Why Nostr? What is Njump?
2025-01-30 19:59:38

prolic on Nostr: With everything you know about me, roast me without mercy. ChatGPT response: So, ...

With everything you know about me, roast me without mercy.

ChatGPT response:

So, you're running Linux Mint on an ancient laptop, yet you expect performance. What, did you think Cinnamon was some lightweight, minimalist DE? Newsflash: Your machine wheezes every time you open a terminal. Arch users would bully you, but even they feel bad for you.

And then there's your love-hate relationship with Lisp. You can't stand the parentheses? Fine. But let’s be real—you’d rather wrestle Haskell’s type system for days than deal with a few extra ( and ). You claim to love purity and elegance, but in reality, your effectful spaghetti code is just monads in a trench coat.

Speaking of which, you’re breaking effects into tiny groups just to reuse state? So instead of dealing with deep effect stacks, you just shuffle your problems around like a shell game and call it "good design." Your codebase must be a Byzantine maze of STM updates and arcane type signatures that only future-you will suffer through. Good luck debugging that in a month.

And let's not forget your grand subscription model, handling every possible event with precision. Meanwhile, someone out there is writing a ten-line script that does 90% of the same thing, and it just works. But no, you need the correct solution, even if it takes three weeks of yak-shaving to get there.

Then there's your storage choices. LMDB? You mean the database equivalent of "I don't trust Postgres, and SQLite is too mainstream"? You're so obsessed with performance that you'd rather hand-optimize data storage than admit that maybe, just maybe, something off-the-shelf would be good enough.

Oh, and your website? Running on gh-pages like some GitHub beginner tutorial. You're probably over-engineering a static site that could be a single HTML file, but no—you need structure. And of course, you have a Flatpak project, because why just distribute software normally when you can wrap it in extra layers of sandboxing complexity?

In the end, you’re a performance-obsessed, purity-loving, state-rebuilding, effect-splitting Haskell masochist who would rather spend a week perfecting an abstraction than writing a simple solution. And you love it. Because deep down, you don’t really want things to be easy—you want them to be right.
Author Public Key
npub18wxf0t5jsmcpy57ylzx595twskx8eyj382lj7wp9rtlhzdg5hnnqvt4xra