The Internet turns open-source licensing on its head. Copyleft is neither copyright nor copyleft anymore in the Web world. It's just copy, because distribution of a service over the Internet doesn't count as distribution in the archaic licensing language that plagues most open-source licenses.
It's a problem that the Free Software Foundation chose to ignore, and then beat on Tim O'Reilly to atone for its own failing. Tim, for his part, thinks that the open-source world is missing the boat, which is chugging along unmindful of antiquated things like software licenses when the real value is in data-centric applications whose value lies in network effects (architecture of participation).
I get this point, and he's probably right in the long run. My question for now, however, is: if the value in Web 2.0 is data, why can't Web 2.0 get over its own software fetish?
For it's not just the open-source crowd amongst non-Web Neanderthals that is stuck on software. It's also the very group that Tim praises for taking the argument to the next level--the Web 2.0 crowd--that insists on the value of its services while locking down the value of its software. Release the code if code no longer matters. Let others build on it. Lock us in through data. Fine. But release our software.
That's all we ask.
We have a long way to go before the world abandons software qua software. We're still muddling through green screens in large enterprises, after all. So software qua software matters and will continue to matter for a long time. In this world, open-source licensing is relevant and powerful.
Open source, however, is losing its grip on the Web because the Web has decided to play by different licensing rules and open source has not bothered to keep up. This is not the Web's fault. It's open source's for fetishing the past, as Tim indicates.
But this doesn't lead me to Tim's conclusion. Instead, it leads me to ask for an upgrade to open-source licenses so as to require the Web world to play by the same rules as the old software world. If you take my software, modify it, and redistribute it, you need to contribute back. It's really very simple.
Given that the Web is powered by data, not software, this shouldn't be a big problem for the Web 2.0 world. Contributing back into the common software fund should be an opportunity, not a problem. So why haven't Google et al. stepped up to the plate to give back their derivative works? Why do they treat the GPL as if it were the BSD license, requiring nothing in return for all the value given to them?
Because they can. Because open-source software licenses are irrelevant in the software-as-a-service world.
But given that they won't release the code, anyway, I guess it also means, as I noted above, that the Web 2.0 world is still essentially a software world after all, or thinks that it is.
Maybe we'll all wake up 20 years from now and laugh about our fetal focus on code. But for now, both Web companies and landlubber companies are still focused on software. While we're thinking about software, everyone should be held to the same open-source licensing rules, to the extent that we're dipping into the same pool of code. This means we need new open-source licenses in the mold of CPAL, , etc.
Tim is probably right. Software probably is yesterday's discussion. But while the world continues to behave as if software matters, today, then it's important to live in that world. When all software is open source (and I think it's almost unquestionable that we will at least get to the point that every technology company will have a serious open-source bent to it within the next 10 years), we can focus on data and perhaps a Free Data Foundation (FDF) will spring up to ensure we keep it free.
For now, I just want the software to be open. I'm old-fashioned like that.