Sorry man but what I see hurts my heart, localisations are in a single 1.5 mb file and the entire source is spread across two files, one 18kLoC and another 2k - that's another 900 kb. It's surprising this hasn't kicked you into the butt yet honestly. IF you code this with AI, which I very much assume so, you're also wasting a huge, huge lot of tokens I bet. I'm leaving that sarcastic comment out but that's why I'm glad I know some basics about software engineering and archetecture, AI doesn't always steer you the right direction.
I'm aware I'm probably wasting lots of tokens, thankfully however I have 5 accounts so I never reach any limit whatsoever. I will not go into detail on how you can achieve this and not spend hundreds of dollars a month, but your point about only two files still stands and actually I will most likely completely change how the program is built, it has not kicked me in the butt just yet since on my machine it works fine, apart from random bugs which appear pretty much every time I add a feature. I have not had the need to change the architecture of the program yet, which does not mean that it will not happen in the near future. I have 4 codex accounts plus AntiGravity, and I might probably stick with that now.
honestly this is kind of a time waster for me, since I am nowhere close to being a developer AI has finally allowed me to build whatever the fuck I want, and that's the only thing I care about really.
LOL Funny, second time I've seen ChatGPT stick to completely redicolous development patterns. A friend of mine also used it and the codebase is just a tiny bit better - there are lots of files well beyond 400 kb size which easily could've been split. @22 do you just use a vpn to get it cheaper somewhere else? If not maybe we could spin this up in pm
Probably cheap and (usually) expiring after some days accounts from g2g, eneba or other account seller website.
-- (rudolf): LOL Funny, second time I've seen ChatGPT stick to completely redicolous development patterns. A friend of mine also used it and the codebase is just a tiny bit better - there are lots of files well beyond 400 kb size which easily could've been split. @22 do you just use a vpn to get it cheaper somewhere else? If not maybe we could spin this up in pm
yeah not stable, but codex does not rely on an account, you can access your chats from any account so it is relatively simple to just switch if in need, but none of my accounts have expired yet. And yes, currently using kinguin, I have 18 months of Gemini which I picked up for 10 bucks and it's been working for over 2 months already, so my investment has paid off at least.
Yes, in that case it's definitely AI. I hope the source code is in one large file and all the possibly sensitive tokens aren't hashed LOL Edit: before some youngsters come here again to whine about me and AI, I have no problems with anyone coding with AI. What I have problems with is badly managed AI code and the inability to judge if AI is doing it right.
-- (rudolf): Sorry man but what I see hurts my heart, localisations are in a single 1.5 mb file and the entire source is spread across two files, one 18kLoC and another 2k - that's another 900 kb. It's surprising this hasn't kicked you into the butt yet honestly. IF you code this with AI, which I very much assume so, you're also wasting a huge, huge lot of tokens I bet. I'm leaving that sarcastic comment out but that's why I'm glad I know some basics about software engineering and archetecture, AI doesn't always steer you the right direction.
--
Όταν θα κοιτάς την θάλασσα Χωρίς να το θέλεις, θα βλέπεις εμένα Και τα μάτια σου θα ‘ναι θλιμμένα Όταν θα ξυπνάς στον ύπνο σου Χωρίς να το θέλεις, θα σκέφτεσαι εμένα Και θα νιώθεις μεγάλο κενό
because I dont want to deal with people code contributions and broken code when I just want to use apricot player as a normal youtube player, dont want to have to deal with downloading sourse codes and all this other shit to get the thing to work right that's why.
What screen reader are you currently using? I assume NVDA. In case that's news to you, NVDA is open source. Do you have to go through that described process for NVDA? No. So why should you here? -- (Aidenn): because I dont want to deal with people code contributions and broken code when I just want to use apricot player as a normal youtube player, dont want to have to deal with downloading sourse codes and all this other shit to get the thing to work right that's why.
Uh, so this is Cameren Beaver? I'd say nice to meet you but not the topic for that and plus I've heard people say that you arnt nice so. -- (cookie_krumble): S T U P ID A P P GO A W AY YOU BITCH
--
Let the stars be your guide. Let your mind just relax. And when in doubt, meditate.
the faided one, the forgoten, but the second some one's computer breaks, it's natan natan natan help. and i can not handel this angelsclan problem, all my friends are leeving
he is a bad person. deleteing servers, blaiming, telling peeple to format servers
the faided one, the forgoten, but the second some one's computer breaks, it's natan natan natan help. and i can not handel this angelsclan problem, all my friends are leeving
I completely agree with you on this one, and I just want to reiterate that since this is my first ever actual project and I have no prior coding experience, I stupidly relired on AI for the logic, basically the only thing I provided was the design and how the program should handle user inputs. To set the point straight, I was actually planning to completely refactor the code, since I knew myself that it wouldn't take long for the code to start breaking due to its one file nature, but I intended on doing that just before public release. But after random bugs started popping up when changing things around, I did actually refactor it completely, but AntiGravity with Gemini completely messed up the entire codebase, so I went back to Codex which I will probably use for the foreseeable future (fast, generous with the $20 plan and very competitive when compared to Claude). The program was very unstable for a few days but I am happy to report we're back on track, and I can finally start adding more features that will setthis program apart from others, not revolutionary features, but I'm focused much more on actual usability, reliability and above all, speed. -- (ArcticMoon): Yes, in that case it's definitely AI. I hope the source code is in one large file and all the possibly sensitive tokens aren't hashed LOL Edit: before some youngsters come here again to whine about me and AI, I have no problems with anyone coding with AI. What I have problems with is badly managed AI code and the inability to judge if AI is doing it right.
-- (rudolf): Sorry man but what I see hurts my heart, localisations are in a single 1.5 mb file and the entire source is spread across two files, one 18kLoC and another 2k - that's another 900 kb. It's surprising this hasn't kicked you into the butt yet honestly. IF you code this with AI, which I very much assume so, you're also wasting a huge, huge lot of tokens I bet. I'm leaving that sarcastic comment out but that's why I'm glad I know some basics about software engineering and archetecture, AI doesn't always steer you the right direction.
--
--
When you're stuck, just try harder.
This site uses cookies to enhance your experience.
