agnishom

22 ms·

361.
by agnishom 2y ago
Remember when Omegle shut down recently? https://web.archive.org/web/20231109003559/https://omegle.co... It seems that some people are convinced that the benefits of having strangers interact with each o
362.
by agnishom 2y ago
I love Tintin too. But I recently asked two different French individuals about Tintin, and they are not a fan -- they think Tintin is racist.
363.
by agnishom 2y ago
Don't worry, it is not a real account and does not send an email to you
364.
by agnishom 2y ago
How many parameters did ChatGPT have in Dec 2022 when it first broke into mainstream news?
365.
by agnishom 2y ago
I have never tried solving a jigsaw puzzle. I have a question: Given a solved jigsaw puzzle, is it possible to swap the position of two pieces such that it fits but does not preserve the picture?
366.
by agnishom 2y ago
So this is the modern day equivalent of an O'Reilly book with the title "Mastering LLMs"?
367.
Haskell Discourse: Controversy Surrounding Anduril Industries(discourse.haskell.org)
2 points by agnishom 2y ago|0 comments
368.
Reachability Analysis of DNS(arxiv.org)
12 points by agnishom 2y ago|0 comments
369.
by agnishom 2y ago
Fascinating. I did not know that you can create presentations with Typst
370.
by agnishom 2y ago
It's a good thing there are Coq proofs. However, I would still like to see the paper though. A paper is supposed to be more than just a syntactic theory. Typically there is some motivation included as well.
371.
by agnishom 2y ago
In this case, they are not really democratizing metatheory. Am I understanding metatheory any better now? I do not know what it would even mean to democratize functions. Functions just are. Can you democratize the number 5?
372.
by agnishom 2y ago
The homepage says "Democratizing Functions" and "Democratizing Metatheory". Whatever that means, I have a strong feeling that this is an abuse of the word "democraztizing"
373.
by agnishom 2y ago
I googled that, and I found that it is a sister conference of POPL: https://popl25.sigplan.org/details/pepm-2025-papers/5/Typed-... I was inclined to think that Barry Jay is some kind of crackpot
374.
by agnishom 2y ago
No, I do not work at a bank. What do you mean?
375.
by agnishom 2y ago
Thanks, I did not know that you could use Canva for presentations. However, it is still proprietary software. (Worse, it lives on the web)
376.
by agnishom 2y ago
Looks interesting. I am on the hunt for clean tools for producing presentations. I really like Powerpoint, mainly because of their animation and vector editing features. However, I don't want to keep using a proprietary tool.
377.
by agnishom 2y ago
I don't really understand how this would work. Writing long paragraphs to prompt the AI is much more tedious than writing a few bullet points for the slides. If you need the AI to help you brainstorm a good narrative, that is a differe
378.
by agnishom 2y ago
Can you please share with us what is so bad about this feature?
379.
by agnishom 2y ago
What is wrong with the requests for money? Wikipedia ads are the least intrusive of ads.
380.
by agnishom 2y ago
Surprisingly progressive policy. I know this is mostly tangential, but I think that this should be the way humanity should benefit from technology. Better technology shouldn't just exist to make capitalists richer, but it should exist
381.
by agnishom 2y ago
I agree. If I made 20% less money and worked 20% less at my main job, I would still need to make up for that 20% of my income. That could be a difficult position to be in.
382.
(Coq based) Verified Matching of Regular Expressions with Lookarounds(github.com)
2 points by agnishom 2y ago|0 comments
383.
by agnishom 2y ago
I am not sure Markdown is a curse because of the gap that it creates in the chart. But I agree with one thing: we do need to explore interesting DSLs for producing beautiful documents.
384.
by agnishom 2y ago
That made me chuckle. It is like showing how good my programming language by building an interpreter.
385.
by agnishom 2y ago
Nothing, but arguably a language with dependent types is more Haskell than Haskell
386.
by agnishom 2y ago
It doesn't have great support for Dependent Types
387.
by agnishom 2y ago
> I’m interested in building stuff, not sitting near my pond on a warm summer day, thinking if TypeFamilies + DataKinds would be better than GADTs for making illegal states unrepresentable. I feel differently. I would rather sit by the p
388.
by agnishom 2y ago
I feel that the main thing that can change it is having a plethora of examples. Also, I think math has much better examples than programming.
389.
by agnishom 2y ago
you mean the usage of Crystall-Ball as notation for oracle?
390.
by agnishom 2y ago
I feel that the LLM community has been very interested in Retrieval-Augmented-Generation, recently. But as far as I have seen (not very far), they are not very formal methods oriented.

More ›