HNHacker News
TopNewBestAskShowJobs

JohnKemeny

1,135 karma · joined December 22, 2021

Scientific pontificate, in-pat living in London.
submissionscomments
JohnKemeny··on Bat v0.26.0
Just a nitpick. They said the classic file viewer is cat, not that cat's primary purpose was viewing files.
JohnKemeny··on Norway says 'mission accomplished' on going 100% EV, proposes incentive changes
From TFAE

Gasoline and diesel cars are now obsolete in the Norwegian new car market, with a few hundred new cars per month, while EVs represent roughly 95-97%.

The current incentive program maintains the crucial VAT exemption for EVs, but only up to a purchase price of 500,000 Norwegian kroner (approximately $49,000 USD). This move is designed to target more expensive, luxury EVs, ensuring that the incentive benefits a broader range of consumers.

JohnKemeny··on Faster CPython team layoff at Microsoft (May 2025)
Three Core Python developers (including Eric Snow and Mark Shannon) from the Faster CPython team were fired from Microsoft in May 2025.

Quote from Mike Droettboom (CPython Core Developer):

It's been a tough couple of days. Microsoft's support for the Faster CPython project was canceled yesterday, and my heart goes out to the majority of the team that was laid off. A hard day for me, but even harder for others.

Related:

Implementation plan for speeding up CPython (Oct, 2020)

245 points, 137 comments

https://news.ycombinator.com/item?id=24837309

JohnKemeny··on LaTeXpOsEd: A Systematic Analysis of Information Leakage in Preprint Archives
Most people I know simply use `latexpand` which "flattens" all files into one tex-file and by default removes all comments.
JohnKemeny··on Modern Linux tools
That's the Lindy effect: old tools like ls last because they've already lasted, while modern ones often don’t stick around long enough to.

    the future life expectancy is proportional to its current age
https://en.wikipedia.org/wiki/Lindy_effect
JohnKemeny··on Deloitte to refund the Australian government after using AI in $440k report
The Albanese government is the sitting federal government of Australia, led by Prime Minister Anthony Albanese of the Labor Party.
JohnKemeny··on How the AI Bubble Will Pop
Almost everyone who has interacted with a blockchain ended up losing money.
JohnKemeny··on Computer Vision: Algorithms and Applications, 2nd ed
There's even a HN post from almost exactly 5 years ago:

Computer Vision: Algorithms and Applications, 2nd ed (szeliski.org)

0 comments

https://news.ycombinator.com/item?id=24945823

But anyway; why not? Yes, add (2020) to the title, by all means.

JohnKemeny··on Gamebooks and graph theory (2019)
Regarding the Cycle Removal Problem. This problem is actually called Feedback Arc Set (it is a set that "hits" all "feedbacks" (i.e. cycles) with arcs, i.e., edges). Its undirected variant is called Feedback Edge Set and is actually simply Minimum Spanning Tree, i.e., a problem solvable in O(m log n) time, compared to the NP-completeness for the directed case.

While the cycle removal algorithm is a fine heuristic, it can perform arbitrarily bad.

JohnKemeny··on DeepMind and OpenAI win gold at ICPC
Do you also think that IBM Watson won Jeopardy primarily because of its ability to reason?

Or due to its massive internal database combined with superhuman button pressing skills?

JohnKemeny··on DeepMind and OpenAI win gold at ICPC
> ICPC = The International Collegiate Programming Contest. These are college level programmers, not elite competitive programmers.

ICPC finalists are very much in the world elite of competitive programmers.

JohnKemeny··on Unified Line and Paragraph Detection by Graph Convolutional Networks (2022)
It's like rai-ain, on your wedding day.
JohnKemeny··on How I, a non-developer, read the tutorial you, a developer, wrote for me
I think Java is dying.

If you want to teach algorithmic thinking, teach Python.

If you want to teach hardware and low-level systems, teach C.

JohnKemeny··on DeepMind and OpenAI win gold at ICPC
What makes you say that they are easier? Are there more people who manages to solve a problem from ICPC than from IOI?

How do you compare those?

There were at least 2 very simple problems in IOI this year.

I haven't read the ICPC problem set, and perhaps there are some low-hanging fruits, but I highly doubt it.

JohnKemeny··on DeepMind and OpenAI win gold at ICPC
I prefer not to due to privacy concerns. Perhaps you can try yourself?

I will say that after checking, I see that the model is set to "Auto", and as mentioned, used almost 8 minutes. The prompt I used was:

    Solve the following problem from a competitive programming contest. Output only the exact code needed to get it to pass on the submission server.
It did a lot of thinking, including

   I need to tackle a problem where no web-based help is available. The task involves checking if a given tree can be the result of inserting numbers 1 to n into an empty skew heap, following the described insertion algorithm. I have to figure out the minimal and maximal permutations that produce such a tree.
And I can see that it visited 13 webpages, including icpc, codeforces, geeksforgeeks, github, tehrantimes, arxiv, facebook, stackoverflow, etc.
JohnKemeny··on DeepMind and OpenAI win gold at ICPC
There is a clear difference between what OpenAI manages to do with GPT-5 and what I manage to do with GPT-5. The other day I asked for code to generate a linear regression and it gave back a figure of some points and a line through it.

If GPT-5, as claimed, is able to solve all problems in ICPC, please give the instructions on how I can reproduce it.

JohnKemeny··on DeepMind and OpenAI win gold at ICPC
I went to ICPC's web pages, downloaded the first problem (problem A) and gave it to GPT-5, asking it for code to solve it (stating it was a problem from a recent competitive programming contest).

It thought for 7m 53s and gave as reply

    # placeholder
    # (No solution provided)
JohnKemeny··on ReMarkable Paper Pro Move
What do you mean by creation.

I'm a mathematician and my "creations" are small notes, comments, and very simple figures.

I've never (besides flipping pages) experienced any noticable lag or delay.

JohnKemeny··on ReMarkable Paper Pro Move
I travel a lot and bring it on the plane. It has thousands of pages of handwritten notes, and hundreds of text books and papers.

In the airport or on the plane, I download papers I want to read or have to review.

On touchdown, my reviews are synced and I can send the annotated paper by email.

It does a bit more than a literal notepad.

JohnKemeny··on ReMarkable Paper Pro Move
> "Even Apple"

Literally the company of all time with most money

JohnKemeny··on ReMarkable Paper Pro Move
Could it be caused by dry skin?
JohnKemeny··on The wall confronting large language models
Look at their actual papers before making a comment of what is or isn't their core field: https://dblp.org/pid/35/3081.html
JohnKemeny··on The wall confronting large language models
They are physicists and will therefore talk about things that make sense to them. What they aren't: computational linguists and deep learning experts.
JohnKemeny··on The wall confronting large language models
I asked ChatGPT and it agrees with the statement that it is indeed wild
JohnKemeny··on The wall confronting large language models
Turing, Church, and even Knuth got their degrees before CS was an academic discipline. At least I don't think Turing studied Turing machines in his undergrads.

I'm saying that lots of people like to post their opinions of LLMs regardless of whether or not they actually have any competence in either computational linguistics or deep learning.

JohnKemeny··on The wall confronting large language models
As you note, some physicists do have computing backgrounds. I'm not suggesting they can't do computer science.

But today, most people hold opinions about LLMs, both as to their limits and their potential, without any real knowledge of computational linguistics nor of deep learning.

JohnKemeny··on The wall confronting large language models
He's a chemist. Lots of chemists and physicists like to talk about computation without having any background in it.

I'm not saying anything about the content, merely making a remark.

JohnKemeny··on How to Give a Good Talk
But then the claim is meaningless.
JohnKemeny··on How to Give a Good Talk
> The audience cannot care about your work until they understand the ‘value proposition’, to use still more contemporary capitalist jargon.

Hard disagree.

It is easy to imagine a problem that just sounds FUN.

JohnKemeny··on How to Give a Good Talk
> Every talk must begin with its motivation

Must is a strong word. Surely there exist good presentations that begin with something else.

← PreviousPage 3 of 15Next →