I have found LLMs struggle with Rust's constraints - they are optimized to produce code that passes the tests, not necessarily good code. So instead of working out lifetimes and borrowing, it will be happy to copy a buffer many times without thought. This means I have to still go through line by line to review and often rewrite either by hand or with another LLM iteration.
There may be some prompting that can help with this but I suspect there is a fundamental tension between writing working code vs good code in LLMs. Go is popular for being simple, making it easy to jump in and write something fast and stable - minimizing the gap between working and good code probably helps out the LLMs a lot.
I notice Gemini is essentially a frontend to Reddit at this point. I'm guessing that was on TikTok and also Reddit, while the malicious Wikipedia edit doesn't make it through a subreddit.
Too much muscle memory to change yet but I continue to want to at least try changing my default search engine to Reddit Answers for a trial to see how it compares to Google, probably at least as good.
I think this is precisely why I don't mind it that much. I can't audit a huge codebase like a JavaScript runtime, whether the code is by a human from scratch or not. I just have to trust it as a black box.
I've seen LLMs produce terrible code indeed, but I have also seen humans produce terrible code. I haven't dug in to JS runtimes specifically but have read plenty of code in openjdk and cpython - there are many points that could be done better, but there's also no point since it's working, and keeping working code unchanged tends to be a smart decision in software engineering.
So of course the last point brings up whether it was a good idea to rewrite bun if it was working. Apparently the bun team thought the difficulty in getting changes in zig upstream meant it is. I don't intend to hold LLM code to a higher bar than human code - notably if the runtime continues to work, that is as good as I can expect from what is otherwise a huge black box of extreme programming (not that agile kind).
I agree with the sentiment and don't like Elon at all. But I think there is a case to say the Full refers to features, so not only lane change / parking, but actual driving fully. If it said Perfect FSD or even High Quality FSD, that's definitely worse, but we have to acknowledge FSD can fully drive in many cases, not enough for me to be comfortable enough maybe, but comparing to rat poison seems like a stretch.
If the latter is no, I do hope the US finally ends military aid to Israel. I really doubt it will happen, but if the record says it should be ok, then that would be really great.
I don't know of anyone accessing a database like Postgres with just the stdlib in real apps. Or maybe the API is just serving data out of a JSON file checked in the repo?
I agree the stdlib is great but the reality is production apps cannot be built using only it.
Most codebases have dependencies. Most accept dependency updates from renovate etc without much thought. Go can remain secure if goproxy has thorough security scans (I think they do but am not sure if it's unbeatable) and if the fact that any attack will be in OSS code itself, rather than packaged just-in-time in a workflow, makes it visible before significant damage. I have no idea if the latter is true but it might be. Of course NPM packages are just as visible as long as someone is scanning them so maybe not.
At this point I assume anyone with more than five years at Google is morally bankrupt to at least some degree. I don't want to paint them as bad people or anything but can't find better words for it - it's so obvious that if you stick around, you are accepting it and likely contributing to some degree. People I still talk to are generally honest about accepting it. I wonder if they will all reneg with this sort of quitting post when the time comes.
This project seems to get shared every month or so. It's really sad that HN's signal to noise has fallen this hard... While I continue to have it in my feed, I'm more and more opening up Reddit since it just feels more robust now adays.
AFAIK the pension system is quite well invested and can continue for many decades even with current trends. Where do you get that it is crumbling? There is a silly law forcing recalculation of pension payments vs inflation to happen too infrequently causing issues right now but that is probably a different issue.
Isn't the big drawback not having a swappable GPU? Perhaps that's not as important anymore but I'm not sure we've confirmed the market demand for that.
Yeah, while the world is mostly clickbait at this point, "magically makes them eligible for inclusion in the S&P500" is probably some of the best (i.e. worst).
I want to buy an index with generally fixed rules. I don't know enough of the topic to know how bad it is but if Nasdaq is literally manually tweaking SpaceX's weight, just because? Then it's not buying the market.
20 years now but rikaikun / chan were great at the time and I suspect still are for learning. Hover over the words you can't read, no corner cases really since it shows all the possible readings and meanings, not just one. I would say that extra context is useful for learning the word completely, not just being able to read some content (can just fully translate an article for that).
The best part is the feeling of hovering less over time until finally removing the extension.
https://www.irs.gov/businesses/small-businesses-self-employe...