... i always thought Linux was opensource.
It is a remarkably common misconception that open-source is antithetical to copyright. It's not true, even though I have read articles, in apparently authoritative magazines intended for a business readership, that make this assumption. Yet a few minutes reading the major opensource license, the
GPL, is sufficient to dispel the notion!
Perhaps the explanation is that people who support the GPL and software freedom tend to be acutely aware of the hazards of copyright, and the abuses by corporate interests that are such a prominent part of contemporary cultural life. There is money in ringfencing ideas and expressions for one's own benefit, and to the detriment of the wider society, most especially if the state can be persuaded to pay its officials and police to enforce this!
Apart from that, an uncritical acceptance of the notion of copyright can lead to absurd and even dangerous situations. For example, the dangerous cult of Scientology has aggressively used copyright legislation to prevent people finding out just what it's really about.
If there were no state interference in the free transmission and use of ideas and expressions, then the opensource movement would still be able to function, it's true (although Scientology might not, for it would find itself widely, perhaps fatally, exposed to ridicule). People would still be free to co-operate with each other, and to share code. But the code they published would likely be taken and used in products and programs which kept their workings hidden from their users. Copyright allows us to prevent our co-operative efforts being privatised in that way, and kept open for the benefit of everyone, if we so choose.
Not all opensource coders make this choice -- in part it depends on one's purposes. A remarkably good way to explain a protocol, and to have it coded correctly by everyone, is simply to publish source code that expresses the intended protocol exactly, and allow others to use it without any restriction at all. Then people and corporates who want to publish their programs in binary form only can just take that source code, and incorporate it into their own source code. The compiled program only (not the source) can be distributed or sold, and although its working are opaque, one can be reasonably certain that the protocols are executed correctly. There's a fair bit of this sort of "opensource" code in Windows, as it happens (although one cannot demand the source to the MS program, one is of course still free to use the original code that MS has incorporated).
