Spring, open source, and the tragedy of the not-quite-commons

What does the popular Spring application framework, and the minor blowup ongoing over changes its pay-for-maintenance plan, have in common with community activism and city planning in Baltimore? Well, stay with me, because I think the parallels are instructive.

I live in Baltimore, and one of the live political controversies in the area is ongoing in an upscale neighborhood called Roland Park. The west edge of the neighborhood is dominated by a country club; for decades, the club has kept much of the vast open space on its grounds open to the public, and it has been a popular hangout for dog walkers and, when it snows, sledders. Now the country club wants to sell the land to a company that plans to build a retirement complex on it, and neighborhood residents are up in arms. After all, they've become accustomed to having access to the open space, the presence of which improves their lives and their property values. Their rallying cry is "Keep the park in Roland Park" -- even though the land in questions is not a public park of any sort, but a piece of privately owned property.

So what does this have to do with open source software? Consider Spring, an open source Java application framework that has quickly become very popular; Spring skills are much in demand [1], and of course there are any number of fine JavaWorld articles on the subject (see for instance here [2] and here [3] and here [4]). And many other open source developers have based their own projects around Spring.

Thus there was much consternation when SpringSource, the for-profit, VC-funded company that puts Spring out, announced that you'll need to buy a maintenance policy if you want bug fixes and updates more than three months after a major release [5]. The resulting blowup was rather predictable, with paranoid musings on what would be charged for next [6] and even calls for a Spring fork [7].

As Spring creator Rod Johnson points out in TheServerSide thread where much of the drama played out, all updates will still be available for free immediately in source code form [8]; this is a requirement of the license, obviously. In fact, this is what open source is really about, and Johnson somewhat snidely contrasts those who are interested in downloading, compiling, and tinkering with the code with those who just expect to be able to download the binaries for free (as in beer). The problem lies not in Spring's adherence to its license, which is impeccable; for anyone other than a certain hard core, it's not even that SpringSource took money from VC investors who expect to make it back and then some. Like the residents of Roland Park, developers who use Spring have gotten used to something -- something that could have been taken away at any time, but that they never expected to lose. And once someone gets accustomed to something, it's really, really hard to take that away without getting some -- or a lot -- of blowback. It's all rather instructive as to the human dynamics that underlies open source software projects, and it will be interesting to see if Spring developers fork the thing, or grumble and accept the new reality.