20 September 2018

Critique: Enemy myna

Today’s poster comes from Jennifer Pannell. Click to enlarge!


Before I get into some details, Jennifer noted that this poster isn’t exactly the way she wanted it to look:

I had some problems with the font, though. Scribus won’t embed them and so they look terrible unless you zoom in 100%, so it might look terrible as a png.

I don’t know if the export problem might explain a few little issues, like dumb quotes instead of curly quotes. Or that there are lines after paragraphs on the left, but not the bottom right. “Fig” should probably have a period after it throughout.

Jennifer’s poster has a clear two column format, with some attractive graphics to bring the casual browsers on board.

There are some positioning choices I find odd, though. There is more open space on this poster than many I see, which is good, yet somehow items still seem to end up feeling crammed together. For example, here’s a close up on the upper right corner:


There is open space around the images of the bird, and that’s good. But there are at least three points of things touching or almost touching that makes it uncomfortable to look at. Two of those could be fixed by making the birds about 90-95% of the size they are now.

  1. The “g” in “findings” in the title is almost touching the “J” below it.
  2. The upper wingtip is almost touching the institutional address. Why the address is split over two lines I cannot say. It seems like there is enough space to put it on one line.
  3. Most seriously, the beak of the lower bird is overlapping with a data table and partially obscuring a number. (A statistically significant one, no less!)


Speaking of the table, I have mixed feelings about having the table and figures on white backgrounds. If you’re going to use a gray background, I’d be tempted to use that gray throughout. Instead of white, I might have tried either a transparent background or a very light gray (maybe 10%) to make the figure edges less obvious.

The gray background is an interesting choice. It means that you can use either back or white as text and it will still be readable, which makes it more visually interesting and open up some options. But the contrast is half what it would be if the background was solid white or solid black.

Figure 3 is missing some information needed to interpret it. There are no standards for what box plots show. Is the line dividing the box the mean or the median? What does the box show? 50% of data? Do the whiskers show a calculation of variation (like standard deviation) or a representation of actual data (95% of data)?

I do like the big circle acting as a way in to the poster, and the little decorative touches like the branches in the left column.

No comments: