Interesting thread. Took a while to get around to the point, but it did, eventually, get there.
> Polycultures are important. People have different skills, perspectives, and connections. They have different institutional incentives and roles, and collaborate with others in diverse ways.
I’ve always found this to be true. I’ve spent my entire life, in fairly “heterogeneous” cultures.
Mixed teams can be difficult to manage (which is a big reason for today’s insistence on monoculture). Weak management tends to result in chaos, but a well-managed mixed team can do amazing things.
> Specialization is productive and efficient [...]
... When the solution fits inside the lines of a particular specialty; otherwise, you need a different specialist, and knowing which one is not at all trivial. From time to time, the solution doesn’t fit within any established lines, in which case you’ll never find it at all this way.
Which is why the commonly expressed sentiment that domain experts should stick to their domains of expertise drives me up the wall. Especially the experts in question are scientists and the specific lines can be ... less than well established, to put it mildly.
That is not to say there aren’t plenty of examples of people being wrong outside their domain, but that’s not even a problem. Being confidently wrong would be one. Unfortunately, that’s not how the incentives are set in most places.
The fact is we need both. A team of generalists and specialists should be able to achieve far more than one or another. Specialists have deep domain knowledge and experience, but at the same time they are "standing too close to the elephant". Generalists, especially those that have shallower experience but over a wide range of different domains can provide a framework of analogies that can bridge the gap between domains.
Agreed, and that was the sentiment I had in mind (among many other contexts) in writing the thread. Monocultures can be "productive" in a narrow sense but then are less adaptable and do not cope with unknown unknowns particularly well.
> Polycultures are important. People have different skills, perspectives, and connections. They have different institutional incentives and roles, and collaborate with others in diverse ways.
I’ve always found this to be true. I’ve spent my entire life, in fairly “heterogeneous” cultures.
Mixed teams can be difficult to manage (which is a big reason for today’s insistence on monoculture). Weak management tends to result in chaos, but a well-managed mixed team can do amazing things.