A confession: case types revisited
In Using design to prevent people from sharing stuff they didn’t read, Giulio Michelon wrote about… well, how to prevent people from sharing stuff they didn’t read. I actually read the article and liked (most of) it. Basically, I agree with his basic premise but am not sure that the solution he describes is optimal. But that’s for another article.
In this article I want to tell you how, shortly after reading Michelon’s piece, I committed the very crime he describes. I had been reading 100 Things Every Designer Needs to Know About People (great book! highly recommend) by Susan Weinschenk and in her discussion about case types, she writes, “There’s no research showing that the shapes of words help us read more accurately or more quickly.” That seemed super strange to me. I’ve been dabbling in the whole case debate for a while — nothing too serious, but it’s been on my radar — and I couldn’t imagine that was true. Furthermore, right below that statement, she gave the most comprehensive research review on the topic as a reference for further reading. The fact that such a paper even exists made me even more incredulous that “there’s no research…” could be true.
And then I did it. I tweeted the review without reading it. Now, in retrospect, I didn’t really need to out myself because I went back and read the paper, and then went back and reviewed my tweets, and they all made perfect sense. In other words, I pretty much “pulled it off”, unlike all those NPR readers who really looked foolish… But I have decided to come clean nonetheless, and not only to admit my wrong, but to attempt to compensate by sharing my thoughts on the article where it all began.
The Science of Word Recognition by Mike Jacobs at Microsoft
The following thoughts are in the order they arose while reading the article. There is no topical segue between them.
Respect
After summarizing and arguing against a model for understanding how people read words, Jacobs offers an explanation for why the original researcher might have supported it. Jacobs in no way wavers from his opinion that this first researcher is dead wrong, but he gives context to the findings, and in so doing, shows respect for his historical colleagues.
“Cattel [1886] supported the word shape model because it provided the best explanation of the available experimental evidence.”
Obviously the subtext here is that if Cattel had access to subsequent evidence, perhaps he wouldn’t have supported the same theory. Respectful disagreement with fellow tradesmen, even those who are long dead, is to be emulated.
Research
First, as Jacobs reviews the history of the research on how people read words, I found myself exuberant that such a breadth of research exists! I had no idea so many people cared so much and I was thrilled to find out.
Second, I started to think about the role of those of us in the biz when it comes to research. It seems to me that the studies that Jacobs reviewed were conducted by academics. But in the industry, a) there’s more money, and b) we are probably the biggest beneficiaries at the end of the day. Do we have a moral responsibility to invest in this research and to compartmentalize it from our individual businesses to work collaboratively for the good of the industry as a whole? If anything like this exists, I’d love to know about it. As an ex-scientist, I would welcome the opportunity to run experiments while keeping my day job.
In the section about neural network modeling, Jacobs describes a theory that reminds me a lot of (what I imagine) Facebook’s newsfeed algorithm (to be). He talks about synergetic negative and positive feedback loops: if a signal is important, the infrastructure for sending that signal is strengthened; at the same time, if a signal is not important, the infrastructure for sending that signal is weakened. Isn’t that how it works on Facebook (and the like)? You see more of what you see and less of what you don’t. Same concept, different context. Wonder what we can learn from that.
So there you have it. My sincere apologies and attempt at reform.
