In association with heise online

Attribution and reputation

I agree that attribution is crucially important – indeed, I think it's actually the key issue that needs resolving in terms of digital materials being shared online, since deriving economic benefit depends on it. But enforcing attribution through licences may not be the best way to do that, as Asay explains:

In important respects, the IP approach to open models of innovation [that is, traditional open source licensing] fails to fulfill the roles assigned to it. For instance, with respect to attribution-only licenses, in most cases the attribution is buried somewhere in legal documentation so that any recognition that may accompany such attribution is minimal at best. This reality suggests that those who contribute under attribution-only licenses, while perhaps motivated by some form of recognition, in most cases are likely motivated by a different type of recognition than what the IP approach provides. In the FOSS world, tools such as GitHub, a widely used social coding tool, might better provide the recognition programmers seek. The fact that more and more software is contributed via GitHub without IP notices or license information at all suggests that the “prize” of an IP notice in obscure legal documentation is not much of a prize at all, at least to those contributing.

Indeed, people and companies have already started using things like GitHub profiles as a way of respectively displaying and gauging coding ability. I suspect this will become the norm soon, with more formalised ways of establishing reputation in the coding world appearing alongside the informal ones dictated by social norms within the free software community.

Having tackled the two key objections to moving on from traditional open source licensing, Asay then looks at what the public domain would mean in practice:

A PD approach, therefore, would need to effectively override any automatic copyright rights, waive any patent rights (both with respect to any patent rights already obtained as well as prospectively), and relinquish any remedies that come with either. Trade secret rights, if any, would be relinquished as soon as the rights holder released the software or content to the public. Arguably waiving any trademark rights is not only unnecessary but inadvisable, since others could then use the marks to confuse consumers as to the source of the software or content. Indeed, this is precisely why CC [Creative Commons], which includes a PD dedication tool in its repertoire of legal documents, expressly exempts trademark rights in the tool.

That last point about trademarks is important. Although it may seem strange to argue that of all the intellectual monopolies, trademarks should be retained, that's because they serve a very different purpose from copyright or patents. Trademarks are designed to protect consumers from fraud, rather than seeking to exclude competitors, even though that's how they are often used today.

But for open source projects, trademarks are purely about reputation – that is, they become guarantees of quality when applied to a program. Anyone may take that code and use and adapt it in any way, but they may not apply the project's trademark, since that would imply that it was the official, "approved" version, which is obviously problematic for a variety of reasons, not least security.

Asay also discusses an important objection to his thesis that placing it in the public domain would be a better way to distribute free software: if that's the case, why does everyone stick with the GPL or Apache? As he points out:

In the FOSS world, for instance, there is no recognized or widely used PD dedication tool. Instead, the Open Source Initiative and the Free Software Foundation — the two leading FOSS advocacy organizations in the world — vet and approve open licenses for use in the community. While it is true that various projects could simply ignore these recommended licenses and adopt a PD approach— and some have attempted to do precisely that — that sentiment assumes that the organizers of such projects understand how to do so.

That is, it is currently quite difficult to place software in the public domain – first, because there is a cultural bias against doing so, even within the free software community, and secondly, because legally it is an awkward process. Indeed, Asay suggests we need new legislation – what he calls the "PD Act" – to make it easier. This would obviously need to take into account the different copyright systems around the world – for example, those that place more emphasis on "moral rights".

A few years back, I asked Richard Stallman for his views on how copyright should be reformed, particularly for software. Here's what he replied:

For most kinds of works, I think copyright would be acceptable if we (1) made it shorter (my suggestion is 10 years), (2) permit noncommercial redistribution of verbatim copies, and (3) define transformative "remix" uses clearly as fair use.

However, I think software and other works of practical use must be free.

Note that he, too, thought that software should be free of copyright – in other words, in the public domain – but he added a few caveats:

I would be glad to see the abolition of copyright on software if it were done in such a way as to ensure that software is free. After all, the point of copyleft is to achieve that goal for derivatives of certain programs. If all software were free, copyleft would not be needed for software.

However, abolishing copyright could also be done in a misguided way that would have no effect on typical proprietary software (which is restricted by EULAs and source code secrecy rather than copyright), and only undermines the practice of copyleft. Naturally I would be against that.

Putting free software into the public domain would be equivalent to abolishing copyright for those programs but leaving proprietary code unaffected. But would that be such a problem? Personally, I don't think so, for the reasons that I have mentioned: any company that takes PD code and encloses it loses all the benefits of openness. It's true that there remain legacy programs from old-style software houses that have always been closed, but their existence doesn't really affect the larger free software world, which has now grown to independence and self-sufficiency. What Microsoft and its ilk do is pretty irrelevant these days.

Of course, moving to PD wouldn't mean that today's free software licences disappear – they will still be there for those who wish to use them. As ever, choice and personal freedom are crucial. But I hope that people will think twice about introducing new licences, or even updating old ones. In particular, I hope that there will never be a GNU GPL version 4. Instead, we need to complete the revolution that Richard Stallman began nearly three decades ago by making free software truly free, placing it in the public domain, and severing the chains that still bind it to that three-hundred-year-old monopoly called copyright.

Follow me @glynmoody on Twitter or, and on Google+. For other feature articles by Glyn Moody, please see the archive.

Print Version | Permalink:
  • Twitter
  • Facebook
  • submit to slashdot
  • StumbleUpon
  • submit to reddit

  • July's Community Calendar

The H Open

The H Security

The H Developer

The H Internet Toolkit