Aaron and Adrian have kicked off an interesting discussion about value on Planet KDE. Having an MA in philosophy, I thought I'd follow up on Adrian's quandary, that both Aaron and Andrew Cowie are right about value in software.

To begin with, a very simple answer to the quandary is that they are discussing different conceptions of value. Andrew Cowie is interested in the exchange value of his code, which is a feature of market economies and dominates the developed world in late capitalism. Aaron is more interested in use value, which preoccupies many Marxian, anarchist and liberal thinkers who reject the supremacy of the market. An object can have both exchange and use value, as well as other kinds of value, and sometimes those considerations coincide. So, for example, useful software is more likely to sell than useless software.

The problem is that life is never as simple as many free market economists would have us believe. There is a connection between exchange and use value, but it's not a direct correlation. To begin with, both the creator, passive users and fellow programmers will have different ways of judging the use value of the software. The KDE4 snapshots are of no use to me, but of significant use to Aaron and Adrian. But the dynamics of free software communities mean that the snapshots probably don't have much exchange value for Aaron or Adrian - that is they wouldn't pay for it - and neither would minor bug fixes, even if they have considerable use value.

Things are further complicated by the fact that, especially in free software communities, we are interested in political and moral considerations, not just economic or technical.

Thomas Scanlon developed an interesting theory of value, saying that it is underpinned by reasons (for philosophers, normative considerations are prior to the domain of intrinsic value). Scanlon defines a reason as "a consideration that counts in favour of some attitude or action" (other philosophers might lend more weight to moral considerations, e.g. John Broome says that a perfect reason is "an explanation of why you ought to" do something). We can then argue that things or actions have value because we have good reasons for them. This may sound pretty obvious, but it rejects a common belief that things may be inherently valuable.

Bastardising his work, then, we can say simply that Aaron and Andrew have different reasons that count in favour of sharing or witholding code. They aren't just different in the sense of different business models, but also in that Aaron brings in wider ethical, political and technical values, themselves based on a complex and (one would hope coherent) set of reasons.

A common challenge for free software advocates is in ensuring that exchange value coincides with our ethical, political and technical values. That is, can we still put food on plates and roofs over heads whilst sharing and developing code in open communities?

Further reading: Reasons, values and agent-relativity, by R. Jay Wallace.