🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

So many good answers! For many of these, it took lots of years for me to see how these "best practices" are toxic. The rest are probably for me to learn in the next many years.

I was hoping there was a good answer. This shows how much value could be unlocked by creating the right tool. Imagine just the total amount of time and cognitive load saved by doing this right.

New name for ML engine? I mean AI Platform... I mean Vertex AI...

I want to be that. So far I'm mostly trying to do prevention: - This tool could help us with X, should we use it? - Probably we don't need it. - But it also has feature Y and Z. - Then definitely no.

This makes me think: why isn't there something like jOOQ for Python, Go, etc.? Ibis with generated code based on schema would be ace (even in a dynamically typed lang like Python).

I feel priviledged that I can install and use Linux on my laptop. Plus a custom rom on my phone and blocking trackers on web. I live in a bubble.

This is so true about software in general. Most of the time it's an imperative for me: do not bulid abstractions on complex systems. Very rarely you can make good abstractions. But you really need to know what you're doing. (The original tweet was about abstractions on Git.)

Why aren't these best practices highlighted in k8s docs? (Also, for memory it should be request=limit.) It took a while for me to figure them out and it's very difficult to find a use-case where it should not be this way.

About why we need to choose a Mastodon server when registering. This pretty much answers it to me. I can get behind this.

This is especially true in this age of more and more YAML. We do more wiring microservices et al. than coding (with exceptions, of course).

I need to love this! I have been waiting for something like this.

Lost my job today as Beat decided to shut down its business. Anyone hiring for senior data leadership roles (preferably remote from Germany) ?