In 10 years of software dev I saw apps rewritten from Angular to React, endless fights Vue vs Angular, and wars Java backend vs Node middleware.
But you know what? The client didn’t care. They just wanted business value.
❌ When tech becomes art for art
I was guilty of this too. As junior (and early mid), I didn’t care about domain at all. I could code anything really.
But the tech… oh yes, it had to be newest and coolest.
If a new state manager got popular, I wanted it in project next day. Because, hey, if everybody use it, they must be right!
Luckily my seniors cooled me down 😉 But some devs never grow out of it.
⚠️ Examples of pointless refactor
- Adding state manager when app is small and no problem with component communication
- Rewriting Angular → React just because syntax looks nicer
- Creating more abstraction layers / monorepo / packages without real need
That’s just burning client’s money.
✅ When refactor actually makes sense
Of course, refactor or migration is not always evil. It makes sense when:
- rewriting old React classes to hooks (yes, still happens 😅)
- adding/changing state manager when project grows and current solution is too weak
- adding new “toys” step by step, if team decide they help
A real example from my work
We had legacy frontend with no typing at all. Introducing TypeScript wasn’t about hype — it reduced bugs, improved onboarding for new devs and made codebase future-proof. That’s business value.
🎯 Final thought
Let’s leave fights about which framework shine the most to juniors and community stars.
We — let’s just focus on delivering business value.
Top comments (0)