Undo the underline

Next time you’re reading a book – or, indeed, any professionally typeset publication, whether it be a journal or a magazine or a newspaper – look for something. Look for underlined text. You know, like this.

I pulled a dozen different books off my bookshelf while writing this post. I opened them up at random, to a different two page spread. I scanned the pages. The number of underlines I saw was zero.

Yet I see people underlining stuff on posters all the time. There are two reasons for this.

1. Underlining is the quickest and easiest way to emphasize text in handwriting. But everyone makes conference posters with computers, where other formatting tools are are available as air.
2. The silly little underline button is visible in a prominent place on every piece of basic office software. It sits there, tempting you to press it. “Come on, baby. You’ve pressed pressed bold. You pressed italic. Why can’t you press me, too?”

Pro typesetters normally use italics for emphasis, particularly for long texts. You also see bold used for emphasis, but less often. Bold text is good for posters, however, because it is more recognizable when skimming text. An underline crosses and obscures the shape of descending letters, like g, j, and q, making the text harder to read.

Underlining is one of those little signs that scream “Amateur!” once you recognize it. Just don’t touch it.

Related posts

The data prison

Critique and makeover: Healthcare

Today’s poster comes from reader Chris Skedgel, and is shown with permission (click to enlarge):

Chris wrote:

I recognise that it is heavy on text, and although I appreciate your advice to think of a poster as a business card rather than a condensed manuscript, I am loathe to hang a poster that I believe doesn’t sufficiently explain what I'm doing or my results. My compromise was to use a font large enough to be read at a distance, with some whitespace to avoid the dreaded “wall of text.”

The tension between completeness and readability is very real. Some of the earliest posts on the blog were about the decision to “writing down everything” and “write very little.” It takes a lot of practice (and ruthlessness) to cut, and cut, and cut, and say the most important stuff in the fewest words. Academics don't always get the most practice at being concise. Usually, comprehensiveness is valued more than concision.

Chris continues:

I also know your thoughts on logos, but I stuck with the institutional template rather than rock the boat. It does crowd out the title a bit, but I also think it looks sharp.

This is one case where I actually like the logo. First, that the white box bleeds off the page. That makes it clear that it’s a deliberate design element, not slapdash addition. Second, there's a single logo, rather than the usual bookends. No duplication. The logo makes its point once, and is done with it.

Here’s a makeover, with some mild chances to the text.

First, the underlined text went away. You almost never see underlines in professionally typeset text. Bold or italics do the job. When I showed this, Chris replied, “I see your point about underlining – it does look a lot cleaner now.”

Speaking of emphasis, I removed the bold from the Conclusions, and the "read more". The less bold, the more punch the remaining bold has. When everything is emphasized, nothing is.

The title, authors, and institutions looked far too crowded; I opened up some space between them. Because the logo was offset and only on one side, centering the title in the remaining space at top made little sense because it didn’t line up with anything below it. The difference is not huge, though. Likewise, I removed the shadowing from the title, but I don’t mind it with the shadowing. Just trying alternate looks.

Personally, I find the font for the main text to be a little fussy and doesn’t read terribly well from a distance. It does have a bit of personality, so I didn’t change it, not wanting to mess with the poster’s style too much.

I tweaked the placement of the uni logo very slightly to align with the edge of the main text box.

One thing I could not do with the file Chris sent me, but would like to, would be to make the font in the graphs the same as the rest of the poster.

Update: At the suggestion of Mike Taylor in the comments, here is a version of the poster without the frame:

Lessons from Samurai Jack

The animated television show Samurai Jack (created by Gennedy Tartakovsky) won rave reviews for its bold, distinctive designs. A key element to the show’s look was the extensive use of colour holds (at least, that’s what I’ve heard it called in comics). See if you can spot the difference. Here’s Jack:

Now compare Jack to a previous Tartakovsky project, The Powerpuff Girls. What’s different?

Here’s an explanation of colour holds (my emphasis):

“Overlays” or “color holds” (where there is no black outline or the outline is in a specific colour) are done on a separate sheet of acetate or vellum overlaying the original art. This is usually done by the penciler or inker as a special effect – simulating invisibility or colours in a fire or explosion.

Now back to Jack. The only place you see black lines is to delineate his eyes and a couple of other facial features.

Early in the “making of” clip below, creator Tartakovsky talks about the design, saying:

If you look at cartoons, every character has a black outline around them. For us, we took the line completely off, so if it’s a white robe, you just see the white shape, you see no linework around it.

I seem to remember a longer version of this interview where Tartakovsky said this style, with no lines around objects, was something you saw a lot in kid’s books. And this gave that kind of artwork a real charm.

Here’s a clip that shows the astonishing design and graphics sensibilities the creative team brought to the show. It takes about a minute for the ball to get rolling, but when it does...


Note that this amazing action sequence depends on characters not having lines around them.

I thought about this when I received this request for feedback from Svetoslava Antonova-Baumann (as always, click for a closer look). She wrote:

Before coming across your blog last week, I hadn’t made a single poster in my life. Armed with your advice, I managed to produce my first specimen today.

My first thought was, “Oh no, not boxes again.” Lines around every section! I went Samurai Jack on this, erasing the first, most obvious, set of black lines around the boxes:

Just that one change immediately lifts and lightens the poster. Next, I eliminated the horizontal dividers within the columns:

Then I thought, “Maybe we can get rid of some more black outlines in the flowchart in the center.”

But I will add lines back in, in this case, to create an explicit X axis in the graphs at right, while taking out the horizontal gridlines:

Then I went about removing the white box around the institutional logo, and resizing and moving the author’s picture so that they both sit at a more comfortable distance from the title.

And a final, small move to the last “Conclusions and future work” section heading, again moving it away from the text.

Differences in colour alone can do the job of dividing spaces just fine. Black lines are almost always gratuitous.

I sent this last version to Svetoslava, who replied:

I really like the new version without the black frames. It feels somewhat “fresher”.

Samurai means “to serve.” I live to serve.

Related posts

Boxism
Critique and makeover: Aptamer biosensors

Link roundup for April 2013

Hat tip to Matt Thompson for spotting this comic:

An interview with Amanda Cox on data visualization. Cox is the graphics designer for the New York Times, which has a reputation for fine graphics. Lots of very interesting ideas here. For instance:

The ability to ask good questions is really what we start with. I come from a statistics background, and I’m finding statistics students’ portfolios are crazy weak compared to the computer science students, even though they’re playing with the same problems. I think it’s because comp sci students are encouraged to play, whereas stats majors it’s, “here’s your rule book, now make things.”

Don’t do this. Logo fails. Each one makes an interesting study: can you figure out why they failed?

Seth Godin talks about rules for public things. To sum up, with my comments in parentheses:

1. The more often a device is used by first-time users, the more standardized the interface should be. (Most academics at a conference are early career researchers.)
2. Who gets left out is the most important question. (No poster is meant for everyone. Also, think about people with less than perfect vision.)
3. The best interface is no interface.

The Scholarly Kitchen has an interview with Michael Bierut about type. It includes this bit about science:

Q: The Higgs Boson — perhaps the most important scientific discovery of the past 50 years — was announced via a PowerPoint deck that used the Comic Sans typeface extensively. Why do you think the scientists chose this? Was it a wise choice?

A: Well, as I understand it, the scientists at CERN were actually surprised that people commented on this. Reportedly Fabiola Gianotti, the coordinator of the CERN program to find the Higgs Boson, was asked why she had selected Comic Sans. She simply said, “Because I like it.”

Dennis Eckmeier provides a nice examination of the shortcomings of the standard bar graph.

Elizabeth J. Petro tweeted how to put a poster away (snipped):

Just observed an incredibly clever poster rolling method: leave one side attached to board while rolling; then detach.

Oh, and while I have talked about the usefulness of QR codes here on the blog, even I have to admit this post kind of has a point:

Critique: antifungal drugs

The call came out on Twitter:

presenter wants feedback on layout of poster. Thoughts?

The tweet contained a link to this picture (click to enlarge):

The image size prevents a more detailed critique, but I sent back links to four posts here on the blog.

Abstract abolition: I put this one first, because just days before, I had done a critique where the main pathway to improving the poster was getting rid of the abstract. The abstract here is chewing up about 10% of the main text for no good purpose.

The epic logo post: The institutional logos make me cringe. These huge slab serif logos bookending the title completely overpowers the title, which is barely readable in the photo. A logo should never, never be more important than the title.

The data prison: Dense tables are the enemy of attracting viewers. I continue to be surprised by the fact that nobody seems to notice how journals lay out tables, with only a few horizontal lines.

Should your first presentation be a poster?: I provided this as an example of the problem of too much text. (I could have also used this post.)

As a bloggy bonus link, I add:

Boxism: This particular poster isn't bad, because at least the boxes are only one set deep. Boxes around just the columns would be an improvement over every single item on the poster.

And I don’t know what that bar over the three righthand columns at the top is doing.

That said, it appears the reading order is clear, and I also like the consistent colour scheme.

Hat tip to Biochem Belle for bringing this to my attention.

Critique and makeover: atlases

Today’s poster comes from Justin Ducote, who was kind enough to give me permission to show this poster. Click to enlarge:

Justin made with a PowerPoint template, and sent me the original file. This made it easy to do a fast makeover. I opened up the file, shuddered a bit at the vast amount of text, and went at it with two major goals in mind.

1. Fewer words.
2. Bigger words.

It was easy to hit those goals. First, I removed the abstract (goal #1). Remember what abstracts were created to do: to summarize an article when you couldn’t read the rest of the article. Abstracts make no sense when the “rest of the article” is on the same piece of paper.

The abstract was chewing up a fifth of the poster. Removing that abstract gave lots of room to maneuver on the left side of the poster. I made the head shots as big as I could, as they’re the most recognizable and attractive graphic on the poster.

Originally, the headings measured 25 points in size and the main text was 23 points: almost indistinguishable in size. I made the headings 44% bigger (goal #2), increasing them from from 25 point to 36 points. Similarly, the main text got 21% bigger, moving from 23 to 28 points.

Having said goal #2 was to make words bigger, it might seem contradictory at first that I made some of the print smaller. I shrunk references and figure legends by 22%, down to 18 points from 23.  This allowed the main text to fit, and it created a visual hierarchy. Instead of two text sizes that were almost the same size, there are three text sizes that are all distinctly different, clearly signalling their relative importance.

The right side of the poster required only a little more finessing. I reduced the contact information, and cut out one phrase in the Discussion to make the text fit.

I rearranged the title and the logos. The logo bookends were forcing the title to be off-centered. Given the logos were so different in proportion, the simplest solution was to embrace the asymmetry and put the title on the left. I could make the title bigger by removing the logos entirely, but I wanted to work with the original style as much as I can.

The University of California logo is obviously informing the colour palette of the poster, so I thought, “Why not just go all the way with it?” Originally, the scatterplots on the right had the only red on the poster. The revision uses blue and gold, like the rest of the poster.

Additional: Radical_Rave on Twitter offers some more suggestions.

Related posts

The epic logo post
Abstract abolition!

De-cluttering

“Cluttered.”

This is one of the most common problems with academic posters. “Clutter is a failure of design,” as Nancy Duarte wrote in Slide:ology. The good news is that you don’t have to know that much about design to fix this problem. All you need is one guiding principle.

Take out the trash.

Whether in a room or on a poster, trash stinks. Trash is the excess, the non-essential, the old stuff past its prime. Clutter is often just an accumulation of trash that people haven’t categorized as trash yet.

In a poster, trash is the long blocks of text nobody’s going to read. The institutional logos. The abstract. The 3-D perspective effects in your bar graphs. The nested boxes around every individual piece of the poster.

The art of cleaning is about making decisions about what you need and what you don’t. Once you think hard about what is essential, cleaning up your poster becomes much easier.

And as the caption to the picture above says, “If you have the right attitude, even taking out the trash is fun.”

Top picture from here; second photo by Ed Yourdon on Flickr; used under a Creative Commons license.