I have a job stalled right now, I believ because a jackass manager decided to get an opinion from AI on the design for a larger steel ramp structure, which is wrong, but the jackass has zero capability to respond now that I have asked for clarification as to which numbers were used in his "calculations".
As of right now I have a whole string of jobs hung up because of errors in designs and blueprints that are bieng sent to me to fix, somehow, but with, again, zero capacity to make descisions, deal with the cold, hard
reality of figuring out how to build something out of metal, something they need, now, but only one unit, which has certain indeterminate issue, requirement, or aquardness that will tie up and jam a whole large organisation, where quite litteraly a broken door handle repair, ends up getting bumped, up, and up, to the top, as there is no PO, part numbers, mission statements, or glad handers, and NOBODY wants it on THERE desk.
so the front doors, on a major asset located on very prime real estate, sit there flapping, and getting a hunk of chain wrapped around them for months, and now they cant figure out how to pay for having that, suddenly fixed.
if I was less busy, I would be angry
but it is turning out to be an amusing and bemusing viewpoint of the lurching ,jerking, debaucle brewing itself up.up↑ there
metal, the bending, joining, of metal for humans to use for something™,who find me through the interwebs , which I have been useing since the dawn, off and on, clumsily, but since grade school. the apple store was one room above a chinese resturaunt and had painted chip board walls.
I have two web sites, one is a rental and I own the other, but I am focusing more and more on my core strengths in dealing with physical realities, which sometimes I call "applied geometry", though often there are curves and shapes that dont realy have names.
But as a good deal of the work is designed and comunicated about with the use of computers and phones, I also spend a lot of time thinking about how that could be better, so hanging out here , trying to fight the good fight, is part of most days.
Even if it did work well, with such a rough start and years of false promises, who is really going to trust it? Everyone who does seems to be riding the hype wave.
Nobody with any sense believes any of the hype. It's the boy who cried wolf effect: after years of improvement it still sucks and can't get work done, so why on earth would I trust in the future when the AI bros claim "no this time it really is good"?
“It passed all the unit tests, the shape of the code looks right," he said. It's 3.7x more lines of code that performs 2,000 times worse than the actual SQLite. Two thousand times worse for a database is a non-viable product. It's a dumpster fire. Throw it away. All that money you spent on it is worthless.”
“This magic that literally didn’t exist two years ago in more than a toy state is moving at such a rapid rate that it couldn’t even reproduce sqlite three months ago and only got better enough in those weeks to produce a bad version of sqlite! Clearly useless! It has no value, no one is using it to do any work and won’t get better over the next three months or three years!”
You’re shifting the goalposts. The initial point was that the Rust regeneration of SQLite was wasted money, because it’s unviable due to its slow speed. You’re trying to shift it to be about how it may get better over time. Do you have something that is more specifically refuting the initial quote that doesn’t involve anything about potential improvement?
The point wasn’t to make better SQLite, it was to make a functioning rust SQLite. Which it did. Badly but you don’t start at race cars. No one was assuming production SQLite.
Well... the actual problem is, imho, that it looks like the LLMs seem to have reached (or are close to reaching) a plateau.
You might be right about the "three months ago it could not produce a working implementation of a DBMS... but what if in 3 months (or 3 years) it stays stuck at the 20K slower threshold?
People have been saying, without any evidence at all, it's reached or about to reach a plateau for years now. We are clearly still seeing significant forward progress. While it's reasonable to think it will hit some plateau eventually, there's no reason to think that right now just happens to be as good as it's ever going to get.
I think that's adequately addressed in the article:
> "The other way to look at this is like there's no free lunch here," said Smiley. "We know what the limitations of the model are. It's hard to teach them new facts. It's hard to reliably retrieve facts. The forward pass through the neural nets is non-deterministic, especially when you have reasoning models that engage an internal monologue to increase the efficiency of next token prediction, meaning you're going to get a different answer every time, right? That monologue is going to be different.
> "And they have no inductive reasoning capabilities. A model cannot check its own work. It doesn't know if the answer it gave you is right. Those are foundational problems no one has solved in LLM technology. And you want to tell me that's not going to manifest in code quality problems? Of course it's going to manifest."
You can argue with specifics in there, but they made their case.
I have a job stalled right now, I believ because a jackass manager decided to get an opinion from AI on the design for a larger steel ramp structure, which is wrong, but the jackass has zero capability to respond now that I have asked for clarification as to which numbers were used in his "calculations". As of right now I have a whole string of jobs hung up because of errors in designs and blueprints that are bieng sent to me to fix, somehow, but with, again, zero capacity to make descisions, deal with the cold, hard reality of figuring out how to build something out of metal, something they need, now, but only one unit, which has certain indeterminate issue, requirement, or aquardness that will tie up and jam a whole large organisation, where quite litteraly a broken door handle repair, ends up getting bumped, up, and up, to the top, as there is no PO, part numbers, mission statements, or glad handers, and NOBODY wants it on THERE desk. so the front doors, on a major asset located on very prime real estate, sit there flapping, and getting a hunk of chain wrapped around them for months, and now they cant figure out how to pay for having that, suddenly fixed. if I was less busy, I would be angry but it is turning out to be an amusing and bemusing viewpoint of the lurching ,jerking, debaucle brewing itself up.up↑ there
Your job sounds really different from what's typical here on HackerNews. I'm really curious - can you tell us more about it?
metal, the bending, joining, of metal for humans to use for something™,who find me through the interwebs , which I have been useing since the dawn, off and on, clumsily, but since grade school. the apple store was one room above a chinese resturaunt and had painted chip board walls. I have two web sites, one is a rental and I own the other, but I am focusing more and more on my core strengths in dealing with physical realities, which sometimes I call "applied geometry", though often there are curves and shapes that dont realy have names. But as a good deal of the work is designed and comunicated about with the use of computers and phones, I also spend a lot of time thinking about how that could be better, so hanging out here , trying to fight the good fight, is part of most days.
„companies will ask for discounts when they know a service company is using AI tools“
„Insurance underwriters are seriously trying now to remove coverage in policies where AI is applied and there's no clear chain of responsibility“
I see a future coming, where everyone uses AI but nobody admits it.
You mean like today?
Even if it did work well, with such a rough start and years of false promises, who is really going to trust it? Everyone who does seems to be riding the hype wave.
Nobody with any sense believes any of the hype. It's the boy who cried wolf effect: after years of improvement it still sucks and can't get work done, so why on earth would I trust in the future when the AI bros claim "no this time it really is good"?
This is a hilarious article.
“It passed all the unit tests, the shape of the code looks right," he said. It's 3.7x more lines of code that performs 2,000 times worse than the actual SQLite. Two thousand times worse for a database is a non-viable product. It's a dumpster fire. Throw it away. All that money you spent on it is worthless.”
“This magic that literally didn’t exist two years ago in more than a toy state is moving at such a rapid rate that it couldn’t even reproduce sqlite three months ago and only got better enough in those weeks to produce a bad version of sqlite! Clearly useless! It has no value, no one is using it to do any work and won’t get better over the next three months or three years!”
An amazing take.
You’re shifting the goalposts. The initial point was that the Rust regeneration of SQLite was wasted money, because it’s unviable due to its slow speed. You’re trying to shift it to be about how it may get better over time. Do you have something that is more specifically refuting the initial quote that doesn’t involve anything about potential improvement?
The point wasn’t to make better SQLite, it was to make a functioning rust SQLite. Which it did. Badly but you don’t start at race cars. No one was assuming production SQLite.
Well... the actual problem is, imho, that it looks like the LLMs seem to have reached (or are close to reaching) a plateau. You might be right about the "three months ago it could not produce a working implementation of a DBMS... but what if in 3 months (or 3 years) it stays stuck at the 20K slower threshold?
People have been saying, without any evidence at all, it's reached or about to reach a plateau for years now. We are clearly still seeing significant forward progress. While it's reasonable to think it will hit some plateau eventually, there's no reason to think that right now just happens to be as good as it's ever going to get.
7 replies →
Opus 4.5/4.6 are definitely not plateaus. Definitely a threshold of quality improvement.
20Kx slower is still faster than my manager could write it.
Where are my flying cars?
Flying cars exist? You can get one.
Why do I want to reproduce sqlite? It’s a library. The point of it is to be already written.
Maybe a native rust version of it has value for some people? :)
Pretty big jump from "we don't yet know the best ways to use this new tool" to "the tool doesn't work well"
I think that's adequately addressed in the article:
> "The other way to look at this is like there's no free lunch here," said Smiley. "We know what the limitations of the model are. It's hard to teach them new facts. It's hard to reliably retrieve facts. The forward pass through the neural nets is non-deterministic, especially when you have reasoning models that engage an internal monologue to increase the efficiency of next token prediction, meaning you're going to get a different answer every time, right? That monologue is going to be different.
> "And they have no inductive reasoning capabilities. A model cannot check its own work. It doesn't know if the answer it gave you is right. Those are foundational problems no one has solved in LLM technology. And you want to tell me that's not going to manifest in code quality problems? Of course it's going to manifest."
You can argue with specifics in there, but they made their case.