SOTGO

I see "This plan may include ads" under the $8 Go tier (accessed https://chatgpt.com/pricing/ from the US) going back to January. Is the behavior change recent?

I think curving has its place. One of my math professors explained that in his opinion an effective test should differentiate performance as much as possible. The top students should score very well and the bottom students should score very poorly. If all the scores are clustered near the top (>80% for example) then it's hard to tell who really mastered the material and who just muddled through. Then, once you've sorted the students you can apply an appropriate curve. He did not have pre-defined thresholds, for each exam he would evaluate when he felt like the quality of work changed from an A to an A-, A- to B+ etc. The curves were very fair; he wasn't trying to force some number of As Bs or Fs, but it did increase my stress levels not knowing in advance how well I needed to do on each exam

"Best" may include some new posts, I actually haven't checked, but the thing that stands out to me is how old many of the posts are. Ever since Reddit made "best" the default sort on the app I notice that any new subreddit I go to will show me at least some posts from more than two weeks ago. It's really baffling that Reddit seems to think it should be preferred over "hot".

For count 3, the prediction markets consider the "bets" to actually be futures contracts, and futures contracts are regulated together with commodities (in the U.S. by the CFTC). There is ongoing litigation about whether this is the proper designation, but that is the U.S. government's position. Insider trading rules are more lax for futures than other products, but I believe this case likely does violate existing rules.

Anyone who treats Geekbench as a meaningful benchmark (i.e. not without a huge disclaimer or with other more meaningful datapoints) is not to be trusted. You can only really trust it for inter-generational comparisons within a single architecture.

Democratic maybe, authoritarian definitely

I think it's that assumption is the problem. Most social systems are predicated on having enough net contributors to provide for net recipients, but with a declining population the ratio of contributors/recipients can get small. There may be solutions to this, but current social systems will likely fail if left unchanged. That doesn't mean the only solution is population growth, but we do need to do something

I can't say for sure about the Wang terminal keyboards, but what you're describing sounds a lot like a mechanism from some IBM Model B keyboards (usually called Beamsprings). I have an IBM 5251 keyboard that has a solenoid that hammers the side of the metal case whenever you type, and I've heard that it was added as users would have been used to typewriters and wanted to know for sure when they had registered a keypress

I haven't looked at any court documents, but the WSJ article from Wednesday reported that "Last year, Google sued the anonymous operators of a network of more than 10 million internet-connected televisions, tablets and projectors, saying they had secretly pre-installed residential proxy software on them... an Ipidea spokeswoman acknowledged in an email that the company and its partners had engaged in “relatively aggressive market expansion strategies” and “conducted promotional activities in inappropriate venues (e.g., hacker forums)...”"

There was also a botnet, Kimwolf, that apparently leveraged an exploit to use the residential proxy service, so it may be related to Ipidea not shutting them down.

There can be other causes. See https://en.wikipedia.org/wiki/Illusory_palinopsia. I think mine is caused by HPPD, but I can't say for sure

Not to avoid the point of the article, but GroupMe is sometimes used for academic purposes. In the 2010s I used it in school for clubs, sports, and group activities, so that may be why it wasn't blocked.

To prove something is transcendental we would need to know how to compute it exactly, and I’m struggling to see how that would come up frequently in a physics context. In physics most constants are not arbitrary real numbers derived from a formula, they’re a measured relationship, which sort of inherently can’t be proved to be transcendental

It's probably possible to use timestamps, but I suppose you would have to handle ties in more places, with sequence numbers you only break ties once. It appears that the FIX specifications allows up to microsecond precision, but given the volume of messages it's still likely a problem. It's also easier to work with integer sequence numbers than timestamps, but that's also a small consideration.

I'm almost surprised that Gemini 3 uniquely has this problem. I would have expected that responses from any LLM that require complex math notation would almost certainly be LaTeX heavy, given the abundance of LaTeX source material in the training data. I suppose it is a flaw if a model can't avoid LaTeX, but given that it is the standard (and for the foreseeable future too) I don't know what appropriate output would look like. For "pure" mathematics or similar topics I think LaTeX (or system that represents a superset of LaTeX) is the only acceptable option.

I get that the author wanted to explore constraint solvers, but why can't you use a greedy algorithm for this problem? Sort the inventory slots by how much bundle space they consume, and insert the cheapest slots. The only way I see this failing is with multiple bundles, but in practice in Minecraft (which is admittedly not really part of the constraint problem) bundles only help when you have many distinct items but a large number of items occur only a few items. In that case it isn't hard to find combinations that fill each bundle completely by only inserting all of a given item (as opposed to inserting only part of an inventory slot) since many items will have only 1 or 2 copies.

There's enough places where em-dashes are inconvenient to type that I find it to be a reasonable indicator, particularly on the web. I don't think most people know how to generate an em-dash with a hotkey, so if I see one in a Reddit comment for example there's a high likelihood that the comment was either LLM generated or at least copy-pasted from somewhere else. Generally speaking in the past I observed a low prevalence of em-dashes on the internet except in more formal writing, so if I see an em-dash in a context where I ordinarily wouldn't expect one I do get suspicious. It's the same thing with the green check emoji, it's possible that a regular user typed it, but pre-LLM I can't recall ever seeing them, so these days I automatically assume it's AI generated content

I'd be interested to hear someone with more experience talk about this or if there's more recent research, but in school I read this paper: <https://research.cs.wisc.edu/vertical/papers/2013/hpca13-isa...> that seems to agree that x86 and ARM as instruction sets do not differ greatly in power consumption. They also found that GCC picks RISC-like instructions when compiling for x86 which meant the number of micro-ops was similar between ARM and x86, and that the x86 chips were optimized well for those RISC-like instructions and so were similarly efficient to ARM chips. They have a quote that "The microarchitecture, not the ISA, is responsible for performance differences."

I think what they meant is that the platforms are being performative by attempting to crack down on those specific words. If saying "killed" is not allowed but "unalived" is permitted and the users all agree that they mean the same thing, then the ban on the word "killed" doesn't accomplish anything.

From a summary on HHS.gov it says "De-Identified Health Information. There are no restrictions on the use or disclosure of de-identified health information." Maybe someone with more knowledge could expand on the limitations of what counts as "De-Identified" but I think that might work. I followed the reference and nothing in the CFR jumps out at me, but I'm not a lawyer so who knows.

I thought the section on finding bugs was interesting. I’d be curious how many false positives the LLM identified to get the true positive rate that high. My experience with LLMs is that they will find “bugs” if you ask them too, even if there isn’t one.

In my experience it would not be typical to use a wedge to represent a cross product. Typically a wedge is used to refer to the outer/exterior product, which in three dimensions would correspond to a bivector as opposed to the vector you get from a cross product.

I'm glad this article included an example of what happens to their eyes when they read text in dark mode. I get the same afterimages and it's incredibly disorienting when it happens and dark mode makes it way worse than normal. As an aside, does anyone know if that effect has a name?

Douyin is the Chinese TikTok equivalent. China isn't opposed to the concept of short form video, they just want to segregate Chinese users into their own app

I believe that they are required to have no more than 20% ownership by "foreign adversaries"

I think they're trying to say that you don't respond to bad behavior (China banning apps) with your own bad behavior (US banning apps). If America is opposed to the way China handles social media then we shouldn't seek to emulate them

Couldn't you argue the opposite? That is, if we are so opposed to China then shouldn't we do the opposite of them? I don't think it seems very American to change our policy to be more like the "enemy"

The HHKB variants can be nice keyboards, but they're definitely not for everyone. I have a Leopold FC660C that has the same silenced Topre switches as some of the HHKB variants and they're really nice, but the boards that come with them quite expensive for what they provide. The HHKB layout can also be a bit annoying to get used to too, which is why I use the FC660C, since it has arrow keys.

In practice I find modern custom keyboards to have higher upside than Topre keyboards, since you can build one with your preferred layout and switches. If you get a QMK enabled board you can also fully customize the functionality of the keyboard, which I find immensely useful. Unless you specifically want a 60% keyboard with light tactile switches I wouldn't recommend an HHKB, but if that's what you want it's hard to do better.

How can you verify a proof though? Pure math isn't really about computations, and it can be very hard to spot subtle errors in a proof that an LLM might introduce, especially since they seem better at sounding convincing rather than being right.

Anecdotally I have found this to be the case for the students I tutor. When I introduce a new topic I always start with worked examples, and I find that students are able to learn much more effectively when they have a reference. Poor pedagogy is also one of my biggest gripes with my undergraduate math program too, where the professors and textbooks often included too few worked problems and proofs, and the ones they did include were not very useful. What I found especially frustrating was when a worked example solved a special case with a unique approach, and the general case required a much more involved method that wasn't explained particularly well. Differential equations seems to be a particularly bad offender here, since I've had the same issue with the examples in many texts.

There's a significant contingent of people who find typing numbers with the number row cumbersome (me included). If, for whatever reason, you need to type a lot of numbers, it can be quite inconvenient to not have a numpad.