eptcyka

5 ms·

1.
by eptcyka 6d ago
No, most often I listen to my own collection.
2.
by eptcyka 6d ago
I wouldn't be surprised if they are optimising for producing more code, because in the long term, more existing code means they can sell you more tokens to maintain it.
3.
by eptcyka 6d ago
You do realize that letting the LLM produce more output means that maintenance will be more expensive? I can easily see a world where claude and gpt are producing more tokens to sell you more tokens.
4.
by eptcyka 6d ago
I never paid for the licence for I never used iPlayer and the investigators were welcomed into my house to take a look at the chromecast dongle.
5.
by eptcyka 6d ago
I stream on youtube to discover and purchase LPs and FLACs on bandcamp and tickets for live shows to support the artist and not the rent seekers.
6.
by eptcyka 8d ago
Thank you for your input - it genuinely is very interesting.
7.
by eptcyka 8d ago
Do you have any thoughts on the question you posed? Amdgpu and mesa is working out fairly well for amd, but yes, there is also still a proprietary userland for it too. Intel I think is using mesa code for their Windows drivers too. How are
8.
by eptcyka 9d ago
I do not understand why vendors believe their software stack is a moat. It is an obstacle to me.
9.
by eptcyka 9d ago
What about support in mesa?
10.
by eptcyka 11d ago
Those services aid will power instead of health directly.
11.
by eptcyka 11d ago
Going vegan 1 day a week is trivially easy - if everyone does this, the emission reduction is not trivial. We need not have to go full militant vegan to make a dent. And I totally agree with energy/transit/real estate planning bei
12.
by eptcyka 13d ago
My practical approach is to have a microvm for firefox and a bubblewrap “jail” for a firefox instance that runs on the host. Of course, the config options mentioned here are good too.
13.
by eptcyka 14d ago
Neither is google.
14.
by eptcyka 14d ago
I never touch a mushroom I am not 100% certain is safe. So far I’ve had a 100% success rate in not poisoning myself or others. Wouldn’t recommend trusting an LLM either.
15.
by eptcyka 15d ago
I have no proof, only personal experience interacting with commenters on HN. No other tech company gets as much positive support when exerting its soft power under the guise of (bullshit) technicalities, e.g. all the issues with EU digital
16.
by eptcyka 15d ago
What specifically is the dispute about factual accuracy at the root of this discussion?
17.
by eptcyka 21d ago
How many times do distinct letters appear in the word mamma? Do not write code to produce an answer.
18.
by eptcyka 23d ago
Sure, cynically, you can use the argument that we can’t just limit the influence of megacorps because the problem is more complex than that. Sure, but limiting them will still help. Why not do the obvious thing that can still help? There ar
19.
by eptcyka 23d ago
If you were to interpret the comment you’re responding to charitably, you’d understand that electronics components are a minuscule minority of all of Chinese exports.
20.
by eptcyka 24d ago
Why would 17 year old do something that only brings them money? I do not think Mr Graham here is advocating for the path that makes most money as a result of learning how to train a model. I assume that tinkering and learning about LLMs is
21.
by eptcyka 24d ago
We’ve received many an email urging us to reach out to Apple via Feedback Assistant. We of course have, multiple times. We’re told that we will strike gold if we just resubmit a bug report to some other part of the company. Yet, the APIs ar
22.
by eptcyka 25d ago
Ye, but signals are cumbersome and a foot gun. Should a library ever install signal handlers? What about languages with runtimes? Ever read the spec on how go handles signals differently depending on how it is compiled?
23.
by eptcyka 26d ago
> You're the pilot, not the compiler. This to me reads like an LLMism.
24.
by eptcyka 28d ago
Pretty sure cameras can be used to read punchcards too.
25.
by eptcyka 29d ago
I wonder if they're looking to achieve easy speedups. I can see lots of value enabling performance gains where normally people wouldn't bother because it's too much effort. I don't think this will take away work from tho
26.
by eptcyka 1mo ago
Every day. Not for knowing how to get to a destination necessarily, but to get routed better.
27.
by eptcyka 1mo ago
Isn’t the conclusion of this paper rather bleak for openai and anthropic? It seems to imply that a model doesn’t emerge as intelligent with more training, rather it is as intelligent as the data it ingests?
28.
by eptcyka 1mo ago
It mirrors the C interface.
29.
by eptcyka 1mo ago
I am using Qubes already ;^)
30.
by eptcyka 1mo ago
Whilst I like getting more control of my hardware, I really prefer if I can contain arbitrary software - sandbox it and compartmentalise it. If arbitrary processes can own your whole system, is anyone truly an owner?

More ›