One of the great things about being the world’s leading purveyor of urbanism for the social media ageTM is that, sometimes, people on social media come to you with interesting questions about how the world’s cities work.
One of the less great things is that, sometimes, you don’t know the answer.
Last night, David Freeborn asked the following:
This is a good question – one I’ve been wondering about myself. Not so long ago, Google Maps showed urban areas in a uniform grey. A few weeks ago, though, it’s started to show some of them in that brown/orange shade.
My first thought, having wasted the 90s playing SimCity, was that this is the difference between residential and commercial areas. The UK doesn’t zone its cities in blocks in the way the game, based on the American experience, did – but there is definitely an identifiable difference between, say, a town centre and a housing estate. So maybe the diverging colour schemes reflect that.
But that, confusingly, would mean the colleges of Oxford counted as “commercial” premises:
...but the area around Tottenham Court Road station in London’s West End doesn’t.
Which doesn’t really stack up.
So, to find out what is actually going on, I – what else? – Googled it. The search uncovered this Google blog post which explains things a bit.
As you explore the new map, you’ll notice areas shaded in orange representing “areas of interest”—places where there’s a lot of activities and things to do. To find an “area of interest” just open Google Maps and look around you. When you’ve found an orange-shaded area, zoom in to see more details about each venue and tap one for more info.
This, then, is why the orange areas are more commercial, but are neither entirely commercial, nor cover all commercial areas. “Areas of interest” are a more nebulous sort of category, covering, basically, the bits that are worth exploring.
So central Oxford shows up because Oxford colleges are the sort of things that might be of interest to the passing visitor; while that bit of central London doesn’t, presumably because construction works means there are fewer shops and restaurants there than in surrounding areas, and those which exist frankly aren’t that interesting.
Or that, at least, is the theory. I’m not entirely convinced by this one.
It’s certainly true that some bits of cities have more stuff to discover, a greater density of incident, than others – and these are more likely to look like downtowns than residential suburbs.
But, to go back to Oxford, why are University College and St Edmund Hall areas of interest, while New College or All Souls aren’t?
By the same token, why is the corner round Tottenham Court Road station grey, while equally unpromising stretches of nearby New Oxford Street are orange?
I’m sure the inconsistencies make perfect sense to the algorithm. I’m just not sure they make that much sense to the human mind.
One factor is probably the types of places Google classes as interesting, which are, more often than not, the sort of places where you can spend money. To quote that blogpost again:
We determine “areas of interest” with an algorithmic process that allows us to highlight the areas with the highest concentration of restaurants, bars and shops.
This probably makes sense if you’re a major multinational corporation selling advertising. But there’s an argument that an algorithm that thinks Trafalgar Square is less an “area of interest” than the restaurants across the road is not fit for purpose.
Most of us, I’m sure, can instinctively tell the difference between an area of a city that is worth exploring on foot, and one that isn’t. But defining that difference, and putting it into words, is rather harder.
And if you can’t define areas of interest, how you can accurately programme an algorithm to look for them?
“In high-density areas like NYC,” says Google, “we use a human touch to make sure we’re showing the most active areas.” Perhaps the human touch is needed elsewhere, too.
Jonn Elledge is the editor of CityMetric. He tweets as @jonnelledge.