Hacker News: floobertoober

New comment by floobertoober in "Mark Zuckerberg explores purchase of Washington DC property"

floobertoober — Thu, 30 Jan 2025 03:38:53 +0000

I think they mean the fees / commission

New comment by floobertoober in "Intuitive introverts lead the most successful teams: study"

floobertoober — Fri, 06 Dec 2024 18:37:06 +0000

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

New comment by floobertoober in "We’re receiving about 3,000 reports/hour"

floobertoober — Sun, 17 Nov 2024 01:53:44 +0000

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

New comment by floobertoober in "Show HN: OnAir – create link, receive calls"

floobertoober — Fri, 15 Nov 2024 15:13:03 +0000

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.

New comment by floobertoober in "Polymarket Paid US Social Media Influencers for Election Content"

floobertoober — Wed, 06 Nov 2024 00:39:01 +0000

Do you feel the same way about insurance in general?

New comment by floobertoober in "Polymarket Paid US Social Media Influencers for Election Content"

floobertoober — Tue, 05 Nov 2024 21:46:40 +0000

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

New comment by floobertoober in "Polymarket Paid US Social Media Influencers for Election Content"

floobertoober — Tue, 05 Nov 2024 21:43:46 +0000

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

New comment by floobertoober in "Using reinforcement learning and $4.80 of GPU time to find the best HN post"

floobertoober — Mon, 28 Oct 2024 23:08:15 +0000

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

New comment by floobertoober in "Using reinforcement learning and $4.80 of GPU time to find the best HN post"

floobertoober — Mon, 28 Oct 2024 22:16:13 +0000

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

New comment by floobertoober in "On the Nature of Time"

floobertoober — Wed, 09 Oct 2024 13:36:22 +0000

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.

New comment by floobertoober in "Gerald Sussman: Programming is (should be) fun (2022) [video]"

floobertoober — Thu, 13 Jun 2024 00:45:18 +0000

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

New comment by floobertoober in "Resume Tip: Hacking "AI" screening of resumes"

floobertoober — Mon, 27 May 2024 23:01:49 +0000

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".

New comment by floobertoober in "Go Browser Package"

floobertoober — Sun, 19 May 2024 00:51:45 +0000

Yeah - maybe "user agent"?

New comment by floobertoober in "Exact binary vector search for RAG in 100 lines of Julia"

floobertoober — Fri, 17 May 2024 22:33:37 +0000

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

New comment by floobertoober in "Show HN: A web debugger an ex-Cloudflare team has been working on for 4 years"

floobertoober — Fri, 10 May 2024 14:41:19 +0000

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!

New comment by floobertoober in "Generative A.I. arrives in the gene editing world of CRISPR"

floobertoober — Wed, 24 Apr 2024 01:48:04 +0000

> 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?

New comment by floobertoober in "Linus Torvalds adds arbitrary tabs to kernel code"

floobertoober — Thu, 18 Apr 2024 05:33:24 +0000

This seems very reasonable, similar to GREASE with tls.

New comment by floobertoober in "Kolmogorov Complexity and Compression Distance (2023)"

floobertoober — Sat, 30 Mar 2024 19:35:35 +0000

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

New comment by floobertoober in "Zig, Rust, and Other Languages"

floobertoober — Sat, 16 Mar 2024 01:45:03 +0000

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.

Complexity bad: An interview with Htmx creator Carson Gross

floobertoober — Thu, 14 Mar 2024 03:14:40 +0000

Article URL: https://www.infoworld.com/article/3713164/complexity-bad-an-interview-with-carson-gross.html

Comments URL: https://news.ycombinator.com/item?id=39700253

Points: 58

# Comments: 15