mawadev

It is either me or you

https://www.youtube.com/watch?v=DCkJ5lGPqFs

That is — incredible

I was quite shocked to witness a watchmojo Top 5 drivers behind software delivery inefficiencies kind of research paper

Very cool and very performant

Very cool then, my personal experience and experience in general with assembling furniture could be different to what regular people who never did it before have, seen it with friends

Very true, I do it with regular screwdrivers but I watched people in the past ruthlessly overscrew and hammer stuff in, its not supposed to be done with force. My furniture so far survived 2 moves and 14 years with full disassembly and assembly in between..

These days I wouldn't recommend Ikea to anybody with the prices and build quality, Jysk is a good and cheap alternative in germany.

I think its AI sentiment generated data, no way someone needs 6.5h for a bed frame. I needed at best 2h...

This is just the cloud area, what if Amazon starts vibe charging regular customers because of some bug? Accounts that are directly linked with regular people's payment methods?

Linkedin is a site where LLMs talk to each other

I think this post completely destroyed anthropics narrative

There are a lot of posts going on, but let me just post this link to a specific file of the bun zig code base so you can have your own impression:

https://github.com/oven-sh/bun/blob/bun-v1.1.42/src/bundler/...

My take is they vibe coded too close to the sun

They are large language models, they can translate very well, but they cannot form original changes without heavy supervision/review and indefinite feedback loops. Most of these projects you see are just a way to gather attention or justify technology that is flawed at a fundamental level for the use case it pretends to be good for. Rust in particular has a very strict compiler and a lot of tokens/language features/syntax that hint at specific behavior tightening the feedback loop.

Maybe I'm very uneducated on this, but what exactly is the problem with postgres in its current state? Why do you need rust to make changes easier?

The only justification I see here is this claim and it is nowhere published or verifiable:

"Update: We're working on a new not yet published version of pgrust that currently passes 100% of Postgres regression suite, has a thread per connection model instead of process per connection, is 50% faster than Postgres on transaction workloads, and is ~300x faster than Postgres on analytical workloads (2x slower than Clickhouse on clickbench and we think it can get faster than Clickhouse). Follow pgrust or join our Discord for updates!"

I sometimes wonder why the algorithms started pushing doomsday scenarios, especially on youtube. Most channels have big red arrows pointing at a miniscule thing and say "its over". Then you get ai 2027 and its more fear mongering, just inside the AI echo chamber on the internet. At work even managers watch odd youtube videos that are radicalized towards politics and so on, but are ultimately head canon slop for ad revenue.

It feels like we are so utterly bored out of our minds and comfort that we make up problems or scenarios, completely detached from reality, or outright irrelevant to day to day reality, just to get through the day and give meaning to our lives.

It is an incredible achievement, don't get me wrong, but what is AI truly going to do when we are already at such a stage of devolution?

It reminds me of the scene in Wall-E when people destroyed the planet and started flying through space to find a new planet, while the captain is an AI and the people became obese in mobility scooters and glued to their screens. I think this is unironically the most realistic scenario for mankind lmao

I guess it is cool to have it around but it comes off as a popular useful case for AI which it really isn't, unless you have a very good test suite to throw the agent against, it is practically useless for rewrites of decades old badly documented code riddled with technical debt. That is where the real value and challenge would be. I see it as marketing and social clout stunt

I was once in a meeting with a guy for a specific purpose and he wasted about 10 minutes lecturing me on why he uses vim, I had no issue with it but honestly that entire world is absurd to me, do what you want as long as it works for you

This entire thread reads like I got flamed by LLM bots

I think the blog post was cringe, but the part with relieving a burden stuck with me. I don't think its acceptable to have trillion dollar companies or companies at all sit on your neck using your project in a way you don't like and subtly influencing its trajectory just because they want to make money. License back and forth, but it doesn't feel right to me personally. Its like the guy who wrote the stuff that is used for the intrusive Intel IME. Now pair that with the tasteless LLM and AI narratives and you got yourself into a big dilemma.

Let's say you make a language and try stuff out but its used by facebook and suddenly you find yourself in calls solving problems for them but you never really wanted to end up in this situation or work with people from such in environment. Meanwhile the compensation/donation does not match the value they are extracting from your work. Hard to say, I'm very mixed on this.

"The primary bottleneck to this growth is the availability of electricity."

We put the cart before the horse, lets stop talking about growth and focus on making it useful first lmao

That took a while for them to catch up on what we all knew back in 2023

The worst part about this is the amount of lag you have when taking a photo. On samsung phones, its quick and seamless, on other android phones the IO is so slow, one picture is taken and before you can tap the shoot button again you have to wait until it saved the previously shot photo. Sometimes there is even a redundant animation that moves the entire frame into the little corner that points to your album to the bottom left. When there is a firework and you want to have a lot of pictures in short timeframes, this kills it. Animations in general are very annoying and inconsistent visually, it makes you unnecessarily wait or take up a lot of resources if the "native" UI is a SPA in a browser instance..

The article is also heavily ai generated, I call bs on every single bit

I love this take. I built some feature and then when I tried to layer new features on top, I saw all the bad decisions surface and errors appear. Its beautiful to walk back and have it "click"

Incredible quote, thank you for posting it

I have to say the LLMs probably can do a lot of these things properly if you had access to the entire infrastructure for a single project but as you said, it is simply not economical to use. I'd rather have people maintain systems before lock in happens and someone hikes the prices and rug pulls me. In almost all cases you need someone experienced anyway to sign off on the changes it makes or to keep guidelines and guardrails intact.

I don't believe the "you are holding it wrong", "works on my machine", "works on this model" or "do this spec structure" type of arguments to compensate the fundamental issues. The tech simply does not do what is advertised and claimed as it is.

Feels like we try to turn anything we do into a career or job, especially with the odd ML angle. How about you "do graphics programming" instead of "being a graphics programmer"? Like start doing simple stuff until it clicks and you see it for being logistics to the GPU, then you can layer on top all the crazy concepts. Its like a small mountain you climb and suddenly everything clicks and you think like "oh my"... the possibilities and things to experiment with...

These companies are so unlikable, I don't understand why they think people want to permantly mount their data collection systems right in front of their eyes. You couldn't give me one of those for free. Its like someone breaking into my car and leaving 2 copilot licenses on the seat: a nightmare

Remember the guy who had an entire island and all his friends literally have had no consequences? You really think the age checks will do what is advertised? Lmao

What if your phone listens to you all the time so it can detect "ok google" or "alexa" 24/7 but you happen to say something illegal with your device ID?

What if you brain storm a book/plot idea with a friend aloud and moments later the police knocks on your door because some system said you are about to commit a crime?

If you look at how workers are mistreated in authoritarian corporations and how some people say it is perfectly fine because you get paid a lot of money to be used and abused, I can't imagine what a reality like this would look like. It does not stop with what you post online, we still can't really tell how our phones know about stuff we talk about irl to serve ads.

Software has always been marketing! Because most software straight up doesn't work or does what has been advertised lmao