wild_egg

3 ms·

1.
by wild_egg 7d ago
Last time I did anything with a BERT, you had to train or fine-tune. Is that not still true? For me the cool bit is that it's all in-context learning or whatever so you can use it in any domain with zero setup. Maybe bert and co. could
2.
by wild_egg 10d ago
Goes both ways. https://terrybisson.com/theyre-made-out-of-meat-2/
3.
by wild_egg 10d ago
What's the point of switching to switch? I've only ever used checkout and not sure what I am missing out on.
4.
by wild_egg 16d ago
As an old person, what's the modern attitude?
5.
by wild_egg 23d ago
It's a limit on input tokens. So that's 3 50k requests per minute. At Cerebras speeds, that's about 5 seconds of usage per minute. I was very excited last year for their coding plan but seeing a burst of requests pulse and th
6.
by wild_egg 27d ago
Don't all the major ones have free tiers now? Pretty sure you can do this direct from the Google results pages even
7.
by wild_egg 1mo ago
Moore's Law has nothing to do with processors getting faster. Dennard scaling stopped working but Moore just slowed somewhat, not stopped.
8.
by wild_egg 1mo ago
Wait, you actually found a viable counter to The Bitter Lesson? Please say more
9.
by wild_egg 1mo ago
Why are you talking about markdown? LLMs output bytes and llama.cpp has to process those bytes correctly. What if the LLM outputs a null byte and causes a buffer overflow somewhere and starts writing into executable memory? Again, we are no
10.
by wild_egg 1mo ago
Article isn't about agents. It's about the inference engine itself being exploited by a malicious LLM output before it is ever sent to your machine or harness.
11.
by wild_egg 1mo ago
You should have just got a reset today at least. There are several sites around for tracking them now. I use this one: https://codex-resets.com
12.
by wild_egg 1mo ago
I would love to pay for Fable at full API pricing but unfortunately it is blocked from working on any of my projects. Looking forward to the end of the year when the truly comparable open models will drop.
13.
by wild_egg 1mo ago
What kind of environments are these? I can't say I have ever encountered emacs in the wild. The only people I have ever met who run emacs did it as an explicit decision because they wanted the malleability. Even having emacs just there
14.
by wild_egg 1mo ago
> Emacs Lisp — millions of people run Lisp every day without knowing it, because their editor is a Lisp interpreter. Are there seriously any emacs users who are completely unaware of elisp? Isn't elisp the entire point of picking em
15.
by wild_egg 1mo ago
You're funny
16.
by wild_egg 1mo ago
Because the system allows for it. Lisp is a far more powerful language which enables faster iteration and development. If you give that to an LLM, it is then also able to iterate and develop faster.
17.
by wild_egg 1mo ago
In this case, the better language buys you increased iteration speed in addition to performance, and that is worth a lot .
18.
by wild_egg 1mo ago
LLMs have been solid at writing Common Lisp since Sonnet 3.5 and have been near flawless since the Opus 4.5 release. The niche language thing is really not a problem at all any more. If you're working in some esolang it doesn't ta
19.
by wild_egg 1mo ago
Ah ok. When I talk about fast I am usually referring to user perception rather than clock cycles spent. Users won't notice 1ms vs 10ms but will notice 100ms or 1000ms. For me, it's a law of diminishing returns thing. I would (an
20.
by wild_egg 1mo ago
View Transitions are your friend
21.
by wild_egg 1mo ago
Who said anything about cloud? I run all my apps on my desktop and access them over tailscale. All web apps and I install on phones as PWAs. Super ergonomic. As for this: > Local software runs faster What? Why? Is this some assumption ab
22.
by wild_egg 1mo ago
It still works exceptionally well in 2026 and generally creates better and faster experiences than an SPA with far, far less code.
23.
by wild_egg 1mo ago
You want client side table sorting? It's really easy to do this with server rendered HTML. Put a link with the sort param in the column headers and be done.
24.
by wild_egg 1mo ago
Data point of 1 but I prefer all my personal software to be web based so I can easily access it from my phone, laptop, and desktop and not worry about syncing things or installing or updating when I switch devices. Also makes it trivial to
25.
by wild_egg 1mo ago
That is not at all how copyright works. If the encoding is irrelevant then virtually all code on GitHub is violating copyright as there are vanishingly small numbers of actually novel ideas.
26.
by wild_egg 1mo ago
As the sibling comment said, the existing precedent is for an entirely different scenario. If a human claims no involvement then it's fair to say they have no ownership. But if I meticulously guide an LLM to implement precisely the cod
27.
by wild_egg 1mo ago
I'm sure we are going to have to wait for some actual court cases to establish precedent but I don't really see why this is the case. If I use a fancy brush in Photoshop to paint flowers into a PNG — do I own the resulting image?
28.
by wild_egg 1mo ago
Relative few that have been around a long time? There is a plethora of 10+ year old incrementals with virtually endless content at this point.
29.
by wild_egg 1mo ago
Right... That's exactly the kind of nonsense that turns people away from ecosystems. Good luck with that.
30.
by wild_egg 1mo ago
I feel like you just directly contradicted yourself so there must be some nuance I am missing. Or is the key word "mostly"? Like I can use LLMs to help the work but still have to type most of the code myself? I have not written mo

More ›