Hacker News: hopfenspergerj

New comment by hopfenspergerj in "College students drown out AI-praising commencement speeches with boos"

hopfenspergerj — Wed, 20 May 2026 16:27:14 +0000

I can be pressured to use it at work to keep up with others who use it, while simultaneously knowing that it is eroding and devaluing my skills, and wishing that we could all stop using it together.

New comment by hopfenspergerj in "Anonymous request-token comparisons from Opus 4.6 and Opus 4.7"

hopfenspergerj — Sat, 18 Apr 2026 17:54:48 +0000

You can't accidentally retrain a model to use a different tokenizer. It changes the input vectors to the model.

New comment by hopfenspergerj in "How to Use Zorn's Lemma (2008)"

hopfenspergerj — Fri, 31 Oct 2025 00:32:48 +0000

We might as well use the axiom that “you can make arbitrary choices at each stage of a transfinite induction”. It’s mostly pedagogical tradition to make students translate that into one of the classical forms of Choice in their proofs.

New comment by hopfenspergerj in "GTP Blind Voting: GPT-5 vs. 4o"

hopfenspergerj — Sun, 10 Aug 2025 12:38:33 +0000

What on earth are these questions? They don’t resemble any real use of an llm for work.

New comment by hopfenspergerj in "Pyrefly - A faster Python type checker written in Rust"

hopfenspergerj — Tue, 29 Apr 2025 16:45:01 +0000

No it is not horrible to work with. Even the “standard” settings are quite lax.

New comment by hopfenspergerj in "Meta is killing off its AI-powered Instagram and Facebook profiles"

hopfenspergerj — Sat, 04 Jan 2025 15:30:51 +0000

If you’re meta and you have to defend the AI by admitting “it’s not really intelligent and everything it says is bullshit”, that’s not a position of strength.

New comment by hopfenspergerj in "Python type hints may not be not for me in practice"

hopfenspergerj — Thu, 28 Nov 2024 01:55:05 +0000

Ruff can automatically upgrade all of the issues you mentioned to match your target minimum python version.

New comment by hopfenspergerj in "My NumPy year: Creating a DType for the next generation of scientific computing"

hopfenspergerj — Wed, 23 Oct 2024 22:48:49 +0000

Are the pandas people considering this as the default string type? Seems like it would be a slam dunk.

New comment by hopfenspergerj in "FastHTML – Modern web applications in pure Python"

hopfenspergerj — Tue, 30 Jul 2024 03:27:12 +0000

I'm looking at the very first example, and I'm a little confused. The function `home()` displays a list of messages, but they aren't passed into `home()`. Instead, `messages` is basically a global variable, and some other functions can append messages to it. Then I went looking at some more examples, and I see this pattern repeated. Is this how you're supposed to build webapps with this package? How does it isolate the list of messages for different users?

New comment by hopfenspergerj in "Evidence of price-fixing in the oil industry?"

hopfenspergerj — Fri, 03 May 2024 19:27:27 +0000

You can obviously compensate by changing other taxes to be more progressive, this is such a silly argument.

New comment by hopfenspergerj in "Ask HN: High quality Python scripts or small libraries to learn from"

hopfenspergerj — Fri, 19 Apr 2024 14:28:16 +0000

This is an example of bad code.

New comment by hopfenspergerj in "More Agents Is All You Need: LLMs performance scales with the number of agents"

hopfenspergerj — Sun, 07 Apr 2024 00:54:27 +0000

Ensemble of any number GPT 3.5 agents is less accurate than one call to GPT-4.

New comment by hopfenspergerj in "Michel Talagrand wins Abel Prize for work wrangling randomness"

hopfenspergerj — Wed, 20 Mar 2024 12:56:15 +0000

That was me on Reddit. I emailed him asking about some of his work on invariant means in the 1970s. He said “I had no taste then”, and told me what a waste of time it was.

At that point, I decided to go into data science instead of trying to get a post doc…

New comment by hopfenspergerj in "Intel Processor Instability Causing Oodle Decompression Failures"

hopfenspergerj — Fri, 23 Feb 2024 13:34:04 +0000

I’ve had instability with my 7700k since I bought it, and 16 months of bios updates haven’t helped. Maybe this latest generation of processors just has more trouble than older, simpler designs.

New comment by hopfenspergerj in "Why is everything based on likelihoods even though likelihoods are so small?"

hopfenspergerj — Sun, 18 Feb 2024 15:06:30 +0000

You work with probability density functions because the probability of observing any given value in a continuum is zero. Density functions may be reasonable to work with if they have some nice properties (continuity, unimodality, ...) The question and answers here seem to be from people that don't understand calculus.

New comment by hopfenspergerj in "I disagree with Geoff Hinton regarding "glorified autocomplete""

hopfenspergerj — Sat, 18 Nov 2023 16:43:26 +0000

I agree. We should prompt the model with the statement of the Riemann hypothesis. If the autocomplete is good, the model will output a proof.

New comment by hopfenspergerj in "OpenChat: Advancing open-source language models with imperfect data"

hopfenspergerj — Mon, 06 Nov 2023 23:01:52 +0000

“All you need is pretraining on the test set.”

New comment by hopfenspergerj in "Stable Diffusion and ControlNet: “Hidden” Text (see thumbnail vs. full image)"

hopfenspergerj — Sun, 23 Jul 2023 04:18:22 +0000

You can squint and see the letters, so I’m assuming this isn’t it.

New comment by hopfenspergerj in "Can Dell’s 6K monitor beat their 8K monitor?"

hopfenspergerj — Mon, 03 Jul 2023 23:34:28 +0000

60 is not a multiple of 24.

New comment by hopfenspergerj in "Pandas 2.0 and the Arrow revolution"

hopfenspergerj — Tue, 28 Feb 2023 13:33:28 +0000

Obviously this improves interoperability and the handling of nulls and strings. My naïve understanding is that polars columns are immutable because that makes multiprocessing faster/easier. I’m assuming pandas will not change their api to make columns immutable, so they won’t be targeting multiprocessing like polars?