Both can be true: - It's useful to anthropomorphize agents when predicting behavior and - we have to use specific language to specify what we mean.

What does the author mean by "confuse the models" ? Are they talking about not picking right information? Picking the wrong information? Losing their previous context / task?

Part of setting up a proper eval is also deciding what we actually mean ourself. What are we actually optimizing for? It's not, e.g. % confusion, %rubbish, etc.

The article does point to it: retrieval latency, accuracy, etc.

The incentives you're talking about -- they're missing because of NIMBYist overregulation. The whole point of NIMBYism is to use regulation to hamstring the positive incentives in the market. "There's demand for twenty units here but the place is zoned for a single unit." or "There's demand for twenty units but the city demands that if we build a multitenant unit, we have to do a twenty-year environmental survey first".

Do you live in a place with a homeless crisis. Guess what: You're a citizen and you have some agency. Democracy can be a backstop to "pure" (or mis-regulated) market forces. I, for one, enjoy clean drinking water (and also: a good deal from a healthy competitive market).

Or: maybe there is no large cost trade-off. Do beautiful manhole covers seen all over the world significantly affect the total cost of ownership?

Fwiw the same thing happened at my last startup. We did a search for a VP eventually promoted from within. I hereby claim, that statistically speaking, it definitely happens sometimes.

I guess the author disagrees with your last statement, > If you start at the bottom then you remain there because your skills are not valuable in the top leadership roles.

She says that she had the space to think more about strategy because the "people at the bottom" were doing an good job of keeping the company's infrastructure stable as they scaled. Maybe it's less about top and bottom and more about what types of problems people are good at solving. My tip: if you like planning and management and strategy, you should try to get roles, in the top, bottom, and middle using those skillsets. Lots of even introductory roles involve "manager", and lots of non-management roles are a great career path for many.

That being said, the task of debugging a deprecated tool version in asdf just one time makes the investment in the nix awkwardness worth it.

the pitch for nix isn't that it's a nicer tool than, e.g. `asdf`, but rather, that it will probably work in five years. frightful flashbacks of of trying to get nokogiri to build.

`make_pizza(["pepperoni"])`

What does `make_pizza()` do? It could be a lot or it could be a little. It could have side-effects or not. Now I have to read another function to understand it, rather than easily skimming the ~four lines of code that I would have to repeat.

I think the article fails to show particularly problematic examples of DRY. E.g. merging two ~similar functions and adding a conditional for the non-shared codepaths. shudders

Without a legal system, e.g. crypto, if you solve my puzzle, then you deserve the reward. It's just math!

There's something fundamentally undemocratic (and perhaps, sadly, realistic) about this take. Will advocating for broad de-regulation and business rule actually reduce cronism? Will it / can it reduce negative externalities?

I just think ... we've tried that. It's been the experiment of the last 40 years. What's next?

In the end, we hire interesting, curious devs, and ask them to use tooling (elm) that has a pretty shallow learning curve.

I think bootstrapping a project in Elm without any experience on the team could be tough, but once you have some expertise, the seeds you plant in yield abundant fruit. At NoRedInk it's been a great success -- especially with regards to "finding other developers".