Thanks for clarifying that. I always thinking it was due to my bad hearing or lack of fluency! English is my second language, and without subtitles, I'm having trouble to understand the dialogues without increasing the volume until I reach a level where I also got disturbed from the sound effects, bg music etc. It's a bigger issue with movies, not only due to what you explained about sound mixing, but also error correction of my brain works worse without the context, unlike series where I know the plot and the characters.
Watch out for any show with Christopher Nolan as director. He is notorious for terrible audio mixing, and having actor mumble and otherwise dictate horribly.
As an american and English as my only human language, even I need subtitles.
And the movie "Tenet" was so bad that I ragequit 30 minutes in. Horrible horrible audio.
It's funny how this coincides with a time when BT headphones have finally become cheap, reliable and capable enough. I recently bought two different sets from Lidl: one for €8 and the other for around €12. Both have ANC and a battery life of around 5 hours, and the sound quality is quite respectable. I've been using headphones all the time since I was 11, so that's 37 years with many different kinds of headphones. Even now, I have more than ten headphones that work. IMHO, Bluetooth headphones have never been closer to becoming a natural counterpart to mobile phones for everyone.
In this age, rich TUI's feels wrong to me. Tools that expose a minimal web server with a lightweight UI is much more welcome than a complex TUI. But for most interactive terminal apps, it feels more natural when there is a single input at a time, like a wizard interface.
Qwen3 Coder Next and Qwen3.5-35B-A3B already very good and can be run on today's higher end home computers with good speed. Tomorrow's machines will not be slower but models are keep getting more efficient. A good sw engineer still would be valuable in Tomorrow's world but not as a software assembler.
Even cutting edge models are not very good. They are not even on mediocre level. Don’t get me wrong, they are improving, and they are awesome, but they are nowhere near good yet. Vibe coded projects have more bugs than features, their architecture and design system are terrible, and their tests are completely useless about half the time. If you want a good product you need to rewrite almost everything what’s written by LLMs. Probably this won’t be the case in a few years, but now even “very good” LLMs are not very good at all.
Not sure why you're being downvoted, this is very much my experience. When it matters (like, customer data is on the line) vibecoded projects are not just hilariously bad, but put you in legal danger.
We've so far found that Claude code is fine as a kind of better Coverity for uncovering memory leaks and similar. You have to check its work very carefully because about 1 time in 5 it just gets stuff wrong. It's great that it gets stuff right 4 times in 5 and produces natural code that fits into the style of the existing project, but it's nothing earth-shattering. We've had tools to detect memory leaks before.
We had someone attempt to translate one of our existing projects into Rust and the result was just wrong at a fundamental level. It did compile and pass its own tests, so if you had no idea about the problem space you might even have accepted its work.
With Claude Code now having a /plan mode - you can take your time and deliberate through architecture and design, collaboratively, instead of just sending a fire-and-forget. Much less buggy and saves time if you keep an eye on the output as you go, guiding it and catching defects, imho.
For that you need to create something which you know exactly how you want to code, or what architecture is needed. In other words, you would win basically nothing, because typing was never the real bottleneck (no matter what VIM and Emacs people would tell you).
LLMs also make mistakes even way lower level than those one pagers allow you to control with the planning mode. Which I use all the time btw. And anyway, they throw the plan out of the window immediately when their tried solutions don't work during execution, for example when a generated test is failing.
Btw, changing the plan after its generation is painful. It happens more than not that when I decline it with comments it generates a worse version of it, because it either miss things from the previous one which I never mentioned, or changes the architecture to a worse one completely. In my experience, it's better to restart the whole thing with a more precise prompt.
Ah, this is true - for my purposes, I've been directing the design and deliberating on the constraints and specifications for a larger system in tandem with smaller planning sessions.
That has worked well so far, but yes, you are totally right, there are still quite a few pain points and it is still rather far from being fire-and-forget "build me a fancy landing page for a turnkey business" and getting enterprise quality code.
edit: I think it is most important that you collaborate with Claude Code on quality in a systematic way, but even that has limits, right now - 1M context changes things a little bit.
For me, personally, I'm building things that would have been impractical for me to do as cleanly within the same amount of time - prototypes in languages that I don't have the muscle memory for, using algorithms i have a surface level understanding of but would need time to deeply understand and implement by hand, and, at my pace, as a retired dev, is probably quantified in terms of years worth of time and effort saved.
edit: also, would I take the time to implement LCARS by hand? No. But with an LLM, sure, took it about 3 minutes or less to implement a pretty decent LCARS interface for me.
The way it's going, the AI hyperscalers are buying such a big portion of the world's hardware, that it may very well happen that tomorrow's machines do get slower per dollar of purchase value.
Not my experience. Current Qwen Coder is noteworthy but still far from good. Can't compare them with current commercial offerings, it is just different leagues.
I do exactly same, even uninstalled their shitty tv app which managed to be stay slower than other streaming apps for many years, even on fire tv stick.
I only needed help of this banana boy twice, it managed to disappoint me each time. The most recent one, I was trying different beard and mustache styles on myself, on a photo I imported from my own Google photo gallery, and it consistently rejected me, claiming I'm a public figure. Nobody ever told me that I look like any famous person, so that's googles own bananination. ChatGPT nicely handled the job.
In my opinion, for open-source projects, scoring the project's AI sloppiness based on the timeline of commits would be a good indicator. If it's completed within a few days, it should require more thorough human review. On the other hand, if the project has been active for a while and received contributions spread throughout that timeline, I think that would indicate accumulated effort (human and/or AI) and higher quality.
> In my opinion, for open-source projects, scoring the project's AI sloppiness based on the timeline of commits would be a good indicator.
You can’t necessarily judge by timeline. I’ve always developed my projects privately and then squashed to one initial public commit. I’ve got a private repo now with thousands of commits developed over years and I still intend to squash.
Show HN has never been restricted to open source projects and it would be weird to make the criteria more restrictive for open source than closed source work.
I thought so too, untill I looked for 1-point Show HN posts with a repo with a long commit history. Some of these are really cool (see my article), but others were not compelling at all, at least to me.
Eh IMO any metric like this can be gamed. My project that reached hn front page was coded in a short time (and yes some ai was used), but otoh I think it was something that showed hey you can do this really interesting thing (in my case vlm based indoor location).
Also its not uncommon for weekend projects to be done in a shprt span with just a "first commit" message dump even pre-AI.
Yes, any metric can be gamed. But I believe measuring the entropy of a repository, comparing state of the code-base over time can be done deterministically, which would make it harder to game it.
So either we are going to completely avoid automation and create a community council to decide what deserves to be shown to rest of the community or just let best AI models to decide if a project is worth show up on front page?
Isn't it possible to fabricate the timestamps on commits and then push them up all at once? If you're planning on literally checking that the commits are publicly available for a certain amount of time, that seems like it would needlessly punish projects that someone worked on offline and then happened to push up once it was completed.
Last time I checked, having a continuously running background process considered as a daemon. Using SQLite as back-end for storing the jobs also doesn't make it queueless.
If putin, then trump and their people agreed on that we are no longer living in a rule based world, patents, licences etc. would hold little value. Realpolitik of the globe will kick everyone's ass.
US let Russia take a chunk of ukraine and China and Russia to certain extend let US control its own hemisphere
just eat up that some major power always playing geo-politic war games that exert its influences
they maybe have a friction and want to mess with each other but the domain of influence is always there and they generally dont want to cross the line for it
reply