uwMike.com

I'm in Waterloo at the moment, and next available to work in September 2008.

Ambigrams Again

September 21st, 2007

Tron Hoodie

A couple years ago I became interested in ambigrams, particularly the work of John Langdon. At the time, I got a hold of his Wordplay book, and began experimenting with some of my own designs.

I recently had a breakthough in figuring out how to create a surprisingly readable ambigram for my mechatronics engineering class. There was a general approval, and we ended up having the design embroidered on about 80 quality Campus Crew hoodies.

As someone with an amateur’s interest in typography and letterforms, I’ve found the doodling of these invertible designs to be an incredibly practical way to teach oneself the fundamental definitions of letters. Specifically, for each character, you can quickly figure out what strokes are inessential, and what areas tolerate the addition of extra decoratives that don’t impact the recognizability of the letter.

As an example of the first of these, consider some variations on the very robust capital R character:

Capital R Variations

It seems clear from this that the vertical stroke is completely disposable, but the diagonal is important to distinguish from the P, and the loop is necessary to distinguish from L and others. But consider what some seemingly innocent variations do to seriously affect the recognizability of the R:

Capital R Variations gone wrong

With only a very small additional stroke, the first has been almost completely transformed into a B. The second and third are now straddling awkward lines between the R and an H and X, respectively. In all of these cases, context could help establish the identity of a letter, but why depend on context when a very subtle change might have an incredibly clarifying effect?

Unless, of course, that innocent change has a devastating effect on the letter going the other way, in the case of ambigram construction. Designing an ambigram is a puzzle in balancing between making changes that clarify a letter in one direction at the expense of obscuring it in the other.

Another interesting form to look at is the that of the of the lowercase t, particularly as it compares to the f:

F and T

Given a basic cross to start from, the top has the choice of no flag, a forward flag, or a trailing flag; similarly, the bottom can be no tail, a forward tail, or a trailing tail. Of the nine possible combinations, which are recognizable as either an f or t, and which just look like nonsense? Based on the above image, it seems like the default is a T, and a forward flag on top denotes an F, unless there’s also a forward tail on the bottom, which pushes it back into T-territory. Try to imagine that character third from the left in front of the letters ree. What does it say? Tree? Free? Or is just a messed up capital E? Rorschach test?

tree-free

Discussion

  1. That letter you singled out looked a lot more like a “t” before I read you asking if it looked like an “f”. Now it looks like both/either.
    I think I want pattern recognition skills!

    Posted at 6:59 pm on September 26th by Christine.

  2. The fact that it’s ambiguous is generally a Bad Thing, but part of the art of ambigram design (the part that I get the least) is cloaking uncertainties using tricks in the surrounding letters… if the overall script has forward flags on lettertops, then you’re more likely to accept one on a letter that doesn’t ordinarily have it.

    Posted at 9:42 pm on September 26th by Mike Purvis.

Leave a Reply

You can use Markdown for style. I love hearing from readers, but please don’t hijack the discussion, use offensive language, or try to sell anything.

© 2004-2008, Mike Purvis, some rights reserved. I'm running Wordpress, and I have an RSS feed.