jerf

60 ms·

961.
by jerf 1y ago
"It's heavy, and any accident would render a chunk of Mars radioactive for quite some time." If you model the rest of the universe out of Earth's highly protective environment as "already being the result of a major
962.
by jerf 1y ago
Honestly I think that was a stretch all along. It is conceivable that two views could present a similar enough interface that the same controller could drive them, e.g., an isometric and an orthographic view on to the same object might be
963.
by jerf 1y ago
This is one of the reasons whenever the topic of math curricula comes up on HN I've tended to argue to remove some things. The system is full up at the moment and you can't really add anything without removing some things. But I c
964.
by jerf 1y ago
You can write a fast algorithm to map distances to points on a Hilbert curve on a computer: https://stackoverflow.com/questions/499166/mapping-n-dimensi... It's complicated to just glance at, and the algorith
965.
by jerf 1y ago
I don't think it was "missed" so much as "the author chose not to discuss that". Just as it isn't productive to say that the Minkowski dimension, the main point of the article, "misses" that concept..
966.
by jerf 1y ago
One of the other markers of "true MVC" I look for is that you ought to have pervasive mixing and matching of the pieces. It is common for models to see some reuse in multiple "views" and "controllers", but if a
967.
by jerf 1y ago
My guess is that it means that not only does "Henna Virkkunen, Executive Vice-President for Tech Sovereignty, Security and Democracy" write her quotes with AI without reading them carefully, but that she's also doing it with
968.
by jerf 1y ago
If you've already got it, I'm not exactly sure what you think the toil is? You just copy it places. Maybe just the MP3s for things like phones. Then use an MP3 player. If you mean the inaccuracies of the metadata, again, you just
969.
by jerf 1y ago
If you just want independence, just start collecting MP3s or CDs or whatever. I've been collecting physical music since the mid 90s and my whole MP3 collection is still under 128GBs, so I just copy it anywhere I want it now. Unless you
970.
by jerf 1y ago
It wouldn't matter if they trained on a quadrillion tokens, or another ten orders of magnitude. There's only so much information you call stuff into a given set of numbers. But once again I am reminded, never make arguments based
971.
by jerf 1y ago
You know, three years ago I would have said that I can give you a pretty good architecture fairly quickly but if you just want banged-out code I'll be beaten by someone who just plows forward for at least a couple of months... but afte
972.
by jerf 1y ago
GPT-4 is about 45 gigabytes. https://dumps.wikimedia.org/other/kiwix/zim/wikipedia/wikipe... , a recent dump of the English wikipedia, is over twice that, and that's just English. Plus AIs are expec
973.
by jerf 1y ago
I'm surprised they even pursued this line of research, though they may be considering it just as a basic territory claim that they don't have a high expectation of turning into anything. Research into "implicit parallelism&qu
974.
by jerf 1y ago
It isn't. The point is that it establishes that they are litigious. There's a tier above them (Oracle, for instance), but they're pretty up there in their willingness to head to court not just with other big corporations but
975.
by jerf 1y ago
I'm not sure what point you intend, since Factor has never been popular. No offense intended to the authors and no shade intended, but I think that's pretty clearly the case. There are other not-popular languages that build up fro
976.
by jerf 1y ago
https://youtu.be/gPIRLQZnRNk , 7:20 specifically for cubes. I knew I'd seen it before too so you nerd-sniped me.
977.
by jerf 1y ago
But it does leave room for "UserHasVerified = false, UserVerificationDate = 2025/08/25" and "UserHasVerified = true, UserVerificationDate = NULL". The better databases can be given a key to force the two fields
978.
by jerf 1y ago
They can still miscalculate the intensity of the backlash or the willingness of people to do something about it. Many such stories. "The enemy has a plan so let's do nothing" is a great way to get consistently rolled in the w
979.
by jerf 1y ago
If you're working in Go, you don't need to put together a proof of concept. Very basic project tooling in conjunction with "go mod vendor", which takes care of copying in the dependencies in locally, will do what you&#x2
980.
by jerf 1y ago
Typescript is a bad introduction to the world of static types in a lot of ways. Typescript is incredible for what it is doing, but still, at its core, laying down a static type system on top of a language ecosystem that was dynamic for year
981.
by jerf 1y ago
I think reading philosophical arguments masquerading as novels, or any sort of fiction, is an intellectual trap in general. Anything can be made to work in a fictional work simply by saying that it works. It means nothing. "Genghis Kha
982.
by jerf 1y ago
The analogy to the Dot Com bubble seems strong. Pretty much everything promised in the Dot Com bubble has come to pass. It just didn't come to pass on the time frame the stock valuations implied. And given that the time value of money
983.
by jerf 1y ago
As powerful as they are, this is something that I don't think we can trust LLMs with. With the architecture of an LLM, and the fact that at the core there is no such thing as an "out of band" with them no matter how hard you
984.
by jerf 1y ago
I have definitely found utility in modeling certain words and phrases as having a value for marketers (and by extension, politicians) that acts much like a natural resource that they can "use up". It's a tragedy of the common
985.
by jerf 1y ago
The dominant business school philosophy in the West is that 1. any reputation you have with your customers is a monetary asset and 2. therefore you should sell it for profit because it's greater than the long term expected monetary val
986.
by jerf 1y ago
The real problem is you need a real-number-valued universe for this to work, where the measurer needs access to the full real values [1]. In our universe, which has a Planck size and Planck time and related limits, the statement is simply u
987.
by jerf 1y ago
That's a question every current junior should be asking themselves. If you want to be well-paid, you need to be able to distinguish yourself in some economically-useful manner from other people. That was true before AI and AI isn'
988.
by jerf 1y ago
The information theoretic barrier is in the information content of your prompt, not the ability of the AI to expand it. But then I suppose I should learn from my own experiences and not try to make information theoretic arguments on HN, sin
989.
by jerf 1y ago
While the general question about string encoding is fine, unfortunately in a general-purpose and cross-platform language, a file interface that enforces Unicode correctness is actively broken, in that there are files out in the world it
990.
by jerf 1y ago
I agree that the level of complexity of task it can do is likely to rise over time. I often talk about the "next generation" of AI that will actually be what we were promised LLMs would be, but LLMs architecturally are just not su

More ›