Oooeeh, that is especially something I always fix for my software!

From the iPhones, 4s ftw! I loved it in design and size. It's really small by today's standards btw. I mainly used it as 2FA thingy. Disabled the internet, lasted around a month on battery. (Sounds extreme now i'm saying it, so just checked it, is actually possible indeed. On top of that, I actually got the phone for having battery issues :P But it seems as you literary would use it for only 2FA with airplane mode it would drain 1% per 12h, making it indeed possible, even with a lesser battery. More people have seem to have reported such thing)

@12bitfloat same for python actually, I enjoy both almost equally, but i tend more to the python ecosystem because it's closer to system development. It's really good for local and web. If I would use js to do something on a terminal it feels like I am hacking something. And python has a reflection system what js only can dream of. Like, i can extract anything from all classes and methods and it has by default it's own lexer / ast system built in. Python comes with all the stuff included to diagnose himself. Give me a hour, and you'll have a nice python documentation system. The power that python has over itself is amazing. So I prefer python. But Node has the beautiful package website. Contributing to it is also very comfortable. Ever seen the python package website? It's just sad.

It's just the unhappy pills :P They're not for helping you, but the environment around you :P The actually describe it also to kids until 18 years that have extreme behavior problems. So, you can thank them for the compliment :P Also to people to calm down when they die actually.

@D-4got10-01 you said nothing? :P But based on this application is https://ideas.app.molodetz.nl made. I know, it's all worthless, FOR NOW. You have seen the bots at my social media site (the Reddit-like one) and they're getting really good in behaving and posting. I am just making more sources for them. Their own ecosystem of information. It's nice because it's up-to-date real life information where it all is based on. So, if someone would join my social media platform, they're actually up-to-date with tech news of the latest hour wherever publised (because it's rally unique amount of quality sources it uses). Read that site, it's not BS (the news one, the ideas one is so far). This news site is objectively probably one of the best sources to stay up to date in technology sector.

@lorentz True in whatever you say. I personally vouch for @blindXfish his product because it's clearly looking better than the things I make. For me were other things more important, like a literal clone with devRant api support what I made a while ago. But the blind fish seems to really upgrade the site in a positive way while remaining a devRant look and feel. Snek became a close active community but yeah, it's chat and it has a layout that will just stay there and no, it will never be possible to delete your messages or change avatar :P Because I personally like it :P But it does get frequent nice stuff. And that is a fun thing about being part of Snek community, you see daily to weekly progress and the struggles often of making such platform. Also, the worth of AI becomes very clear to people who've spend much time in the channel since every premium LLM is automated there in some way so far and everyone had pros and cons. I'm very happy with all my testers / fellow dev.

@antigermanist I did stuff before it was cool. Anal for example.

A working MVP of Reddit, built by my system costs around 60 cents and is covered with playwright fucking tests. It is extremely efficiently made. My older versions costed often 60 cents per minute :P I think I've discovered a golden hammer, also for continues improvement of this system. I don't work anymore, also don't want to. But this system is like wow, would be sad if I won't make it commercial :P But I have all the time in the world and will just work on it, improve it. These days, I do not believe that new-comers enter the business still. So, why even bother. Fucking marketing world.

@afaIk there is a chance that I buy it again, but recently I also have seen system86 machine that could be the only correct replacement. I am looking for months for different machines and i just do not like them. But this one seems to be very interesting: https://system76.com/laptops/...

The x270, I just have to admit that by now I also notice myself that the screen is a bit old. The HD is a bit small and don't want to fuck with that shit myself anymore and so on. I mainly live in the terminal and do a lot on server because I use vim anyway. And for just running terminals, it's like a monster beast :P There is no upside to upgrade enough for now. But this one: https://system76.com/laptops/.... What a cuty.

Because of my low end laptop, I decided to make my own roocode and my own version is actually now better than the roocode itself. It completely works in terminal. It builds software, and runs playwright browser for you to show it works. Sick huh.

@jonathands In general, if you have nothing better to waste money on, just do the computer. Why not. I can't look in your wallet. I personally work on a x270 and work on AI 24/7 and I've spent around 800 this year. I would not say that is that much. I just ran:

```

Total tokens: 327761

Total AI call time: 691.89s

Total elapsed time: 706.76s

```

Fetched financial data from perplexity:

If all tokens were output tokens

327,761 tokens × ($1.50 / 1,000,000) = $0.49

If all tokens were input tokens

327,761 tokens × ($0.20 / 1,000,000) = $0.07

Typical blended scenario (3:1 input/output ratio)

Using the industry-standard 3:1 ratio:

Input: 245,821 tokens × ($0.20 / 1,000,000) = $0.05

Output: 81,940 tokens × ($1.50 / 1,000,000) = $0.12

Total: $0.17

So, to give you an idea, i spend literally that much time on full time AI calling and it costed me 17 cents in Blended scenario (out/in balance) and around 0.50 in worst scenario.

@afaIk uhm, owning something yourself? I mean, someone is making profit on it, so it should mean you can do it yourself better and cheaper. But, in case of AI, that is very not true at the moment. But yeah, if I find a way to drop the cloud i would do it immediately. I also do not do AWS or whatever and configure my dedicated servers myself and have never unexpected costs or whatever weird shit. It's mine and that's comfortable.

So many reasons not to choose the cloud, but at this moment, we're tight into it.

On the subject btw: models are getting lighter and lighter. One of my favourite is Gemma3:12b to automate. That's amazing. A year ago, that was impossible with such model.

But if you code well, automate certain stuff yourself, you do not need really heavy models.

Currently playing with grok-code-fast-1 and that is by far the best bang for the buck I have ever seen. No Opus for me. But the new one from Claude (uhmz,) Sonnet 4.5 is also amazing. Sine then no Opus.

But what AI makes slow, uhmz, that is as you know, the context length. And if the context length is not important or keeps small it can be a good choice I guess, but it will take a decade before it will be a good financial choice in comparison to the cloud based solutions. I personally moved everyone to Ollama and ended up vouching for the cloud solutions. Even in a financial matter. But be aware, there is a difference between OpenAI and OpenRouter for example. On OpenAI is blast requests of concurrent using a gather(). Meaning, i do 20 requests at once regularely. Grok is fine with, OpenAI is fine with it. But besides that, NO ONE You'll be rate limited, on any other provider. And groq, the one most famous about their inference, is literally the worst in delivering up to their promise. Fast inference, but no concurrency. What is the point ffs.

But please explain what you actually want with the AI. Your subject is very important to discuss.

Automation takes small context norm.

If the AI is a real goal, it is an interesting bet.

Because, the AI you are used too, is absolutely not the AI on a local machine. The local AI's on default graphic cards can answer a question for sure, but real automation takes more than that. Also, the cost of the local machine is several thousand euro's. Well, do you have any idea how much AI you could buy from that? I am a mayor spender on AI. Sometimes because i ran a loopii in the background or was just inefficient. The costs I made are more on me than the supplier. Also, such light model qwen, maybe he wins according to some statistic chart, but I actually doubt if it would even beat gpt-4o-mini or gpt-4.1-nano. I'm not kidding. Benchmarks do say in AI world completely nothing, especially not automating with it if that is goal.

Your comment remaining claude / codex, that's not just model. It's a very good algorithm behind it in pure code. At this moment, i destroyed roocode with my own build system using light model.

@djsumdog yeah, sure and fine that you have issues with it, but blame the developers (or their companies they're working for), not the technology. The technology is freaking amazing in the good hands. I may hope that you're seeing that at least.

That's a dilemma, because people normally don't want to be neither of them. We want to be special so we color our hair and do drugs. Pay attention.

@sandeepbslan webdav does get requests with a body. Ever heard about the http methods PROPFIND PROPPATCH LOCK MKCOL?

That reminds me, i have to add them in my https://molodetz.nl/dashboard reverse proxy, i have now my webdav servers down i guess.

That reverse proxy really costed more time than expected, but it does some magic causing me to interpret the HTTP protocol and all it's exceptions. Websockets, chunked, SSE. Whatever. Grr. Written in C, does not cause any overhead. My python version sometimes had a hard time, see what data is going over the line there often.

@sandeepbalan it's actually the best web protocol that windows supports. It's old, but battle tested. You can mount is as hard drive in windows and Linux. Next Cloud also supports it. SSH is actually better, it offers also more features. The webdav engine (dav2fs) is not so sophisticated. But I worked on such project several times and Snek has it also built in, self written in aiohttp. My first server was many years ago. It was in C single threaded with Linux select system call. What did I spend a lot of effort to stream multiple files at once buffered. The performance was amazing. But the final version was in the end over-engineered and even had a web-interface and got lost. Very sad about that, because it was my first 'bigger' project in C. In C such project is quite big. Also, it was a challenge to make it working on Windows as drive and on Linux and on Cadaver (webdav native client). With webdav, you can also tag files and you have locks for writing to prevent issues.

@SidTheITGuy you really should start to do nice again. What is wrong with you lately. You're sleeping on the couch or something?

On the subject, i was making fun to a (and about a) German coworker (on a new place I worked) during a break and one of the HR's looked at me like: "Oooh, i let it slip because you're new, but else..". My god, like if a German is a minority. If we can't make fun of the Germans anymore, what to do with them?

What is your worth anyway there if there is time for such shit. Also, I feel sad for the people who think about slavery when they hear the name 'master'. It totally do not notice the word.

I assume that was not the only reason you left.

I found the idea to change from master to main a sign of incompetency, how much things would've broken it? On top of that, it's like batletasted C (gnu) apps replacing with Rust. Real incompetency. Especially these days with our 'all software contains bugs'-attitude. Good luck.

Btw, did it actually change? No right? Maybe at github? My git is always still master, and molodetz advises to make main the default branch in the docs after creating a repository. But that's it I guess. Git is unchanged? Or does a rainbow fork exist?

Swastika, it will make her think like "Wow, how did he do that?" and she'll be impressed for ever.

Fuck, it's just the last optimization i guess. Fuck. Will check tomorrow. Now working with roocode on some stuff. It's getting fun.

Cross-Network Discovery:

How to connect phone and desktop on different networks?

→ A lightweight Redis broker that just handles handshakes.

omg, it's all just a joke huh? Someone is testing his AI bot.

I actually would really like the stuff, but why aren't you writing about it yourself? I understand people correcting their English and punctuality, do minor upgrades. But if you rape your post like this, what do you expect? I see so much shit that aint even true.

Why did you actually specifically want copilot in your pocket? It's just claude/gpt right? Could've been a direct connection with no home setup.

I like the leader board. I down voted people to come on top. Sad that it's not allowed anymore to up vote yourself. Posting, up voting myself, while voting others down was my strategy and it broke now. Please fix.

@c3r38r170 nah, you can't expect from people to know your experience level. That is always the big question with meeting people online.

I am actually right that he says 'know how to code'. Because i sell that as argument as well, if you know your shit, you can explain it very detailed which is exactly what an LLM needs. But since you can, there is no issue right?

Sometimes, just watching what an LLM makes of it is also fun, just for concepts or to get an understanding on what level the LLM actually is.

I have no idea why not to use an LLM if you get expected results. Whatever your expectations are. I expect specific structure, layout and such is often just by LLM and for me it'sy fine because it's mostly better than my own visual designs. I suck that hard at that.

@devux-bookmark my prompts take hours. How long does yours take? Also, the AI (in case of claude) does exactly what you want (especially when you tell him to) and you just have to describe literally everything. And it simply does what you say. I do not have much difference between code I wrote myself and AI. AI is actually doing it a way better, more exception handling and such. Just invest more time in it, finetune your self written prompt with AI, to ask what is unclear about it, update your prompt until everything is clear and you won't get surprises. I do use this technique a lot and I once had like 50 answers to give. So 50 things would've been unclear and thus i would've had a completely unexpected result if i did not let review my prompt with the AI. You actually don't want the AI to make any decision. You do yourself, but you have to be specific about it. Consider it like this: would a different person understand your explaination? Explain it as to an junior.

@jestdotty well, if you just deliver the expected quality in whatever way, nobody cares. So sure, if you achieve with an LLM to write the code for ya, you sure can get a job like that.

@djsumdog all those things you fix afterwards, you could've told the LLM up front, and in case of claude, it will perfectly listen an do exactyl what you say.. Also regarding conventions. Good prompting is a process for hours, not of minutes. Because you have to tell that thing literally everything what you want to achieve. Also, you can ask the LLM, what is unclear about your prompt and it will tell you, and you can address the answers and ask to update your prompt with given answers and then you have a complete one. So, not making prompts with LLM, but finetune your ones. Never let the LLM decide anything. It should "Do as i say, nothing more, nothing less". My default sentence at the end. That prevent's slop.

@AlgoRythm regarding that, how can't you not get caught, review process right? I mean, with decent use of AI you won't see a difference. You force it to work as you do yourself. The comments are not end of the world - but you can disable those. I do not mind that much anymore. I accepted that the world of programming is changed and don't care anymore if someone knows that I use AI. I did do many years without, i know the struggles, and with that knowledge I am able to instruct LLM's now and they are just the best tool to work with for me. So, why wouldn't i use it :) It would be shooting in my own foot. Then i can type everything myself again. Why would I, I still type enough but at least not the 1000th docker configuration over and over again.