Of the interesting comments left on my post about Mark Noad's redesigned London Underground map, one of the most interesting was left by eminent typographer and designer Erik Spiekermann. In it, he stated what he called a "common misunderstanding" about Harry Beck's legendary wayfinding display: that it isn't a map at all, "it's a diagram. Not meant to show geographic relationships, but connections." Spiekermann criticized Noad for mixing the two concepts in his redesign, which attempted to combine Beck's clean lines with added geographic accuracy. And when tech blogger (and occasional Fast Company contributor) Tim Carmody responded to Spiekermann's comment via Twitter, Spiekermann derided him for using the word "map" again. Harrumphing and profanity ensued.
Twitter slapfights don't usually get me thinking, but this one did. Regardless of whether you think Noad's experiment is successful, interesting, ill-conceived, self-aggrandizing, or something else, his intent — like any designer's intent — was simply to create something useful. As a visual display, the London Underground diagram/map (don't hit me, Erik) has a use: to help someone see where they are and decide where they're going. That's the end; its map-ness or diagram-ness is simply the means. Matching ideal means to useful ends is "what a designer does." But if a redesign doesn't preserve the purity of the original, successful means (in this case, Beck's diagrammatic approach) does that make it an a priori failure?[Click to view larger]
Designing is often described as "solving," but isn't it more about engaging?
I agree with Spiekermann to a point — I don't find the London Underground diagram confusing as-is, and adding "some" map-like geographic accuracy (but not "all" of it) could easily send a very conflicting message to the user about what to expect, and how to use the display. But here's the thing: the Underground diagram already does send very conflicting messages — it's constantly saying, "I look like a map, but I am not one. Psych!" And yet if you ask 100 people what the London Underground display "is," I'll bet cash money that 100 people will call it a map. Because that's its use. And if it's use-ful in that way, aren't designers' definitions about what it "is" or "isn't" simply academic?
Spiekermann (who also designed Berlin's subway diagram) is right that we "misunderstand" the London Underground diagram when we think of it as a map. But so what? If that misunderstanding makes no huge difference to its usefulness — which is probably the case for most people, who use the diagram (and think of it) in a basically maplike way and get on fine — the misunderstanding is moot, neutral, meaningless. But if that misunderstanding does make a difference — for example, makes it more difficult for a user to do what they want to do, which is see where they are and see where they're going — isn't that a shortcoming of the design, not the person? And if that misunderstanding is indeed "common," as Spiekermann says, mightn't it make sense to bring the design more in line with users' maplike expectations? Maybe, maybe not. That was what Noad was experimenting with.[The old version. And the new...]
Noad's redesign is of questionable utility, but every redesign is: That's the point. The questioning. "Is this artifact a diagram or a map?" is an interesting question, but it's not really the question that Noad's redesign is asking. It's asking, "could an artifact that's a diagram and a map be useful?" I live in New York, where our subway display has been a map/diagram mix for decades, so I'm inclined to think "yes, it would be useful." Most native Londoners, who've been trained to think "diagrammatically" about their subway system for 80 years, might think otherwise. But that's who defines the answers (note the plural) to that "questionable utility" in the end: the users, not the designers.
Designing is often described as "solving," but isn't it more about engaging — connecting with users, conversing with them even, via these artifacts? That's not to say designers and design-observers shouldn't think critically about why artifacts "work" or not. But no design is "right," "true," or "done" for everyone at all times; the conversation will change. Applying definitions and categories and semantic hair-splitting can end conversations with users or usefully extend them. When in doubt, isn't the latter a better way to go — even if it means changing the subject?