HNHacker News
TopNewBestAskShowJobs

marquisdepolis

13 karma · joined February 7, 2021

strange loop canon
submissionscomments
marquisdepolis··on Semantic ablation: Why AI writing is generic and boring
Me too, though it feels just a grift.
marquisdepolis··on Semantic ablation: Why AI writing is generic and boring
This article is entirely AI generated, making this either a rather brilliant piece of meta commentary or just pure grift. Alas, my hopes are low.
marquisdepolis··on Iceberg I/O performance comparison at scale (Bodo vs. PyIceberg, Spark, Daft)
A performant pythonic way to read and write data into Iceberg.
marquisdepolis··on Polars Cloud: The Distributed Cloud Architecture to Run Polars Anywhere
This is very interesting, clearly there's a major pain point here to be addressed, especially the delta between local pandas work and distributed [pyspark] work!

Would love to test this out and do benchmarks against us/ Dask/ Spark/ Ray etc which have been our primary testing ground. Full disclosure, work at Bodo which has similar-ish aspirations (https://github.com/bodo-ai/Bodo), but FOSS all the way.

marquisdepolis··on Seeing Like a Network
As the author, I think this is my favourite comment I've ever received. I won't even try to argue with you about my bias!
marquisdepolis··on Seeing Like a Network
They were!
marquisdepolis··on What can LLMs never do?
Author here. This is super interesting, and while I am mostly a lurker here welcome back?
marquisdepolis··on What can LLMs never do?
This is very interesting thanks Shawn. I did email William Merrill to see his thoughts but didn't get a response yet.
marquisdepolis··on What can LLMs never do?
They are librarians, just that it happens to be the library of Babel.
marquisdepolis··on What can LLMs never do?
(Author here). I tried creating something similar in order to solve wordle etc, and the interesting part is that it is insufficient still. That's part of the mystery.
marquisdepolis··on Four Magic Words
An excellent short story about an ASI getting super-aligned through four magic words, and its aftermath
marquisdepolis··on Before Altman’s ouster, OpenAI’s board was divided and feuding
It's quite literally not. After she wrote that it would be silly if you didn't even have discussions considering this.

Actively working would require active efforts to make this happen.

marquisdepolis··on Before Altman’s ouster, OpenAI’s board was divided and feuding
In this instance, "Sam is trying to take over the Board on a flimsy basis" is an reasonable reason to remove him. Started leading discussions about whether she should be removed, is also very very far from actively working to remove her.

This is amateur hour and considering what happened she probably should have been removed.

marquisdepolis··on Some things to learn from the British East India Company's growth and demise
Fwiw I wrote right upfront that it's of course terrible. But the assumption was that now that we've acknowledged that we can also look at other aspects. I get the revulsion though, sorry!
marquisdepolis··on Some things to learn from the British East India Company's growth and demise
Yup, midjourney
marquisdepolis··on Some things to learn from the British East India Company's growth and demise
You're right! Fixed :)
marquisdepolis··on Minsky Moments in Venture Capital
This is a good exposition of a normal dynamic in the public markets. I don't think it will be as dramatic in the private markets because spirals require liquidity and regular price updates neither of which happen in private markets.

That said the idea that once you think you know how to do a strategy (whether that's in bond trading or in growth rounds) its utility decreases as everyone copies it is very true, and a key aspect of how markets get efficient.

marquisdepolis··on Ask HN: Work Gamification
Few years ago I tried doing this with fitbits to encourage exercise within my team. Worked a treat for a few months ...
marquisdepolis··on Story of a place where democratically elected Communist Parties is a real thing
Hahaha that's hilarious! One diff in Kerala is that it has relatively high HDI and no real economy, which makes for interesting dynamics. Also means that for the ruling parties bringing money in is (or was) relatively uncoupled from actual industrial policy.
marquisdepolis··on A Study of Eccentricity
Found out from a random share that this resonated and reached HN! Happy days ... Am the author and glad it sparked a discussion.

Re the eccentricity inflation, I do think the fact that we now know more about everyone and everything means the need for peacocking is much higher! So it does play a part.