HELLO

Okay, I’m being honest with you, I’m so tired of my claude or whatever the kimi deep alibaba agent tries to help me building my projects. It never does. I’m coming back to old ideas, with code quality, that our ancestors, scrum-gurus and whatever nonsense left us with, I believe that cognitive complexion just makes the things unnecessarely harder to understand. Here is why slop.urm8.org/mcp[sse], add it to your claude/copilot: The thing is, you shall always take track of how much of LOC you are producing, the lower the LOC the more readable it is, the more productive you are at your codebase, I still can’t understand why so much people can’t get it. The cognitive complexity is the tuffest part, take care of it, the composition and the overall sequence of the functions you write are always make sense. The code is a novel that you should be able to read, not a fucking sudoku, it is a human/human/human/…/human/cpu/network/ipc/…/user communication.
The code must to be simple, ofc it must solve complex problems, so we have abstractions, but the closer the abstraction to the intimate implementation - the better. The code must follow the concrete order, like in a book or a poem, top-down is the only way for us humans as well as for agents… The code mustn’t have any unnecessary comments, when you write a comment it means that you wrote some bullshit. The code must be extendable without loosing readability, pls, don’t add 1000 arguments to a function that has to do one thing. The code shall be portable, it should never rely on language to language quirks bangs and whistles, we all are people in here to not deal with that BS. Shortcuts are good. Changing the shortcuts also considered good. The dumber and the more obvious the solution the better, only complexity coming from not enough abstractions or not enough effort, in the end each solution is retardly stupid.

FINALLY

OUR TASK IS TO MAKE THE THINGS WORK, NOBODY CARES IF IT IS LOOKING BEAUTIFUL OR NOT BUT FOLLOWING SOME RULES MAKES IT LAST