Once allergic to open source – “let me get this straight…you’re giving the software away? for free?” – venture capitalists have in recent years adapted to the reality of open source, arriving at the realization that the model possesses certain intrinsic advantages in both distribution and community building. So completely have these lessons been taken to heart, in fact, that it was apparent in a few cases that the VC community had actually over-rotated with respect to open source. Tales of powerful VCs insisting that the software they were considering funding be open source abounded – whether open sourcing the software in question was in the interests of the startups they were funding or not. Which, predictably, it wasn’t always.
Consider the case of del.icio.us, as an example. Would open sourcing it have helped it compete more effectively against, say, Furl? Remember them? As much as I believe in open source software, speak on its behalf and rely on it day to day, my personal view has never been that it is the one and only way to develop and deliver software. In the case of del.icio.us in particular, I’m of the opinion that a mandated open sourcing of the code would have negatively impacted the user experience, since the utility of the service depends to some degree on the network effect. The same network effect that multiple, distributed instances would actively act to undermine.
But what was true for del.icio.us need not be for every web application that followed; certainly WordPress’ gains at Movable Type’s expense came largely when the former was open source and the latter was not.
The question now before us is whether open source will be the differentiator for Identi.ca that Jaiku, Plurk, Pownce, and so on have lacked.
Identi.ca, in case you haven’t seen it yet, is essentially a feature-poor alternative to Twitter. But while it lacks features like SMS updates and notifications or a Twitter-compatible API, it’s open source, which means that additions could come quickly. Where quickly can mean “an hour.”
Beyond the obvious appeal of the potential for accelerated development via community contributions, because the project code (dubbed Laconica) is open and available, individuals are free to run their own, localized versions of the service. As Russell Beattie is already doing here. While the impact of that independence on the central Identi.ca service will be both positive (“sweet, we can run our own Twitter”) and negative (“wait, we have to post to and monitor more than one channel?”), it’s undeniably differentiating.
One very interesting note: Laconica is licensed under the Affero GPL. For those less than familiar with that particular license, it’s essentially the GPL reciprocal-style license with one important distinction: it regards network deployment as equivalent to distribution. Practically speaking, this means that anyone – be they a commercial entity or individual – deploying the application in a network context will be required to make their updates or modifications available under precisely the same terms, which is atypical. Lest you think these are obscure obligations, they’re already being run into. This could be an interesting test for the future viability, or at least popularity, for Affero-style licenses that attempt to reframe the concept of distribution by incorporating the increasing reality of Software-as-a-Service delivery models. Unless I’m mistaken, as well, the selection of the APGL for Laconica will mean that this unofficial Google Code repository will shortly be taken down, as the AGPL is anathema to Google.
Will open source be the spark that ignites a mass departure away from Twitter? It’s impossible to say at this point, and despite the remarkable rate of adoption amongst my small community, I’m far from replicating my Twitter network on Identi.ca. Meaning that a wholesale move for me, at least, is unlikely (not least b/c there’s no Twhirl integration). And while many have predicted doom for Twitter in the past, it’s still here and still chugging right along. Well, actually, it’s down at the moment, but you know what I mean.
Wherever the services head from here on out, James is very likely corrrect: Identi.ca is likely to make an excellent personal trainer for Twitter. More so, even, than Jaiku, Pownce and friends. Why?
Because the biggest community wins.