thelittlenag

First, since data centers likely need some kind of treated (i.e. demineralized) water, and they want to use local municipal supplies, then they can do so only if they pay full market rate for that water. I would push for them to pay 1% to 5% above market rate to help offset other costs, but market rate at minimum. At no point should homeowners and/or residential customers subsidize this usage.

Second, commercial users (including data centers) should only get water produced by treatment plants in excess of residential needs. Plants can only treat so much, pipes can only carry so much, and priority goes to residential first.

Third, ground water is off limits for any commercial/industrial evaporative cooling uses. If data centers want to cool that way, then they need a different source of water.

Did I miss anything?

This feels both like a good write-up, but an argument for why you often need to reach for multiple solutions when creating distributed system. Sure a fallback can have issues, but in the example given (website -> cache -> db) I can't help but feel that a circuit breaker pattern should also have been in play to catch other types of failures.

Distributed systems are dynamic systems and rarely will one type of solution or pattern cover the entire space of possible failures.

    Location: Minneapolis, Minnesota, United States
    Remote: Yes (10+ years working remote) (or hybrid/on-site in Minneapolis or St Paul)
    Willing to relocate: No
    Technologies: Scala, Java, Typescript (Effect TS), Python, C/C++, PostgreSQL, AWS, HLS, Real-time ABR
    Resume: https://drive.google.com/file/d/19O9A3zMulkPWb61Ng9FsSZbCPHZms_Qy/view?usp=sharing
    Email: [email protected]

But there are some obvious follow up questions that I do think need better answers:

Why is recovery made so hard in so many languages?

Error recovery really feels like an afterthought. Sometimes that's acceptable, what with "scripting" languages, but the poor ergonomics and design of recovery systems is just a baffling omission. We deserve better options for this type of control flow.

Also, why do so many languages make it so hard to enumerate the possible outcomes of a computation?

Java tried to ensure every method would have in its signature how it could either succeed or fail. That went so poorly we simply put everything under RuntimeException and gave up. Yet resilient production grade software still needs to know how things can fail, and which failures indicate a recoverable situation vs a process crash+restart.

Languages seem to want to treat all failures as categorically similar, yet they clearly are not. Recovery/retry, logging, and accumulation all appear in the code paths production code needs to express when errors occur.

Following programming language development the only major advancements I've noticed myself have been the push to put more of the outcomes into the values of a computation and then further use a type system to constrain those values. That has helped with the enumeration aspect, leaving exceptions to mainly just crash a system.

The other advancement has been in Algebraic Effects. I feel like this is the first real advancement I've observed. Yet this feature is decried as too academic and/or complex. Yes, error handling is complex and writing crappy software is easy.

Maybe AI will help us get past the crabs at the bottom of the bucket called error handling.

Their major complaint of the project approach is not getting signal on adaptability to new codebases. That has never been a first concern at any company I've worked at, and frankly if engineers are touching a new codebase every month then I'm getting a bit worried.

The goal here is to see if the candidate understands the domain (generic distributed systems) well enough on their own. For more senior roles I look to make sure they can then communicate that understanding to a team, and then drive consensus around some approach.