you may not: (a) copy (except for backup purposes), modify, adapt, redistribute, decompile, reverse engineer, disassemble, or create derivative works of the SDK or any part of the SDK; or (b) load any part of the SDK onto a mobile handset or any other hardware device except a personal computer, combine any part of the SDK with other software, or distribute any software or device incorporating a part of the SDK.
These conditions seem totally unacceptable to me and are likely to cause a reaction such as calling the Android SDK proprietary from anyone who values software freedom. However, let's not stop there and let me get back to what is right before these statements in the license text:
You may not use the SDK for any purpose not expressly permitted by this License Agreement. Except to the extent required by applicable third party licenses,
The last sentence is the meaningful one: it means that basically, the restrictions are not applicable to software that is covered by another free software license. So that's basically how Google can avoid breaking other licenses terms. Moreover, Google is not the copyright holder for all of the software released in the SDK, so they basically have no right to apply such restrictions to it. Huh, we're safe, after all, the Android SDK still is free software. But wait, is it really? Are all the files shipped with the Android SDK proven to be free software? If that was the case, then why would Google waste some time writing down these terms if they actually do not apply to anything in the SDK? So that point gives us fair enough reasons to suspect that there is actually proprietary software in the SDK. Yet another good reason to release a free SDK such as the Replicant SDK. Now let's consider the Android SDK manager utility: it is designed, down to the source code, to check for plug-ins and updates from Google. If I recall correctly (once again, correct me if I'm wrong), there used to be a clear license statement for each components: the Google APIs were shown as non-free software in an explicit way and the emulator images were somewhat shown as containing mainly free software. Now Google changed all this, and all the components show the same EULA terms. Now how can the user make any difference between what's free and what's not in that components list? Sounds harder than it used to be, and like a problem to us. That's why the Replicant SDK won't check for new components from Google. So here are the reasons why we call the Android SDK proprietary and why we think that there is a problem with it. Even though not all of this is a sudden change, why would it be any less relevant to try and raise public awareness about the issues we've spotted?
2013-01-06 Update: I've checked the license of the individual software components shipped with the Android SDK and it turns out that all of them are covered by a free software license. What's the point of that overall proprietary license then?
I have been contributing to OpenStreetMap since years now, and only rarely wrote about it, while there is actually a lot to say about the whole experience. I'll be presenting my feedback over the project in a series of articles, starting with this one.
So first thing's first: let's talk a bit about why I care about the OpenStreetMap project. You may have understood from this blog that I am a computer programmer involved in free software development. Though, the reasons behind my involvement in OSM aren't quite exactly the same as the ones that explain why I care about free software. While free software is mostly about giving back to the user control over their computing, OSM is related to data, not software.
Both software and data have legal owners, who can restrict the use that is made of them. In the case of software, that restriction can be made to a point where it becomes obvious that the user doesn't get enough rights to still control the computing they do with such software: we call that lacking freedom. That is why the free software movement was started years ago, coming with a solution in the form of legal licenses assuring the user basic freedom, denying unjust powers to the software copyright holder. That's the main reason why I believe free software matters.
Though, when it comes to data, the question doesn't remain quite the same. First, data released with a permissive license or with a strict license will look the very same, and the license can change without any change made to the data itself. This is not the case for software: a permissive license is pretty much useless if no source code is ever released: that means free software needs to be released with source code, while a music will sound just the same whether it's under a permissive license or not. Another point is that software can harm, data usually can't. Computers make it incredibly easy to spy on people, track them, read their online conversations, etc. That's why it is so obvious non-free software is bad: with such issues at stake, you'd better be in control of your computing. Data is instead pretty much harmless. So it seems clear to me that what is at stake for software makes the need of free software very obvious while for data, the issues aren't as serious and the experience you get from the data will remain the same.
Well, strictly the same isn't quite true. There must be something more that comes with content under a permissive license. Of course, that's the ability to edit, remix, reuse and share the content. Now I think it's all about a personal consideration: whether you think it is ok to forbid the user from doing these things. I personally believe that such rights should be guaranteed and the fact that guaranteeing them would perhaps harm sales rate isn't a good enough justification to refuse them, compared to the benefits of sharing culture among people. And yes, we can find other ways to let the authors of the content live of what they're doing, it just needs to be fair enough between the authors and the public. So this is why I believe data, content should be licensed under permissive licenses. I won't get too much into details there, about under what conditions exactly I find the bargain fair, but it's worth writing about. Also, note that even though I disapprove the licensing of content under restrictive licenses, I do not see any reason to refuse using it under the given terms since, as I wrote before, it seems to do no harm and basically, using it with the involved restrictions seems better than doing nothing. With these elements cleared up, we can get to why I decided to get involved in OpenStreetMap.
Maps are just another type of data, licensed under the same rules as any other content. As a matter of fact, all of the serious maps providers were licensing their maps under restrictive licenses, which I, just like the people behind OSM, disapprove. I could have just decided to pick one and use it anyway, but OpenStreetMap was there. As its name suggests, OpenStreetMap provides maps under a permissive license, that used to be the Creative Commons BY-SA license and recently changed to the Open Database License (ODbL). It's also doing more than just providing content: it is also a community effort, that lets everyone contribute to the map! At some point, I got a Neo Freerunner, with GPS capability, so I just told myself: hey, why not try and contribute to the effort, I have everything that is needed for that (which is basically only a GPS).