zarzavat

27 ms·

691.
by zarzavat 1y ago
This is all very well but exactly what type of IP is CCTV footage? It's not copyrighted. It's not patented. It's not trademarked. ...trade secret?
692.
by zarzavat 1y ago
Fermat was alive in the 1600s, long before the advent of mathematical rigour. What counted as a proof in those days was really more of a vibe check.
693.
by zarzavat 1y ago
Smart^ people don't get banned on Reddit they just get downvoted. People come to HN comments to have a constructive debate, so they are less likely to downvote for disagreement. I often upvote comments I disagree with if the point is w
694.
by zarzavat 1y ago
My father has dozens of notebooks in which he writes prose. I couldn't imagine doing the same. I want my writing to be searchable and digitalized. It's easier to edit text when it's not committed to ink. Nevertheless, I still
695.
by zarzavat 1y ago
I asked GPT-5 to OCR a table for me the other day, it hallucinated perhaps 10% of the values. This was a screenshot of a spreadsheet, with large font, not challenging except for the layout. What's interesting is that I asked it to also
696.
by zarzavat 1y ago
UK prisons are almost full. The last thing the government needs is to jail every 14 year old who sets up wireguard for his friends.
697.
by zarzavat 1y ago
You need to introduce an invasive great firewall before you can effectively ban VPNs, since there's so many different ways to hide traffic. Unlike banning porn, banning VPNs has no political value because the technically inept voters w
698.
by zarzavat 1y ago
I'm more optimistic. Open source and open weights will eat this whole space. Training is capital-intensive, yes, but so far it appears that there will always be some entities willing to train models and release them for free. All it ta
699.
by zarzavat 1y ago
You're absolutely right, but AIs still have their little quirks that set them apart. Every model has a faint personality, but since the personality gets "mass produced" any personality or writing style makes it easier to dete
700.
by zarzavat 1y ago
Why couldn't an AI do all of that? > And I’d bet that the cyborg writers using AI will outcompete the purely AI ones. In the early days of chess engines there were similar hopes for cyborg chess, whereby a human and engine would tea
701.
by zarzavat 1y ago
The assumption here is that LLMs will never pass the Turing test for copywriting, i.e. AI writing will always be distinguishable from human writing. Given that models that produce intelligible writing didn't exist a few years ago, that
702.
by zarzavat 1y ago
I've tried this approach but at some point you're just coding in a less precise language (English), and with more duplication. My current approach is to stub out the architecture I want, write the type definitions I want in full,
703.
by zarzavat 1y ago
"AI" has been writing 90% of my code, on a per-character basis, since IntelliSense came out. The question is not how many characters can a computer spit out, but whether you need a human in the loop or not. That's not going t
704.
by zarzavat 1y ago
Definitions of wealth often exclude primary residence for this reason, it depends a lot on where you live, and it's also not very liquid. There are poor people who own large houses (but can't sell for whatever reason), and there a
705.
by zarzavat 1y ago
The point is that "best model you can train in 5 minutes" is hardware dependent, the answer will be different depending on the hardware available. So it's necessarily a single-player game. "Best model you can train with
706.
by zarzavat 1y ago
Instead of time it should be energy. What is the best model you can train with a given budget in Joules. Then the MBP and the H100 are on a more even footing.
707.
by zarzavat 1y ago
> A major point of discussion when NIST announced the SHA3 finalist being Keccak back in ~2012(?) was that BLAKE1 at the time offered significantly better software performance IIRC, Keccak had a smaller chip area than Blake. Hardware per
708.
by zarzavat 1y ago
To give an example: a few days ago I needed to patch an open source library to add a single feature. This is a pathologically bad case for a human. I'm in an alien codebase, I don't know where anything is. The library is vanilla J
709.
by zarzavat 1y ago
Both modes of operation are useful. If you know how to do something, then you can give Claude the broad strokes of how you want it done and -- if you give enough detail -- hopefully it will come back with work similar to what you would have
710.
by zarzavat 1y ago
> our privacy policy says "for longer than 14 days," as mentioned elsewhere in this thread — we don't actually store them at all, but the 14 day legal guarantee is to make sure that if someone accidentally commits a log st
711.
by zarzavat 1y ago
"Could you wash the dog?" Is not a question asking whether the person is capable of washing the dog. It's a command phrased politely. "Try to wash the dog" "Try and wash the dog" If you had no prior inform
712.
by zarzavat 1y ago
You should have added a duck.
713.
by zarzavat 1y ago
I believe in the UK it works like this: if you leave the country you are not liable for CGT even if you realise gains, unless you return within 5 years then you have to pay tax on any gains you realised while you were away. With some caveat
714.
by zarzavat 1y ago
Never. It's a marketing strategy. Some percentage of users will check these files into their repos, and some percentage of repo browsers will think "what is this X.md?" Given how much money people are spending on these things
715.
by zarzavat 1y ago
The magic is the prompting/tool use/finetuning. I find that OpenAI's reasoning models write better code and are better at raw problem solving, but Claude code is a much more useful product, even if the model itself is weaker.
716.
by zarzavat 1y ago
You can still have ownership issues and leaks even with a GC, if an object is reachable from a root. e.g. object A is in a cache and it references object B which references objects C D E F G ... which will now never get collected. If A owns
717.
by zarzavat 1y ago
The documentary is real. I don't remember a house construction scene but it's been over 10 years since I watched it. It's not an uncontacted tribe, but tribes from Vanuatu and the Solomon Islands. Meet the Natives: https:&#x
718.
by zarzavat 1y ago
Hunger and satiation are two different neural pathways. Hunger is low blood sugar. Satiation is stretching of the stomach. You can still eat even with normal blood sugar if your stomach isn't full.
719.
by zarzavat 1y ago
I didn't mean it doesn't attempt to compete, I mean it doesn't actually compete. Claude code for agents, Copilot for autocomplete (depending on your editor/IDE). For single-line autocomplete, which is how I use it, pre
720.
by zarzavat 1y ago
Github Copilot and Claude code are not exactly competitors. Github Copilot is autocomplete, highly useful if you use VS Code, but if you are using e.g. Jetbrains then you have other options. Copilot comes with a bunch of other stuff that I

More ›