Calculator is a pretty widely used app, too, even by Google engineers. But that hardly qualifies the author to work at Google.

Similarly while Homebrew is widely used and fits a need, it doesn't mean that the author is equipped with the skills that Google wants. A package manager and something like Google Brain don't exactly share much in common. It doesn't mean Google's script failed nor does it mean he shouldn't be pissed off, it's just how things are sometimes.

So what if random Googlers could also have written it? They didn't.

I talk to people about interview stuff, I notice that most people who say "I shouldn't have to know X" seem to be the least intellectually curious, least great problem solvers and just generally combative folks. If you are going into an interview with a serious chip on your shoulder, demanding a job, asking to exempt yourself from having to show problem solving skills, then just how is that going to work out?

Regarding reversing a binary tree, yeah, the interviewer didn't expect you to know it. That's the point. It's something that's a little tricky, not impossible, requires you to be precise about memory ownership and pointers, and algorithmic. It takes no time at all to describe, and isn't impossible to solve. Yeah it's problem solving, but it's not random trivia. It's a functional test of your intelligence and yeah it filters out some who should be at Google, and they know it. However relaxing the bar is fraught with peril. You're quite literally banking the future of the company on it.

It's my opinion that wasn't an unreasonable assessment (from mxcl). Although if indeed nearly nobody at Google uses Homebrew, as now multiple people are claiming, it was, no matter how reasonable, wrong.

I think he saw the question, and basically just refused to grapple with it, thinking it was below him, and then the rest pretty much was inevitable.

The rest of your post is speculation.

No, you actually never earn that right. Seriously.

In fact, it's a sign of good engineers that they just view their stuff as stuff, and not "well, i am doing the super important stuff, everyone else is not".

Note that this is different than prioritizing, etc. If you have a bunch of high level engineers in a room, and they don't value each other's work (IE recognize that a lot of stuff is "important", and there stuff is just part of that) because they think the stuff they do is super-important, that's going to go very badly.

At lower levels, pretty much none of it is so important that there is any value to having an inflated opinion of self importance.

The fact that you've achieved adoption does not necessarily make you a good engineer.

Straw man argument. Nobody stated this.

There's also a difference between thinking that your stuff is super important in isolation versus simply pointing out it's in wide use at a company and hence your previous efforts have in fact been productive and useful for them.

This is pretty much exactly his attitude, even if he did not explicitly add the other 3 words.

"in wide use at a company and hence your previous efforts have in fact been productive and useful for them."

Except that assumes it's literally true. It's not, as a downvoted person said :)

It also assumes that him working on it would be useful or productive relative to doing something else. That is also probably a negative.

Why?

To quote one of the demotivational posters, "just because you are necessary doesn't mean you are important".

https://news.ycombinator.com/item?id=12662182

Cogs in a machine?

https://cdn.shopify.com/s/files/1/0535/6917/products/worthde...

He may be important as a person who has developed software used by a variety of people (including many at Google), but that doesn't mean he is someone who ought to be important for Google to employ.

The number of googlers with Macs is not even 90% so this line of argument is just silly.

Basically the internal dev environment is already fully complete and works great. It's ultra custom and there are strict rules about bringing in open source code. The AGPL and GPL3 are good, but also one has to be wary about them in a commercial environment.