Don't Let Architecture Astronauts Scare You

It’s NOT just an If statement, because Bob our architect decided 10 years ago that it would be fun to have all our logic in database views and procedures. Now Bob had decided it’s not secure for people to change the database without talking to him first. Bob’s a “genius”

There is also a related category of Architects — "Pigeon Architects". Why Pigeons ? Because they fly into your project, do sh!t and fly out.

Dude, this speaks to me. In a previous job I had a guy so fixated on make the DB so intricated he spent over a year just designing the schema… Also we as a company were only bringing in roughly 5m a year gross. There is a reason why it was a previous company.

11:50 Grand Eloquence. It's two words. Don't let literary astronauts scare you by combining words to sound smarter.

We have those in the real world too! I worked for a company which patented it’s own robot architecture. It only had a drawback, it had no “inverse kinematic” (the equation that tells you if you want the hand to be at (x,y,z), you need your motors at angles (a,b,c)), so we had to simulate the whole robot and run a search algorithm to find the closest point we could reach. We substituted a O(1) process for an NP hard problem just so our architect could claim he invented something.

“Don’t abstract at all, until it’s painful not to abstract” love it

Reading Joel on Software is what transformed me from an electrical engineer who writes code into a software engineer.

My favorite Joel Spolsky quote: It’s harder to read code than to write it. I’ve dealt with so many turd burglar engineers who think everything needs to be rewritten. Of course they mean that THEY need to rewrite it. Those are Architecture Astronauts in training.

"Remember that the architecture people are solving problems that they think they can solve, not problems which are _useful_ to solve." Idk if I feel seen or if I should nervously chuckle and reassure myself that my many side-projects are actually somewhat useful.

"All they think about is peer-to-peer this, that, and the other thing..." And my mind went: blockchain, microservices, LLMs. Seriously, our industry has a horrible pattern of discovering (or rediscovering) some idea and then, rather than thinking "this is an interesting solution to a problem," we jump to "this will solve EVERY technological problem that we can think of, EVER."