Marco RogersSep 17, 2024
Okay. I don't really wanna start a fight today. But at some point we're gonna talk about how the last 10 years has seen software developers pushing for premature promotion into Senior or Staff titles. And one of the ways they try to demonstrate that they deserve these promotions is by doing unnecessary "Arkiteksure".
https://phpc.social/@afilina/113153763726491625
🇺🇦 Anna Filina (@[email protected])

Attached: 1 image Almost every day, I see microservices that should have definitely not been microservices. Almost every time, it's a big bad of mud with extra complexity. You can't just split things into random smaller things, slap HTTP in between and call it architecture. In almost every case, cleaning up a monolith would have yielded better outcomes.

PHP Community on Mastodon
Show threadMarco RogersSep 17, 2024

There are a few spicy things here. So I want to unpack a few things just so folks don't misunderstand me. (You can feel free to be mad as long as it's about the right thing)

First, I said "premature promotion". I know a lot of people are going to wonder what that means. I'll try to state it clearly and directly. Before the bottom fell out of the market, devs could make a lot of money by getting raises and promotions. Part of the way to get that was by raising your level by any means necessary.

Show threadMarco RogersSep 17, 2024
I believe we have been seeing an overinflation of levels/titles for a while now. This is mostly because devs were in high demand. They agitate to be "Senior" after only a few years of working. For better or worse, many companies have been making this concession in order to retain people. (whether it actually worked is another convo). But the knock on effects have been severe. "Staff" title has come into vogue as the new "Senior". And the expectations of Senior have been greatly decreased.
Show threadMarco RogersSep 17, 2024
The word "premature" is meant to convey that many devs are reaching higher levels without the actual experience and expertise necessary to inhabit those levels effectively. We see the outcomes in so many ways. Slow bloated apps are pervasive; Data breaches have become normalized; Interviews are centered around specific frameworks rather than foundational knowledge. All of this come from similar root causes in my opinion. Our industry grew too big way too fast. And we didn't train anybody.
Show threadAndrei KucharavySep 17, 2024
@polotek a point Alan Kay made a couple of decades ago as well.
Show threadMarco RogersSep 17, 2024

@andrei_chiffa Yeah I mean this is the elitist version of this take. But I don't really subscribe to it. The fact is that Shakespeare is overrated and shouldn't be taking up so much cultural room. In the same way, I think relaxing the credentialism around "computer science" has allowed us to create a much wider range of value from software.

But just like tv/movies, the level of quality is going to vary greatly. And we need to decide how to contend with that reality.

Show threadAndrei Kucharavy

@polotek oh yeah, totally. I think the “you need to be good enough for it” attitude was what killed Smalltalk in the same way it is killing Haskell today. And if anything I am looking forwards to coding LLMs allowing even more perspectives to come into the software.

But your original comment definitely brought this quote to my mind.

Sep 17, 2024 at 8:24pmToot!