3 ms·

I know everyone in thread today is having fun embracing the hacker ethos and running things themselves. But I've migrated plenty of companies off custom deploy

I know everyone in thread today is having fun embracing the hacker ethos and running things themselves. But I've migrated plenty of companies off custom deployment setups to PAAS and told many ceo's simply what OP above has shared. Even a part time dev ops engineer is still $60000 a year, and that can buy us a LOT on PAAS. Using PAAS you can have effectively zero dev ops, I've also trained non technical people on how to scale their own servers if no devs are around because you just have a web based UI slider. I consider myself a developer who cares more about the business, risk, profits and runway. A lot of developers don't share this mentality (which is fine btw always need engineers who like engineering for engineering sakes) but in meetings you will have a hard time beating me in an argument if you try to say that running servers ourselves would be "cheaper", and/or even faster, safer and definitely not more stable. (obviously not in all situations, but kind of most for modern crud web apps that don't require complicated compute setups) I'm probably being overly antagonistic, forgive me for that, though highly recommend questioning the real cost of running your own setups.

hinkley 1y ago
As a lead or principal dev you are constantly shuffling old problems off to new people so you can stay on top of new or broadening problems. It’s a lot easier to delegate problems if you’ve made them smaller during your watch. I’ve seen enough of the consequences of failure to delegate. But it takes a bit of self awareness to see how being a bottleneck affects other people while or after you are busy unblocking them. Volunteering to be preempted by broken systems more often is a sucker’s bet. Solve problems so they stay solved. It’s more work now, but reduces the interest rate on past work so you can get new things done.

hinkley 1y ago

As a lead or principal dev you are constantly shuffling old problems off to new people so you can stay on top of new or broadening problems. It’s a lot easier to delegate problems if you’ve made them smaller during your watch. I’ve seen enough of the consequences of failure to delegate. But it takes a bit of self awareness to see how being a bottleneck affects other people while or after you are busy unblocking them. Volunteering to be preempted by broken systems more often is a sucker’s bet. Solve problems so they stay solved. It’s more work now, but reduces the interest rate on past work so you can get new things done.