I am now blogging about design engineering in organizations that deal with lots of domain knowledge(s). This post covers metrics that don't do what your manager thinks they do, and why interdisciplinary work has trouble finding a good home. https://designengineers.substack.com/p/the-myopia-of-professionalism

I feel like I understand what you're saying for a second, but then I blink and it disappears. ---

I propose that experts supply and evaluate criticism, just as professionals supply and evaluate metrics.

Are professionals and experts different people here? Earlier in the article it seems like they're just synonyms. --- The part that seemed most fertile was the 3 pictures of domains between the two halves. --- I don't quite see how the two halves connect. The major connection I see is between translations between domains and boundary objects between teams. Is there more? Perhaps what's missing is an introduction or conclusion that lives outside the two halves. Or perhaps they should be two distinct posts?

Yes, that makes sense - thank you for unpacking the disconnect there. I completely skipped over the definition of a domain (following David Ribes' workplace ethnographies). That would set up the second half as a payoff to the first half, because the boundary object solves the problem where the domains hate to explain themselves to each other (because switching languages is really slow and frustrating). The role of the 'expert' is not grafted quite right into the world of the 'professional' from Stengers' essay. It should be 'public intellectuals', university-affiliated or otherwise. I expand on the picture of domains at three 'zoom levels' in the BELIV paper! Which is in a dense academic style unfortunately.

Like @Kartik Agaram, I am a bit confused about the roles. I see the issues you raise every day in academia, between scientists from different backgrounds and also between researchers and management. But if academics are "experts", then we have no "professionals" in this situation. I see researchers taking both roles, though not on the same subject.

Hmm, the tricky thing with researchers is that everyone I've met in academia wears at least four different hats. Multiple inheritance, y'know? Perhaps a set of 'partial roles' e.g. domain practitioner / literature expert / institutional administrator is called for. Absolutely, this means the incentive structures for each role are actually more complex than 'metrics vs. critique'. Otoh, incentives aren't necessarily blended between multiple roles, because people will take on jobs they weren't hired to do on paper. Therefore, within the context of how do I form a stable niche in my institution, it might actually become a simple taxonomy of roles again? Albeit, rather cynical.

Four hats may be true. That's actually an aspect of academia that I like. It's a more diverse environment than the typical business with its more sharply defined roles. And it provides more individual freedom. And yet, as I said, the core observation of your post applies to academia as well. Maybe the incentive structure matters less than we think? Or is more complex than often described?

Yes, absolutely. It needn't be a complex set of institutional incentives, even; but your response is not knowable to me, thus you can form a niche that I wouldn't be able to predict. See, if I ascribe someone a fixed utility function, then - first off, that says more about me than about them. But let's say the point of a given role is that I can predict that person's utility function, i.e. what kinds of resource and relationship I can offer in exchange for their labor. Now if you've got four hats, then I don't know what your switching condition is. Maybe you want to work from home because of family commitments; or conversely, maybe those are pushing you to the office. And all that gives you freedom from the simplifying gaze of post-Fordism and (ugh) scientific management, in terms of being able to defy its expectations. (Not as in being able to make money outside of it.) Now a really important corollary of the 'freedom of multiple hats' is the ability to switch between metrics, critique, and other evaluation strategies. That's going to make up the part of my tenure packet, or grant proposal, that I can actually design on the spot. It's critical that my strategy be legible to the manager, or the funding source, or whoever is evaluating that. And that's how I hit the impasse where some methods are respectable right now, and some other methods are actually productive for my questions.