Visual Vocabulary 9 Years Later

For no apparent reason, I suddenly remembered Jesse James Garrett’s Visual Vocabulary today, which he promulgated almost 9 years ago this October.

I recall at the time that there were a number of people hailing it as the first true user experience documentation standard, and I saw no reason to disagree with them. Yet after a couple of years, I hadn’t really heard of anyone using it for real. Indeed, when it came to visual languages and UX, it was more often than not the dreaded UML that was being bandied about.

So why did this noble attempt at “describing information architecture and interaction design” meet with nothing much more than a collective shrug of the shoulders from the world of user experience practitioners? At least a contributing factor must have been the overwhelming presence of UML when one looked at any visual method of system architecture representation. Indeed, UML had been around for about three years before JJG thought up his Visual Vocabulary, so the latter’s existence was always a bit of mystery in that respect. Odd too that UML isn’t mentioned in this interview with its creator, three years after the VV was launched.

So with hindsight – was the VV basically a problem in search of a solution? Compared to the alternative of knocking up a simple screen-by-screen prototype, what do these diagrams (18K PDF) really bring to anyone’s party? Even assuming the syntax was understood, at the level of granularity we’re dealing with, I think most stake holders could be forgiven for saying  “OK, now show me the screen designs, please.” Developers, if they even looked at it at all, would want to know far more than the “language” could communicate. In terms of the activity of design, abstract boxes and lines are also unlikely to be useful in deciding the best experience for something unless there’s truly a large number of alternatives to how you could (say) log in or do a search. Insulating yourself from context, and the subtle effects that has on interaction design, is not a very good idea.

So this combined with the fact that we now have AJAX and the decline of the entire page-based model, means that I think we can now consign the Visual Vocabulary to the dustbin of quite nice, but useless ideas.

I think Garrett should therefore throw a failure party.