Paul Graham, the Commons, and How Google Stopped Being Google

By Alexis C. Madrigal

Google has increasingly tailored its search service with individuals' data, but clickthrough rates can't reflect a community's desires.

canon_615.jpg

Startup incubator founder Paul Graham has a great post about the "frightening" startups he thinks "could make you a billionaire." Within it, there's a pocket critique of Google's latest iterations that rings true:

[L]ately when using Google search I've found myself nostalgic for the old days, when Google was true to its own slightly aspy self. Google used to give me a page of the right answers, fast, with no clutter. Now the results seem inspired by the Scientologist principle that what's true is what's true for you. And the pages don't have the clean, sparse feel they used to. Google search results used to look like the output of a Unix utility. Now if I accidentally put the cursor in the wrong place, anything might happen.

Thinking about Google's company dynamics, I have a hypothesis about how this might have happened. It has everything to do with the limitations of data-driven design. Let me stress that this is just a guess.

So, Google collects data on user behavior. They make changes to the user interface and collect more data, analyzing what they've got to see what users "liked" more. All these little tweaks and personalizations and speed enhancements got incorporated on the basis of THE DATA. And who could argue with that? Give the users what they want!

But there is a problem. While the amount of data Google can collect is enormous, it all comes at the level of an individual's behavior. There is no place in the data for what the community of Google users might want for the community as a whole, even if it came at the expense of the individual search from time to time. And what many of us wanted from Google, what we loved about Google, was the idea that it showed us The Known Universe of Stuff on a Topic. It was a fiction, but it was also an imperfect, metastatic, hyperlinked version of a canon. And even if I didn't want to click on any of the first 10 things on that list -- and so my search was counted as a failure -- I could still appreciate the idea of that Google and understand its importance within the context of this unknowable mass of pages on the web.

It isn't that anymore, though. Searches are geospecific and social network-dependent. All of which is fine and useful, but that's not what made us love Google's search engine. The more the search engine -- and the web more generally -- adjust themselves to us, the less they represent a collective idea of what is known.

I'm not the first person to make this point, but there is an important lesson that I hope we take from this relatively minor tragedy: The aggregation of individual data does not a commons make.

This article available online at:

http://www.theatlantic.com/technology/archive/2012/03/paul-graham-the-commons-and-how-google-stopped-being-google/254917/