And conductive heat transfer is what’s being measured in the context of this article.

There was a framework I started using because it seemed interesting; but I still used several others.

It aligned well enough with the way I think about software, and I liked its ecosystem and maintenance approach well enough that I started choosing it by default for new projects unless there was a good reason not to.

I was hired for a couple roles where they wanted familiarity with it, and ended up having reason to look more and more under the hood to understand its internals when debugging. Not to mention learning how to and how not to scale it in terms of both traffic and development.

Now I find myself an expert in it and at an advantage when the need is for someone more than just conversant in it. And I still like it enough to look forward to working with it every day.

And I maintained enough broad knowledge and familiarity with other aspects of development not to be a one-trick pony. Which is important for staff and principal level engineering roles.

Another direction and one I’m working on establishing now is to choose a vertical to become well versed in. There are a few — especially in highly regulated environments like fintech or health tech; or where it takes a lot to ramp up on the subject matter, particularly in other science/engineering fields like aerospace engineering, biotech, and the like — where having previous experience in that vertical will give you not just a 10-25% advantage over a candidate who doesn’t (all else being equal) like in your bog standard app startup, but more like a 5x or 10x advantage. It can be tough to get your foot in the door; but once you do that door becomes a bit of a moat.

Oh, never mind, we already know the answer to that…

That said, they’re generally not early stage startups — more typically megacorps whose owners or lobbying arms have sufficient wealth/power/influence to make a meaningful impact that’s harming our economy, democracy, or social order. Sometimes a scale-up whose externalities are particularly destructive.

But under an authoritarian regime, it doesn’t matter whether what the state does runs contrary to statute or constitution because there is no one who has both the will and the ability to enforce any restrictions against the state.

ETA: look at SQLite for an example — it’s a relatively recent and simple entrant in the field and the closest you’ll find in the mainstream to a purely filesystem based RDBMS. How would you provide a stdlib that would let you implement something like that reasonably simply? What would be the use case for it?

There’s really no motivation for China to attack the US (they have plenty of economic leverage and that would be far less costly to them than an invasion, and what would they even want with the territory of a conquered US?) but I suspect many of the reasons the US is assumed to be immune to invasion are far weaker in the face of an enemy with far more manpower and a near-bottomless supply of drones that also has the capacity to cut off our access to key parts, materials, and manufactured goods.