Hacker Newsnew | past | comments | ask | show | jobs | submit | didip's commentslogin

Does it account for errors generated from Runtime bugs which caused rerunning of prompts?

Because that’s what happened in the real world when generating a bunch of untyped Python code.


Who should own the money printer if it’s not the feds?

No one. We should be on a hard money standard. The Fed shouldn't be able to socialize the impact of bad business decisions like what was recently done with Silicon Valley Bank using the BTFP. Sometimes consequences need to be realized, even at the cost of bad downstream impact to those indirectly involved. It's the only way for more resilient systems to arise since our current system interrupts important feedback mechanisms.

So… it’s a good thing that the design emperor is poached by Meta, yeah?

Funny enough, I never suffered this because my mouse pointer has always been configured to be comically large. So I had adapt with inaccurate click area for many many years due to my own cause.


The paragraph that was started with this sentence:

> However, this technology is far too important to be in the hands of a few companies.

I wholeheartedly agree 1000%. Something needs to change this landscape in the US.

Furthermore, the entire open source models being dominated by China is also problematic.


I am super curious about tensor parallelism and the mechanisms behind how some models can activate only some of their attentions.

Early Google did not close their tags, I think it was for the sake of payload size?

That said, your linter is going to drive you crazy if you don't close tags, no?


How the mighty have fallen. A search on YouTube now pulls in 2.14 MB of HTML alone.

This is more of a problem with linters.

How does Tailwind make money?

It’s possible to have something like Express, take a look at Javalin.

But Java problem is not the mechanics, it’s that the community doesn’t want nice things.


Javalin was inspired by https://sparkjava.com/ which was inspired by Sinatra (which I think also inspired Express?).

Anyway, libraries like this were only really feasible after Java 8 because of the reliance on lambdas. Having to instantiate anonymous nested classes for every "function" was a total pain before that.


Kinda now that Java has lambdas, but still async in that disn't work as easily as JS, which is important. This is only recently starting to change with Project Loom.

I have a fun experiment for OP, since you already walked the reverse engineering route.

Why not download the most popular DAV libraries from various languages, Java, C++, PHP, etc. Regardless how ancient they are.

And then have AI like Claude to analyze and bring in the improvements to your own Go library?

I was doing something like that for Kerberos and Iceberg Rest Catalog API, until I got distracted and moved on to other things.


Author here, we don't use generative AI for software development. We've been building since 2018, and our number one goal has always been ensuring our software remains maintainable.

Did you use the 'litmus' test suite? I found it very useful when building Fastmail's (perl) WebDAV file server implementation.

There were also a bunch of fun things with quirks around unicode filename handling which made me sad (that was just a matter of testing against a ton of clients).

As for CalDAV and CardDAV - as others have said, JMAP Calendars/Contacts will make building clients a lot easier eventually... but yeah. My implementation of syncing as a client now is to look for sync-collection and fall back to collecting etags to know which URLs to fetch. Either way, sync-collection ALSO gives a set of URLs and then I multi-get those in batches; meaning both the primary and fallback codepath revert to the multi-get (or even individual GETs).


> Author here, we don't use generative AI for software development.

You don't have to use it to directly write code. You can use it just for the analysis phase, not making any changes.


I've tried that (with Sonnet 4.5 at least, not Opus) and Claude isn't good at code analysis because it's too lazy. It just grepped for a few things and then made the rest of it up.

I think the issue is mostly that it desperately tries to avoid filling its context window, and Anthropic writes system prompts that are so long it's practically already full from the start.

A good harness to read code for you and write a report on it would certainly be interesting.


Those two things aren’t mutually exclusive. It may be worthwhile to at least have Claude (or whatever LLM you favor) to look at the other libraries and compare it to yours. It doesn’t have to write the code, but it could point out areas/features you’re missing.

We know what we're missing (a lot, we didn't implement the full spec). We don't know what weird edge cases the clients/servers will have, and I would bet you decent money a LLM won't either. That's why manual testing and validation is so important to us.

I wouldn’t be so sure about the LLM not helping. The LLM doesn’t need to know about the edge cases itself. Instead, you’d be relying on other client implementations knowing about the edge cases and the LLM finding the info in those code bases. Those other implementations have probably been through similar test cycles, so using an LLM to compare those implementations to yours isn’t a bad option.

> Author here, we don't use generative AI for software development.

How close to retirement are you?


Dumb question, where do you actually buy digital copy of music these days? Or find a big library of music that you can download.

Qobuz and 7digital both have extensive libraries of high quality music.

Edit: and of course bandcamp exists but I wouldn't call their collection extensive.


Don't forget bandcamp.

Yes, but not every artist is on there. I do use it if the artist has one but that's rare.

Bandcamp, and when what I want isn’t there, qobuz usually has it. A couple of times I’ve had to buy the CD off eBay and rip it.

I buy CDs and rip them myself. That's technically buying a digital copy of the music, since CDs are digital.

i use bandcamp whenever possible

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: