Similar to my experience, it works well for small tasks, replacing search (most of the time) and doing alot of boilerplate work.
I have one project that is very complex and for this I can't and don't use LLMs for.
I've also found it's better if you can get it code generate everything in the one session, if you try other LLMs or sessions it will quickly degrade.
That's when you will see duplicate functions and dead end code.
Perhaps need to semi-randomize the file size?
I'm guessing some of the bots have a hard limit to the size of the resource they will download.
Many of these are annoying LLM training/scraping bots (in my case anyway).
So while it might not crash them if you spit out a 800KB zipbomb, at least it will waste computing resources on their end.
Much of this feels like when they did studies on people who take mushrooms for example feel like they are more productive, but when you actually measure it they aren't. It's just their perception.
To me the biggest issue is that search has been gutted out and so for many questions the best results come from asking an LLM.
But this is far different from using it to generate entire codebases.
This is also why I think the current iterations wont converge on any actual type of intelligence.
It doesn't operate on the same level as (human) intelligence it's a very path dependent process.
Every step you add down this path increases entropy as well and while further improvements and bigger context windows help - eventually you reach a dead end where it degrades.
You'd almost need every step of the process to mutate the model to update global state from that point.
From what I've seen the major providers kind of use tricks to accomplish this, but it's not the same thing.
Even Google ended up adding more stuff to their homepage in the end.
For a long time they tried to keep it super minimal- and it still is, but there's footer links and signed in header and a whole bunch of other links as well.
Mind you in the early days pages used to have hundreds if not thousands of text links all over the place, the only sites that do this now are the hardcore conspiracy sites where the author just adds several new links a day.
So in this dimension at least web UIs have changed for the better.
Except that nothing is illegal now, Biden showed that with blanket pardons. So for all intents and purposes they will do what they like and Trump can just wipe it clean at the end.
Goes further than that too, suppose the one working on the wooden part is slow and the one on the metal part is faster. And surely the value of one part or another is also different, even though its the combined value that's relevant.
Suppose as well there are a thousand people lined up to make the wooden part but hardly any for the metal, then surely the ones who work on the metal part will (try to) command a higher wage too.
I actually think out of any language async/await makes the most sense for javascript.
In the first example: there is no such thing as a blocking sleep in javascript. What people use as sleep is just a promise wrapper around a setTimeout call. setTimeout has always created microtasks, so calling a sleep inline would do nothing to halt execution.
I do agree that dangling Promises are annoying and Promise.race is especially bad as it doesn't do what you expect: finish the fastest promise and cancel the other. It will actually eventually resolve both but you will only get one result.
Realistically in JS you write your long running async functions to take an AbortController wrapper that also provides a sleep function, then in your outer loop you check the signal isn't aborted and the wrapper class also handles calling clearTimeout on wrapped sleep functions to stop sleeping/pending setTimeouts and exit your loop/function.
fsync doesn't work here right because unix pipes are in memory?
I've had luck elsewhere with nodejs and WriteableStreams that refuse to flush their buffers before a process.exit() using fsync on the underlying file descriptors.
I think at this point I'd rather pay for a desktop browser, something with blocking baked in at compile time.
It would be such a QoL improvement that it's worth paying a yearly subscription for.
My comment was just on the standard behavior. Not sure when it comes to observables, probably comes down to how you are implementing it. But in general it's impossible to cancel a hanging Promise, you have to have some kind of signalling mechanism that either throws so it rejects or otherwise resolves the Promise chain.
It's also unintuitive in the sense that if you are racing you usually want to discard/stop the other Promises that didn't finish in time.
But the only real way to do this is to pass through an AbortController down into whatever networking/IO task (and your own code for that matter) is happening to "cancel" it that way.
Otherwise you just end up with the result of the fastest Promise and the rest will continue when they get a chance.
The contract is payable, i.e. will accept ether as payment but doesn't actually do anything.
From a glance looks like the withdrawal function is setup to generate the address of the scammer - through all of those obfuscated functions that have hex string slices - so ultimately only they can remove the funds.
I have one project that is very complex and for this I can't and don't use LLMs for.
I've also found it's better if you can get it code generate everything in the one session, if you try other LLMs or sessions it will quickly degrade. That's when you will see duplicate functions and dead end code.