“But now we’ve reached an interesting point in our project’s growth: our grant ends on June 30, and, under the terms of our grant, we’re open-sourcing the EveryBlock publishing system so that anybody will be able to take the code to create similar sites. That’s a Good Thing, in that EveryBlock’s philosophies and tools will have the opportunity to spread around the world much faster than we could have done on our own, but it puts the six of us EveryBlockers in an odd spot. How do we sustain our project if our code is free to the world?” – Adrian Holovaty
Adrian asks an interesting question, one that we actually are posed with quite frequently; most recently during a consult yesterday in which a customer was considering open sourcing certain assets in an effort to leverage the distribution advantages that approach offers. While I would not be presumptuous enough to claim that I have the answer to Adrian’s query, I would submit that I have an answer. Or perhaps part of one.
The answer is dependent on an assertion that is becoming more common these days, the most visible version of which is Tim O’Reilly’s phrasing: “Data is the Intel Inside.”
Consider Google. When people focus – predictably – on the almost childlike simplicity of their ad driven revenue model, lost in the shuffle is perhaps Google’s greatest asset: its data. The kind of data volume that allows them to predict disease. As Nick Carr puts it, “He with the most data wins.”
Which brings us back to EveryBlock and the other would-be open source projects we speak with. As you might have concluded from the above, my contention is that – ultimately – the value of the service will depend not on the codebase, but on the data it generates – and more importantly – collects.
True, the imminent open sourcing of EveryBlock does indeed open the door to would-be competitors, which could potentially have a deleterious effect on the network effects that can make or break web based properties (how many eBay’s can the market bear, and all). But in my view, the code is but a fraction of the value of the service. Of the barriers to entry for the market, far more critical is the traction, adoption and critical mass fostered and maintained by the properties in question.
It wasn’t del.icio.us clones that caused Clay Shirky to leave del.icio.us, after all. That would be Yahoo.
Were Adrian to ask me, then, I would issue the same recommendation to him that I did in my consult yesterday: seek revenue models that leverage the distribution advantages afforded by open source. The most convincing of these, of course, is the telemetry acquisition model with its ad sales byproduct that is Google. In EveryBlock’s case, I might expore opportunities to mine, package and sell telemetry back to the commercial and governmental organizations that are covered by EveryBlock.
Open sourcing an asset may indeed create threats while it manufactures new opportunity. But in many if not most cases, projects would be best served by focusing on the problem of traction and critical mass, because that is an asset far more difficult to assail.
Either way, open source can and should be extremely complementary to your business model. It just not might be the one you predicted.
