floobertoober

I think they mean the fees / commission

I think those attributes might be distributed differently in USAF officers than other branches

I disagree - IMO the idea of having superdelegates is terrible. Any way of picking your party's candidate that deviates from how the candidate is selected in the general is setting yourself up for failure. IMO this is part of why the party has lost steam and alignment with their base. That does fall on the DNC, I think

You might want to investigate this further - IIRC a lot of email services retrieve images in advance server side to avoid leaking whether the client has interacted with the email. I might be mistaken here.

Do you feel the same way about insurance in general?

For me personally, I placed a bet for my non-favored candidate so that either my person wins, or I win some amount of money. I'm using it kind of like insurance or hedging. Unfortunately if enough people use it this way, it probably no longer works as a predictor

But if their having placed a bet makes your opposing bet cheaper, then it can! : )

I think that this assumes a uniform distribution of "interestingness" in the competing posts across all of those dimensions and I wouldn't be surprised if that isn't the case

Maybe it would help to use a box cox transform on the score distribution?

I've actually thought about free will in the context of wolfram's ideas before, and I like the idea that our minds are computationally irreducible - I think it is a very close analogue to free will.

The only person I'd add to that list is Simon Peyton Jones. I could watch them talk all day!

They're saying they put a poison pill in the job listing itself, so if an applicant copy and pasted the entire job listing into an LLM the generated text will have some kind of "tell".

Yeah - maybe "user agent"?

I'm not sure what is meant by "exact" here - do they describe their binarisation process at all? This seems more like an XOR benchmark than a rag benchmark, no mention of recall or other relevant performance metrics

Maybe scrub cookies and authorization headers by default, but make them an opt-in for bugs where they're relevant?

Edit: it looks like this is already the case!

> I suspect we will need to move from sequential modelling to graphical modelling to level-up again, though.

Out of curiosity, would you mind elaborating on this?

This seems very reasonable, similar to GREASE with tls.

With a Turing complete language, you can't know whether a given program eventually yields the string, or continues indefinitely

The more I've used the language, I'm starting to think that this decision makes sense in the context of Zig. Common string operations involve allocation, non-constant time access, etc. I can definitely understand how strange that sounds without the rest of the language around it as context, though.

> Please don't comment on whether someone read an article. "Did you even read the article? It mentions that" can be shortened to "The article mentions that".

From the HN guidelines

I have a hard time believing those are all honeypots. I think it is more likely that deployment configurations (with LBs, proxies, etc) are more varied than the author expects. It doesn't surprise me at all that there are overlapping and missing indicators of different services exposed on a single port

I don't think we should be regulating things from a place of "I can't imagine ever needing X; we should mandate technological controls to prevent it." We should regulate things as a need arises, based on data

Aliasing is one example of the huge-array-of-bytes abstraction being broken

It has been drastic. Some of the UI changes have been off putting to me, but I can chalk that up to taste, or their pursuing other forms of engagement. But along with those changes has come a sharp decline in quality. There are days that it seems like half of my clicks are dead! I would happily move to a new platform at this point.

I would ask for clarification about the rules - it isn't clear to me whether code that is special cased for the known 400 place names (even if it supports additional names via a slower path) would be considered valid:

> Q: Can I make assumptions on the names of the weather stations showing up in the data set?

> A: No, while only a fixed set of station names is used by the data set generator, any solution should work with arbitrary UTF-8 station names (for the sake of simplicity, names are guaranteed to contain no ; character).

What david-gpu wrote matches my understanding, based on his testimony (and his being vouched for by other DoD folks). I'm curious if there are reasons to doubt these parts of his story that I'm not aware of?

> Of course, large language models are (by definition) currently going the other direction ...

How so? Aren't the networks' weights orders of magnitude smaller than the training data?

If space is expanding, does that also mean that gravitational potential energy is always increasing, too? I'd never thought of that before