This puzzle - the apparent conflict between the truth-seeking understanding of science (e.g. Popper), and the sociology approach (which as I understand it doesn't predict that scientists do find even approximations of the truth) - is very interesting.

A philosopher that I spoke with at the Singularity Summit made the observation that everyone there seemed to be familar and comfortable with Popper but there were almost no mentions of Kuhn. My understanding of why the scientists and technologists at the summit didn't mention Kuhn is that his theory of how science works isn't (obviously) usable. There will be paradigms, and paradigm shift - but as a practictioner, what do you suggest I actually do?

Pickering's "Mangle" concept may be applicable. I've been trying to digest his "The Mangle of Practice" into a top-level LW post, but other priorities keep getting in the way.

Pickering is a sociologist (and therefore writes in a style that I find annoying and off-putting) but he includes as "actors" non-human entities (like microscopes or bubble chambers). This makes his theory less human-society-centric and more recognizable and sensical to a non-sociologist. Unlike Kuhn, I think Pickering's mangle might be able to be applied in improved methodology.

As a programmer, the best way I can explain the "Mangle" (Pickering's theory) is by reference to programming.

In trying to do something with a computer, you start with a goal, a desired "capture of non-human agency" - that is, something that you want the computer to do. You interact with the computer, alternating between human-acts-on-computer (edit) phases, and computer-acts-on-human (run) phases. In this process, the computer may display "resistances" and, as a consequence, you might change your goals. Not all things are possible or feasible, and one way that we discover impossibilities and infeasibilities is via these resistances. Pickering would say that your goals have been "mangled". Symmetrically, the computer program gets mangled by your agency (mangled into existence, even).

Pickering says that all of science and technology can be described by an actor network including both human and non-human components, mangling each other over time, and in his book he has some carefully-worked out examples (e.g. he applies his theory to Hamilton's invention of quaternions) which seem pretty convincing to me.