttered. The text is pared down and the graphs are given ample margins to let white space. It is not an intimidating poster. It invites in people, saying, “Hey, I am going to tell you my story and be done in two minutes. It won’t take you half an hour to decipher me.”
  2. Aligned. The columns are clean (though not perfect).
  3. There is no confusion about where I should read next, even when the poster switches to the reading order going sideways (Results section 4 to 5) instead of down (the rest of the poster).
  4. The artwork of the arms in Results section 1 acts as a nice entry point for passers-by.
  5. It has a QR code, which is at the vanguard of today’s conference poster technology (which I say purely and only because I wrote a blog post about them). Just so you don’t have to scan the image above, it links here. This is a full package of everything someone at a poster session might want to follow-up on.

This poster succeeds brilliantly on those points alone. Do not be fooled that the list I am about to give of potential improvements is longer than the list above makes this a bad poster. It does not. The points above weigh heavily. The points below are trivial.

If I were to pick on a few points where maybe things could be improved...

  1. The combination of Times New Roman and Arial is readable, but lacks character. A little something more daring, perhaps for the headings, would make this a little less “plain white bread.”
  2. Arial is used throughout the results, for both sub-headings and graph labels. Those are more text-level than heading level, and putting those in Time New Roman might have been a better match. The poster is perfectly consistent, which is good, but I am not sure if it meets expectations about hierarchy.
  3. The colours of results sections 2 and 3 could be made more consistent.
  4. Similarly, the graphs in results sections 4 and 5 use symbols filled with colour, and the graphs in results section 6 don’t.
  5. Containing only text, that all important left hand column is just a trifle gray.
  6. I remain unconvinced that poster numbers (in the upper right under the QR code) are helpful to viewers.
  7. The first section of Results, um, isn’t results. “Experimental setup” should be in methods.
  8. The two “Experimental setup” pictures are further to the left than the corresponding pictures below them.
  9. The Results section heading could be made even more descriptive. For instance, instead of, “Sustained contraction characteristics,” it could be worth the extra bit of space to say, “Sustained contraction characteristics differ in young and old.”
  10. And now we’re down to truly nitpicky stuff... But I am not alone in disliking a comma before an ampersand symbol (in the list of author’s names).
This is a great looking poster, and I bet it went down well.

02 June 2011

Breaking the hourglass for headings that holler

Making a poster like a journal article is one of the easiest traps for a poster maker to fall into.

Much of the good advice about posters from Mike the Mad Biologist will be familiar to readers. But this one is something I haven’t written about:

Informative headings are your friends. I see too many posters with sections labelled “Abstract”, “Methods”, “Results”, and “Conclusions.” You have a summary! And some methods! Results and conclusions too! Bully for you. Use the section headings to inform the reader, while simultaneously describing the figure or table (see #2). Something like, “xyz genes are found only in clinical isolates” tells me what I should be looking for in the figure.

This is a classic case of doing something for no reason other than habit.

Academics write journal articles. It’s what we do. And journal articles have a very rigid structure that takes practice to master. So much so that it becomes second nature.

The classic journal structure has an acronym: IMRAD. The letters stand for the section headings that Mike refers to: Introduction, Methods, Results, And Discussion. This format leads to people saying that academic papers have an hourglass shape: They start off very broad in terms of problems and concepts (Introduction), increasingly become more specific and detailed (Methods and Results), then widen back out again to big picture stuff (Discussion).

You can even see this tapering from general to specific in the Introduction. The first sentence is usually about some very big issue, like terrorism or invasive species or the nature of sexism. Then it pares down to to a more specific question, then pares down again and again until you get to this particular paper.

This means that the specific research question is usually buried in the last paragraph of the Introduction. this could be two to five paragraphs down in the text. This is what my journalist friends call, “burying the lead.”

But you don’t need to follow that structure on a poster! At least, not to the degree of actually spelling out each heading.

Someone looking through conference posters wants to know the question you are trying to answer immediately. Putting those key points in big text as a heading helps the curious quickly figure out if this is something they want to read. The IMRAD format is so well known to academics that you can dispense with them entirely. I did that on one of the favourite posters that I’ve done, and it worked well, in my opinion.

Instead of, “Introduction blah blah blah research question?”, shorten it to, “Research question.” Leave all the rationale and justification and background and blah blah blah off the poster and say it  to people who visit your poster.

Related posts

Poster Venn
Should your first presentation be a poster?
References on posters

Trap photo by kevindean on Flickr; hourglass photo by bogenfreund on Flickr; both used under a Creative Commons license.