just wrote a 3k word article on why this (really good, i mean really exceptionally awesome) serverless parody of Hamilton has a chorus that i hate a lot.
"i'm gonna reduce your ops..
yeah i'm gonna reduce your ops"
yo, **ops is not a synonym for toil**.
operational engineering is the constellation of all your organizational skills, habits, best practices, defaults, tools, and aspirations around delivering software to users swiftly, efficiently and humanely.
"less ops" isn't necessarily better ops, any more than fewer lines of code necessarily means better software.
in either case, you need ... as much engineering as it takes to do the job well, and not a whit more. why is this hard?
i have mad respect for the serverless folks. but they seem to persistently misunderstand the reason for their own success, which is weird and frustrating.
the value of serverless isn't found in "less ops". the value of serverless is unlocked by **clear and powerful abstractions** that let you delegate large portions of your infrastructure to be run and managed by other people who can do it better (and more cheaply) than you can
to be clear, there's just as much ops going on as before. possibly more.
just _not by you_.
this isn't a reason to shit on ops and call for #NoOps bullshit, it's a reason to reevaluate your lingering wobbles around outsourcing higher-level services and abstractions.
because yes, operations is a cost center. if your mission doesn't involve solving infrastructure category problems, operational costs are a thing to be managed for efficiency.
doing this well is extremely difficult and takes a lot of experience and expertise.
what's the difference between automatically shipping 10x/day, rarely getting paged, and swiftly restoring service when you do,
vs shipping 1x/week (by hand, consuming days of engineering-time), getting paged 20x/day, and always escalating to the same person in a panic?
(ops)
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Well, I for one am not past this bullshit by now. ☺️ EMs who do some hands on engineering are better EMs.
Forbidding EMs from touching code at all is almost as silly and counterproductive as telling EMs that writing and shipping code is a core function of their role.
I say "almost" because if I had to choose one or the other, I would choose the clarity of "EMs responsible for team outcomes, SWEs responsible for technical outcomes" over the muddle of holding EMs responsible for everything and splitting their focus between people and code.
But I don't have to choose! EMs who keep a hand in the code are better EMs. They have more empathy and understanding for their team. They are better equipped to evaluate their engineers, they have more credibility and context. Everyone wins.
I have a new piece up. It's a bit of a rant, even for me, so buckle in.
A lot of "thought leaders" have been making their mortgages lately off of bits on how AI is going to replace software engineers, particularly entry-level engineers.
This is a dumb idea. It bespeaks a wealth of misunderstanding about what it means to be an engineer and write code, and what is valuable and hard about software systems.
But even really dumb, damaging ideas can weasel into people's heads if you repeat them blindly enough times.
Generative AI has made it easier than ever to generate lots of code. @kentquirk says it's "like a junior engineer who types really fast". 🤣
But writing code has always been the easiest part of software engineering -- *always*. And it's getting easier by the day.
It felt, to me, like those participating were stepping very cautiously around a few of the third rails Jaana just tripped over. (💜)
"Work-life balance"
"Working hard vs working smart"
"Meritocracy"
The intersection of company tech cultures and expectations and performance.
These are hard, complicated topics, and there are some very good reasons for speaking carefully. People can pick up a sentence and run in the wrong direction with it, and do a lot of damage.
I have abandoned god only knows how many drafts on this topic, for that reason.
The question is, how can you interview and screen for engineers who care about the business and want to help build it, engineers who respect sales, marketing and other functions as their peers and equals?
It's a great question!! I have ideas, but would love to hear from others.
I said "question", but there are actually two: 1) how to hire engineers who are motivated by solving business problems and 2) aren't engineering supremacists.