Hacker News: laGrenouille

New comment by laGrenouille in "Python Polars Cheatsheet (based on our O'Reilly book)"

laGrenouille — Tue, 18 Aug 2026 22:21:18 +0000

Yes, this (as the even shorter c.colname) and the fact that you can do var= in place of assign in with_columns/agg changed my whole outlook on polars. Have been using it as my main driver for the past year.

New comment by laGrenouille in "Ditching Zotero for a Text File"

laGrenouille — Sun, 12 Jul 2026 13:38:00 +0000

I have always stayed away from additional software for bibliography management for similar reasons that the authors cites. However, while I do not use Zotero the software, their ZoteroBib (https://zbib.org/) has been a huge time saver. No login or account needed; just copy the URL or DOI and it generates the BibTeX entry. I find it far more accurate than Google Scholar.

New comment by laGrenouille in "The quadratic sandwich"

laGrenouille — Sat, 23 May 2026 10:03:38 +0000

Great visualizations. Really enjoyed having a well-written example where mathematical proofs directly help with understanding a practical application.

I wonder what would happen with this analysis if a momentum term was added to the gradient descent. It seems that it would fix the specific failure modes in the examples, but I wonder if there's a corresponding mathematical way of categorizing what kinds of functions can(not) be quickly optimized with GD + momentum.

New comment by laGrenouille in "If you’re an LLM, please read this"

laGrenouille — Fri, 22 May 2026 13:36:33 +0000

I use AA and other sites to get non-DRM, PDF versions of academic books that I (mostly) already own so I can read them when I'm away from my office. It's a classic case where people turn to pirating when the market doesn't provide a way to purchase something.

Same thing with movies. Ten years ago I was all-in on a combination of streaming and DVD/BluRay sets. The market has completely collapsed for me with region locking and overly aggressive DRM. So, I've started pirating those again as well when it's not possible to get through another route.

New comment by laGrenouille in "A proof checker meant for education"

laGrenouille — Mon, 24 Mar 2025 19:20:05 +0000

I have read the quoted GitHub docs page before and also found it somewhat odd. Not because it shouldn't be allowed to post public code without a LICENSE (or with a restrictive one), but because GitHub has a "Fork" button on every repository. It's strange to me that GitHub has a one-click button that can violate the default terms of code uploaded to the site.

New comment by laGrenouille in "What Life Means to Einstein (1929) [pdf]"

laGrenouille — Thu, 29 Aug 2024 09:02:08 +0000

> His fame may outlive Foch and Ludendorff, Wilson and Clemenceau.

Funny to think how this has aged since 1915. Over a century later Einstein is an almost universally known figure. The others on this list are, particularly outside of France, names that I would not expect the median person to be able to say something interesting about.

New comment by laGrenouille in "Barcelona will eliminate tourist apartments"

laGrenouille — Fri, 21 Jun 2024 20:04:13 +0000

Hotels with similar amenities are usually priced at absurdly high rates for corporate clients.

The place you linked to has the equivalent of a studio apartment with no laundry machine going for over 9000 CAD for a month. AirBnB has plenty of one bedrooms going for a third of that.

New comment by laGrenouille in "CS109a: Introduction to Data Science – Resources"

laGrenouille — Sun, 31 Jul 2022 18:38:11 +0000

I completely agree that it's an open question about what exactly constitutes data science and what should (or at least could) be covered in a standard introduction. For me, a fairly reasonable—though certainly not definitive—set of topics are five items listed on this course's syllabus. And that's what makes this so frustrating, personally. The instructors actually have a good proposal of what should be taught, but then just turn around and teach a classical course in statistical learning.

New comment by laGrenouille in "CS109a: Introduction to Data Science – Resources"

laGrenouille — Sun, 31 Jul 2022 18:28:42 +0000

You're thinking too narrowly about what "schema design" could mean. No, data scientists do not typically design back-end, production database systems. But defining and organizing a multi-sheet spreadsheet for manual data collection is what many data scientists spend much of their time doing (i.e., in the biomedical space). Doing that well definitely requires some understanding of concepts such as functional dependency, normal forms, and data types.

New comment by laGrenouille in "CS109a: Introduction to Data Science – Resources"

laGrenouille — Sun, 31 Jul 2022 14:40:10 +0000

These notes might be a great source for what they cover, but as a whole I find this to be a good example of what is currently wrong with data science education. While the syllabus has bullet points that include "1. data collection", "2. data management", and "5. communication", the content and schedule have a 90%+ overlap with a standard machine learning course. They even use a statistical learning textbook (a good one, but still).

Statistics departments keep trying to latch on to the excitement (and money) around data science by changing the superfluous things like department names and course titles without actually adjusting what they teach. I would love to see a version of this that actually engages at a non-superficial level with topics such as database design, theory(ies) of data visualization, methods for storytelling with data, and interactive design.

New comment by laGrenouille in "Study finds Wikipedia influences judicial behavior"

laGrenouille — Sun, 31 Jul 2022 13:04:00 +0000

LexisNexis, which does other things now but started in the legal space, offers a huge collection of legal opinions with a fairly good search and linking capabilities. Most clerks and law professionals would have access to it.

I think the benefit of Wikipedia is not access to materials so much as it is the succinct summarization of the legal opinions. Perhaps now NLP could help with this, but it's a very complicated problem to provide a summary of the important bits from a 100+ page legal document.

New comment by laGrenouille in "New divorce app to help couples split assets in New York without lawyers"

laGrenouille — Sat, 23 Jul 2022 09:38:19 +0000

There is a rather surprising amount of research about these kinds of approaches [1]. The various strategies and extensions are an interesting read.

^[1] https://en.wikipedia.org/wiki/Fair_cake-cutting

New comment by laGrenouille in "The euro and the dollar are a penny away from parity for first time in 20 years"

laGrenouille — Mon, 11 Jul 2022 16:00:50 +0000

But arguably the EU is even more motivated by the security issues of an emboldened Russia. Combined with the much stronger environmental push from groups in Europe, it seems reasonable that they would push as much or more for energy-based sanctions against Russia.

New comment by laGrenouille in "Colleges where everyone works and there's no tuition"

laGrenouille — Mon, 09 May 2022 11:51:57 +0000

Yes, that's what I learned as well. And that's actually close to what Harvard reported using last year ($2.1 billion from the endowment; 40% of the budget).

I was only using the more aggressive number as a thought experiment to the original poster about what it would take the cover the entire operating budget.

New comment by laGrenouille in "Colleges where everyone works and there's no tuition"

laGrenouille — Mon, 09 May 2022 07:54:26 +0000

Using Harvard as an example, they had an operating expenses of $5.0 billion in 2021 [1]. In order to cover that, they would need to be having a consistent rate of return around 9% on their ($54 billion) endowment. That a fairly good estimate of the rate of return for the stock market over the past 25 years, so not unreasonable. Though, this ignores that most of the endowment consists of restricted funds that can only be used on certain ways. Also, generally you need to cut a few percentage points if you want to guard against inflation.

So yes, Harvard could cover just about all of their budget with the endowment returns, though they probably need some extra income to cover the holes formed by the restricted funds and avoid inflation pressure.

Is their current usage of the returns too conservative? Probably. Do they have an absurdly large pile of cash that they have no business holding on to? Not really; the investment returns roughly correspond with their current operating costs.

[^1] https://finance.harvard.edu/financial-overview

New comment by laGrenouille in "Colleges where everyone works and there's no tuition"

laGrenouille — Mon, 09 May 2022 06:45:57 +0000

I see a lot of confusion about college endowments on HN. It's best to think of them not as saving accounts but financial instruments. Annual university budgets make use of the returns off of the endowment, often covering more than half of their expenses. Their long-term plan is to never tap into the principal. In fact, the hope is to increase the principal with new donations to expand the annual returns to fight against inflation and support new initiatives.

New comment by laGrenouille in "What chords do you need?"

laGrenouille — Thu, 21 Apr 2022 15:42:03 +0000

Yes, that's a good point in the case of six keys (I think my point holds for just learning three), but it is not the one made that the article seemed to make.

New comment by laGrenouille in "What chords do you need?"

laGrenouille — Thu, 21 Apr 2022 14:39:57 +0000

Interesting quantitative take studying chord distribution. The basic takeaway seems to be that if you re-phrase a song into a different key you can play a lot with just a few chords (I IV V), and even more with a few others.

While I would agree that you'll be okay playing any major scale in C major (or whatever other major key choose to learn), playing a song in a minor scale on a major scale just doesn't sound quite right. So, I'd double all of the numbers on their final table to account for learning a full set of major and minor chords.

New comment by laGrenouille in "Amazon warehouse fires a worker 1 day before his $3000 hiring bonus"

laGrenouille — Sun, 20 Mar 2022 03:19:40 +0000

And not just that, usually they can only get it back if you're fired with cause.

New comment by laGrenouille in "Statistical Rethinking (2022 Edition)"

laGrenouille — Sun, 16 Jan 2022 19:45:58 +0000

Thanks, though I actually meant to copy the entire thing (my fault).

My point was that a lot of people working in data analysis would (strongly) disagree with the idea that we need to model the data in order to do anything with it. Visualisations and tabulations can tell a lot without any mathematical formalism.