EnPissant

5 ms·

31.
by EnPissant 6mo ago
And yet when you look at corruption perception indexes, they largely track the Hajnal line. As do so many other graphs.
32.
by EnPissant 6mo ago
There are multiple factors at play: - Among those who believe in intellectual differences among human groups, very few believe Europeans are the most intelligent group. The prevailing opinion you would find is that both Ashkenazi Jews and E
33.
by EnPissant 6mo ago
> He characterized Stanford and MIT as “mainly political lobbying operations fighting American innovation at this point” and vowed that universities would “pay the price” after “they declared war on 70% of the country.” Oh? He vowed what
34.
by EnPissant 6mo ago
git sparse-checkout
35.
by EnPissant 6mo ago
If you really wanted, you could implement this on top of a snapshot based system by mixing in git blame information (or an algorithm that is similar). It's not hard to compute that text based dependency graph on the fly, though maybe e
36.
by EnPissant 6mo ago
> 1. Dependencies. While Darcs patches would inherently "depend" on the last patch which affected the same lines, the committer could specify other patches as dependencies, handling exactly the case you described. I don't
37.
by EnPissant 6mo ago
I think patching/cherry-picking is just inherently complicated and needs intelligence applied. I don't think Pijul is going to be any better here.
38.
by EnPissant 6mo ago
From their home page: > Why? > Commutation > In Pijul, independent changes can be applied in any order without changing the result or the version's identifier. This makes Pijul significantly simpler than workflows using git re
39.
by EnPissant 6mo ago
You are a Star Wars Rebel fighting Darth Vader. Good job!
40.
by EnPissant 6mo ago
> Alright, good luck to you. I'm not really interested in talking to people who think they're lawyers for AI providers. Codex is open source, you can inspect it yourself, but let's not let facts ruin your David vs Goliath
41.
by EnPissant 6mo ago
No. Your data or any derivative of it does not leave RAM unless you are detected as doing something that qualifies as abuse, then it is retained for 30 days.
42.
by EnPissant 6mo ago
Yes. That's the rational position.
43.
by EnPissant 6mo ago
Codex just talks to the responses API with store=false. So unless the model detects you are doing something that qualifies as abuse, nothing is retained.
44.
by EnPissant 6mo ago
Not true. https://developers.openai.com/api/docs/guides/your-data
45.
by EnPissant 6mo ago
Do you use macOS? That's exactly how I feel on macOS because it is so so bad at this. Dual 4k 27" monitors on Linux with KDE Plasma near perfect.
46.
by EnPissant 6mo ago
Mate itself has not had a release since February of 2024. Sad to say, I think it's probably mostly dead. IMO, Mate held the crown for most usable Linux desktop during most of its existence, but development has slowed down to the point
47.
by EnPissant 6mo ago
> 4. persisting context across compactions > LLMs forget things as their context grows. When a conversation gets long, the context window fills up, and Claude Code starts compacting older messages. To prevent the agent from forgetting
48.
by EnPissant 7mo ago
It's very workload dependent. It certainly does more than 2% on many workloads. See https://www.phoronix.com/review/amd-ryzen-9-9950x3d-linux/10 > Here is the side-by-side of the Ryzen 9 9950X vs. 9950X3D
49.
by EnPissant 7mo ago
That assumes you have significant work to do between fetches (so you can prefetch while using the current data). With LLM decode you don't.
50.
by EnPissant 7mo ago
You do not provide any comparison to llama.cpp with mmap. You do not explain how any kind of predictor can work for MoE experts. You do not explain how prediction can even be useful. I can predict the layers used in a dense model (all of th
51.
by EnPissant 7mo ago
That's the same reason I don't like Opencode, but Codex doesn't use the alternate screen. I remember it did when it was very very new, but now it doesn't.
52.
by EnPissant 7mo ago
Forget the 80s-90s - Even California passed prop 8 in 2008.
53.
by EnPissant 7mo ago
I really like KDE plasma, it's the best DE out there once configured to mimick Gnome 2 / Mate, but I agree with you on screenshots! Also, Konsole required much configuration to be not way too busy. Other than that I don't hav
54.
by EnPissant 7mo ago
He also went on to say that Lionel Messi asked him for tips on corner kicks.
55.
by EnPissant 7mo ago
Unlike offloading weights from VRAM to system RAM, I just can't see a situation where you would want to offload to an SSD. The difference is just too large, and any model so large you can't run it in system RAM, is going to be so
56.
by EnPissant 7mo ago
Europe used to be a lot like other places a millennia ago (I typo'd century above): low-trust, dominated by extended kinship networks, different marriage patterns, etc. The transformation largely happened between the years 300-1500, be
57.
by EnPissant 7mo ago
The ability to stream weights from disk has nothing to do with MoE or not. You can always do this. It will be unusable either way.
58.
by EnPissant 7mo ago
I have a Macbook pro and a Linux machine attached to my dual 4k monitors. Fonts on Linux (KDE Plasma on Wayland) look noticeably sharper than the Mac. I don't use subpixel rendering either. I hate that I have to use the Mac for work.
59.
by EnPissant 7mo ago
More like a century of no cousin marriage / late marriage / nuclear families via bipartite manorialism has produced a high trust people. Everything you attribute to magic dirt or oppression is really this.
60.
by EnPissant 8mo ago
>> most people in the world don’t and have never lived like Europeans > Yeah, but, as it turns out with modern migration trends, the revealed preference is that they would want to, given an opportunity. Being European, I would also

More ›