ballmerpoint

I like JRPGs from the early 2000s. I haven’t settled on one that I like enough to specialize in yet. It’s a bit of a time commitment, as you can imagine.

I feel ya.

Most of my friends from high school and college are dead from random cancers, or drugs, or both.

I used to love skimming the new preprints on arXiv in the morning but now the signal is buried LLM drivel.

The one hobby I have that is kinda LLM-proof for now is speedrunning. Tool-assisted speedruns have always had their own categories for decades, and even if the LLMs or their descendants are used for routing or glitch hunting there’s still an element of human execution.

We’re down to a handful of breadtubers hiding from the Algorithm on Nebula, and some random podcasts.

Or Brazilian politics. https://apnews.com/article/brazil-elections-amazon-lula-flav...

Y’all are equally incomprehensible to us. Drastically reduced immigration, ineffective policies for incentivizing pregnancy, and no cash incentives for having multiple children (the one intervention that has been found to have a real, albeit modest impact on TFR). Degrowth under those conditions is all but certain.

When fertility rates fall below replacement rate, some amount of degrowth is all but inevitable. Probably multiple generations’ worth, assuming things remain at least nominally normal.

That few in government are readying for that and are rearranging deck chairs instead is not particularly reassuring.

Here’s an opinion poll from 2025:

https://issueone.org/wp-content/uploads/2025/10/Issue-One-Yo...

Take with a grain of salt, of course; N is about 1000 and they were contracted by what I assume was a PAC trying to pass an anti-corruption state issue in Montana.

But that seems to be a significant, bipartisan moiety that disagree with parts of SCOTUS’ opinion.

Your argument is oversimplified. Landlords may be able to extract more money from people who net benefit from UBI, yes, but they will be able to extract less money from those who take a net loss from the policy. (I’m skeptical that landlords will actually be able to extract more because I believe most low income people who can afford rent now are likely unable to afford the means of subsistence in total, but this aside doesn’t affect the argument I’m making here.)

So the inflationary effect will probably be limited to low-income housing, which is already heavily subsidized by taxpayer dollars in the US, arguably even more inefficiently than the situation under a hypothetical UBI. It will probably be coupled with a deflationary effect at middle-to-upper income housing, depending on the nature of the benefit and the structure of the taxes used to fund UBI. If the policy is structured competently, there’s no reason that it is necessarily inflationary as a whole.

Feel free to investigate the relevant case law; there are ample references in sibling threads.

Like everything else about native Julia alternatives to popular libraries, vendor lock-in is strong.

The language of 18 USC 1030 is clear.

I used to deliver Julia models wrapped in mex to customers using Matlab, this was during v0.3 if memory serves. It worked pretty well.

The Julia DSP module is pretty nice for learning. The basic objects are about what you would expect them to be, you can specify filters by their zeroes and poles, etc.

Some of the design algorithm bits are hidden away, but I’d argue that a first course in DSP probably shouldn’t cover optimization details like Parks-McClellan.

I’m still wondering why UT Austin is letting him teach a course (CS395T AI Alignment Theory) so completely outside his field of expertise (Quantum Computing).

I am also not disputing the correctness of the Lean proof. I even emphasized this in my comment: “correct but irrelevant”.

Oh, well. I suppose I should avoid getting involved in these AI threads, but now it’s about half the forum.

This shouldn’t be a surprising result. We’ve known almost since LLMs became a thing that they can “prefer” modifying the terms or context of a problem when they can’t solve it directly (what one might call “cheating” if there were any volition involved). Often that happens in a way that isn’t immediately obvious to the user.

Before it was dropping databases or deleting repositories. Now it’s subtly changing the meaning of math problems to get a correct but irrelevant answer.

I don’t think there’s any “going back” now. This is simply who he is going forward, and we need to update on that.

Not at all. Karesansui almost never exhibit Hamiltonian paths and rarely feature ninety-degree turns.

The process is intended to be meditative, but this maximalist, AI-designed GUI and the “puzzle” framing both obstruct that.

> Vault's 2018 office romance survey received 654 responses from professionals in a number of industries. Approximately 70% of responses came from the Law, Banking, Accounting and Consulting professions. A further 7% came from professionals in the Technology sector.

This survey is weak evidence based on a small number of self-reports. Population was not representative of the American work force c. 2018. Their call for submissions heavily primed the pump:

https://vault.com/blogs/general-articles/take-vaults-2018-of...

> Have you ever had a fling with a colleague, hooked up with a coworker, or met your significant other through work? We want to hear about it (anonymously, of course!) in our latest Office Romance Survey, our most popular study which has been featured by Forbes, Newsweek, and NBC.

So of course the people with strong opinions are overrepresented; the survey was more or less clickbait for marketing purposes. I suggest you find better data to support your own degree of confidence.