Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Cloudlflare builds OAuth with Claude and publishes all the prompts

github.com

Post reply on HN
Print view View original

141–150 of 552 posts

Re: Cloudlflare builds OAuth with Claude and publishes all the prompts

#141

Post by stego-tech »

realreality wrote: post #126
stego-tech wrote: post #120

Earlier quoted context omitted.

You’re not wrong (examples include the US FCC, ZA’s Telekom, ye olde Standard Oil, vertical integrations…the list goes on, and even includes modern cloud services and AI tools, since the regulations they champion are often intended to block competitors with onerous compliance requirements), but in the context of the person I was replying to, they used “regulation” very much in the same context Uber/AirBnB and the SV…

You’re wasting your time fighting a straw man. I never said all regulations are bad. The question was why plumbers are expensive. I assert that it’s not because plumbing is especially difficult.

> You’re wasting your time fighting a straw man.

Smartest thing you’ve said all day. Thanks for reminding me that trying to convince someone of something when they cannot be bothered to do research beyond first order impacts is a waste of my time.

Re: Cloudlflare builds OAuth with Claude and publishes all the prompts

#142

Post by dingnuts »

diggan wrote: post #40
infinitebattery wrote: post #23

From this commit: https://github.com/cloudflare/workers-oauth-provider/commit/... === "Fix Claude's bug manually. Claude had a bug in the previous commit. I prompted it multiple times to fix the bug but it kept doing the wrong thing. So this change is manually written by a human. I also extended the README to discuss the OAuth 2.1 spec problem." === This is super relatable to my experience trying to use these AI tool…

> They can get halfway there and then struggle immensely. Restart the conversation from scratch. As soon as you get something incorrect, begin from the beginning. It seems to me like any mistake in a messages chain/conversation instantly poisons the output afterwards, even if you try to "correct" it. So if something was wrong at one point, you need to go back to the initial message, and adjust it to clarify the promp…

Can you imagine if Excel worked like this? the formula put out the wrong result, so try again! It's like that scene from The Office where Michael has an accountant "run it again." It's farcical. They have created computers that are bad at math and I will never forgive them.

Also, each try costs money! You're pulling the lever on a god damned slot machine!

I will TRY AGAIN with the same prompt when I start getting a refund for my wasted money and time when the model outputs bullshit, otherwise this is all confirmation and sunk cost bias talking, I'm sure if it.

Re: Cloudlflare builds OAuth with Claude and publishes all the prompts

#143

Post by freedomben »

On a meta-note, it's (seriously) kind of refreshing to see that other people make this same typo when trying to type Cloudflare. I also often write CLoudflare, Cloudlfare, and Cloudfare:

> Cloudlflare builds OAuth with Claude and publishes all the prompts

Re: Cloudlflare builds OAuth with Claude and publishes all the prompts

#145

Post by dingnuts »

diggan wrote: post #78
sceptic123 wrote: post #63

Earlier quoted context omitted.

> as long as it produces the results you expect? But it's more the case of "until it doesn't produce the results you expect" and then what do you do?

> "until it doesn't produce the results you expect" and then what do you do? I'm not sure I understand what you mean. You're asking it to do something, and it doesn't do that?

if you give an LLM a spec with a new language and no examples, it can't write the new language.

until someone does that, I think we've demonstrated that they do not have understanding or abstract thought. they NEED examples in a way humans do not.

Re: Cloudlflare builds OAuth with Claude and publishes all the prompts

#146

Post by kentonv »

ZiiS wrote: post #77

Shouldn't they really have asked it to read https://developers.cloudflare.com/workers/examples/protect-a...

The secret token is hashed first, and it's the hash that is looked up in storage. In this arrangement, an attacker cannot use timing to determine the correct value byte-by-byte, because any change to the secret token is expected to randomize the whole hash. So, timing-safe equality is not needed.

That said, if you have spotted a place in the code where you believe there is such a vulnerability, please do report it. Disclosure guidelines are at: https://github.com/cloudflare/workers-oauth-provider/blob/ma...

Re: Cloudlflare builds OAuth with Claude and publishes all the prompts

#147

Post by stego-tech »

TeMPOraL wrote: post #136
stego-tech wrote: post #84

Earlier quoted context omitted.

I don’t consider that “new” research, personally - because AI boosters don’t consider that “new”. The future they hype is one where these LLMs can magic up entirely new fields of research and study without human input, which isn’t how these models are trained in the first place. That said, yes, it could be highly beneficial for identifying patterns in existing research that allows for new discoveries - provided we do…

> Though I question its value to society in burning up fossil fuels, polluting the atmosphere, and draining freshwater supplies compared to doing the same work with Grad Students and Scientists with the associated societal feedback involved in said employment activities. I'd imagine AI is much cheaper on that front than grad students, whether you count marginal contribution, or total costs of building and utilization…

You really should read the papers and reporting coming out about the sheer cost of these AI models and their operation. It might seem significantly cheaper in the context of immediate impact, but those humans provide knock-on impacts that can decrease their environmental impact (especially if done in concert), while the current crop of AI is content burning NatGas turbines and guzzling up groundwater just so a human isn’t tasked with reading a full paragraph of information, or a white paper of important content - and that’s the most optimistic view, at present.

Evaluating a technology in a vacuum does not work when trying to assess its impact, and in that wider context I don’t see the value-add of these models deployed at scale, especially when their marketing continues focusing on synthetic benchmarks and lofty future-hype instead of immediately practicable applications (like this one was).

Re: Cloudlflare builds OAuth with Claude and publishes all the prompts

#148

Post by skydhash »

ookblah wrote: post #118

AI critics always have to make strawmen arguments about how there has to be a human in the loop to "fix" things when that's never been the argument AI proponents ever make (at least those who deal with it day to day). This will only get better with time. AI can frequently one-shot throwaway scripts that I need get things done. For actual features I typically start and have it go thru the initial slog and then finish…

The one thing about concocting throwaway scripts yourself is the increased familiarity with the tooling you use. And you're not actually throwing away those scripts. I have random scripts laying around my file system (and my shell history) to check how I did a task in the past.

Re: Cloudlflare builds OAuth with Claude and publishes all the prompts

#149

Post by Powdering7082 »

dingnuts wrote: post #145
diggan wrote: post #78

Earlier quoted context omitted.

> "until it doesn't produce the results you expect" and then what do you do? I'm not sure I understand what you mean. You're asking it to do something, and it doesn't do that?

if you give an LLM a spec with a new language and no examples, it can't write the new language. until someone does that, I think we've demonstrated that they do not have understanding or abstract thought. they NEED examples in a way humans do not.

https://openreview.net/pdf?id=GTHD2UnDIb

Re: Cloudlflare builds OAuth with Claude and publishes all the prompts

#150

Post by piker »

pton_xd wrote: post #133
jes5199 wrote: post #21

I’ve been using Claude (via Cursor) on a greenfield project for the last couple months and my observation is: 1. I am much more productive/effective 2. It’s way more cognitively demanding than writing code the old-fashioned way 3. Even over this short timespan, the tools have improved significantly, amplifying both of the points above

This mirrors my experience and those I've talked to. LLM assisted coding is a way to get stuff done much faster, at a greatly increased mental cost / energy spent. Oddly enough.

The small dopamine hits you get from "it compiles" are completely automated away, and you're forced to survive on the goal alone. The issues are necessarily complex and require thinking about how the LLM has gotten it subtly wrong.

Painful, but effective?

Post reply on HN