Комментарии:
That's your projects, apparenty
ОтветитьThe correct way to go about this for future proofing is properly modulating.
If you are not modulating your creating the biggest run on possible.
Just make modules of your code.
Break the code down for easy debugging and reuse purposes.
Not rocket science 😅
Yeah. Truth.
Ответитьsame, lucky I refactored all before too late
ОтветитьBuilding the foundation is hard thoughhhhh and then you get your footing and have a codebase to work off of and life is not burning around you anymore
ОтветитьBuilding the foundation is hard thoughhhhh and then you get your footing and have a codebase to work off of and life is not burning around you anymore
Ответитьoh hey cool inverted whiteboard idea, where did you come up with this
ОтветитьThe accuracy wtf 😂😂😂😂😂😂
Ответитьyeah, I'm awful at code, I'll screw up less than a hundred lines
ОтветитьNot for me
ОтветитьWow that makes me feel good. If i do something well with my projects, is thinking ahead and prepare for scale, code and deployment wise.
ОтветитьWhy is he right 😭
Ответитьsmart guy, that's why most of my projects are under 2,000 lines of code.
ОтветитьDude. Tradeoffs and refactoring?
ОтветитьOh no i startet regretting at 4000 lines already.
Then again had i used c++ those 4000 lines would’ve been more closer to 8000 lines or more.
I think I would regretted more if I just vibe coded.
ОтветитьI have 12000 lines in one file. I don’t regret it, I am the regret
ОтветитьNope I found the magic key to getting around that little piece of dread,
I'll share just because I want to make the world a better place.
Make your system modular so that way the initial core is always able to run on its own and everything else you add to your program will be like clothes - shoes, hats, pants, socks.........
You can take off any one of these things but it won't break you.
Not if you're using functional statically typed language
Ответитьthis is why u start trying to think about minimising self input instead of minimising lines, such as python, use for loop for checking if something is valid, if ur using if statements for lets say a clicker game and u have 100 upgrades
use a dictionary, use a for loop to check
you dont want a if statement for every upgrade
4 lines vs 300 lines
which one u want
yet everyone is still using Python and Javascript and PHP and Go and C++ and Java where these issues are arguably the worst 😢
ОтветитьI leave the first 2k lines for later and start with 2k+1
ОтветитьFunny 'cos in 35 years of work I only ever write up to about 2,500 lines on average per app/util. I never write more as I just find that seems to work for me. If I think I need more then I just code up modules and pull 'em in. 2,500 is where I start to loose track of the code and it gets too big to manage in my head, so I break it up and buidl re-usable mods.
Ответитьhe a n008
ОтветитьWhen code starts to matter less, but architecture design takes the role.
ОтветитьI hate how true this is
ОтветитьI can feel that bc :(
ОтветитьUNBELIEVABLY REAL
ОтветитьYeah... relatable..
Ответитьits true if you don't have any experience, try building a house ))))
ОтветитьI have a 3600 line program I made at work using chatgpt. It works, but I have no clue what i’m doing 😅
Ответить10.000?
Dude, I work in a codebase with probably a couple million lines... The CMake files only are way past the 10.000 line mark.
True 💯 😂😂😂
Ответить🤣🤣🤣🤣🤣🤣
ОтветитьThis is why i like OOP
ОтветитьNo
Ответить"wow this does exactly what i want" i think to myself, almost like i wrote it myself
ОтветитьThis is just skill gatekeeping dressed in personal allegory
ОтветитьTrue
ОтветитьDon’t like this. This dude is biting Thor’s flavor, and it sucks.
ОтветитьSo true
Ответитьwhy axis tho
Ответитьin sort term we call that FAFO
ОтветитьI never regretted my design choices. The code I produced is never perfect but is always properly designed more or less with reusable components and functions and can always be improved with some refactoring.
And the bad design choices are due to deadlines to meet so it’s not a regret it’s some hacky things under control that were added but to be consolidated later.
Overall what I don’t recognise these regrets on my projects
Yes, but how do I make it go away? It's consuming my life and making me miserable
ОтветитьTLDR: Technical dept
ОтветитьI worked at a company where every single file was a 10k lines of cluster fuck nested if, nested n^4 loops and nonesense circular dependencies. It was the biggest cluster fuck I ever saw. I quit soon after.
ОтветитьAll my projects are sub 2000 lines. Easy.
ОтветитьI found that this is decreasingly so in more modern technologies, I have a Svelte 5 project with 50k lines going damn well and feels like I can continue to add more and more and it will never become shitty
Regret comes from "new feature requires something that the underlying architecture doesn't have"
It will reach a day where our base shared understanding of "what an underlying architecture actually needs" will include the things we need, when that time comes I think, we may reach programmer Nirvana
I've been looking at Convex recently and I think their hearts are in the right place, I love the project, but it's still very young, but they're walking the correct path, the path of the future I speak of
Yep I can definitely relate to that😅😅😅
Ответить