oersted

34 ms·

511.
by oersted 2y ago
Indeed, but Python is used to orchestrate all these lower-level libraries. If you have Python on top, you often want to call these libraries on a loop, or more often, within parallelized multi-stage pipelines. Overhead and parallelization l
512.
by oersted 2y ago
When it comes to old-web non-social patterns, the situation is much healthier I think. There are many blogs that are having their golden-age right now (Astral Codex comes to mind). I love what is being done in the smolweb, and the whole mov
513.
by oersted 2y ago
The thing is that before most of the Internet population was in those places, and granted the population was much smaller (with pretty good self-filtering). But now everyone is scattered in the social media meme hellscape, and those old coo
514.
by oersted 2y ago
Yeah it's a MrBeast-esque hellscape
515.
by oersted 2y ago
I think he meant it as an absurdist joke, but this is a great response! I looked it up, "gress" comes from "gradi" in Latin which directly translates to "walk". More specifically: con(pro) + gradi -> congred
516.
by oersted 2y ago
I think this site, along with the 100r.co sister site, are my favourite places on the Internet. Digital Garden indeed!
517.
by oersted 2y ago
Witty, I like it. It also highlights the subtle irony of turning "doing nothing" into yet another thing you need to optimize.
518.
by oersted 2y ago
My bad, you are right, should have looked into it better, I was too dismissive. Still I think that highlighting those charts from OpenAI is important.
519.
by oersted 2y ago
I wonder if it supports Structured Output / JSON Mode. That would make a big difference to programmatic use. I guess I will try it later when I have time.
520.
by oersted 2y ago
Kagi is most likely evaluating it mainly on deriving an answer for the user from search result snippets. Indeed, GPT-4o is plenty good at this already, and o1 would only perform better on particular types of hard requests, while being so mu
521.
by oersted 2y ago
If it really is Reinforcement Learning as they claim, it means there might not be any direct supervision on the "thinking" section of the output, just on the final answer. Just like for Chess or Go you don't train a supervise
522.
by oersted 2y ago
Definitely Snow Crash was challenging and shows its age. Some of his older books have similar bizarro vibes, I could not get into The Diamond Age the first try. Seveneves was much more accessible, as was The Rise and Fall of DODO, and the B
523.
by oersted 2y ago
You can definitely hear the difference in his more wacky NileBlue channel.
524.
by oersted 2y ago
The metal coming out of the bottom of the mold was also a frequent issue, and a more dangerous one. More pressure could be bad.
525.
by oersted 2y ago
> Microsoft Visual Studio is where you paint and sketch It's a code IDE, there's very little visual about it. It's a legacy name relating to Visual Basic, I believe it started as a code editor tightly integrated with a vis
526.
by oersted 2y ago
I mean, relative to what? Most serious crimes I can think of have a hard-to-crack black box around them. Is it any easier to investigate violent crimes like murder? Financial crimes? Organized crime? Smuggling? Fraud? Criminal negligence (l
527.
by oersted 2y ago
I've always thought that every authoritarian regime is simply a company whose business is to manage and extract value from the country that they own, like any other productive asset. Some of them are longstanding family businesses (mon
528.
by oersted 2y ago
The Factorio team has kept up a very good devlog for over a decade now. https://www.factorio.com/blog
529.
by oersted 2y ago
Correct me if I'm wrong, but I always took RTBF to mean you have the right to be forgotten by any specific service provider: that you can request they delete the data they have that relates to you, and that they forward the request to
530.
The e-graph data structure(cole-k.com)
1 points by oersted 2y ago|0 comments
531.
by oersted 2y ago
The whole thing is covered in mock latin
532.
by oersted 2y ago
Right, in that case Obsidian might not be the best choice. I'd look for Knowledge Graph editors then, for RDF or OWL knowledge bases, I don't have any specific recommendation, there are many but most are rather old. Alternatively,
533.
by oersted 2y ago
If it's text-heavy I'd recommend Obsidian
534.
by oersted 2y ago
It is somewhat old-school, but Gephi is by far the best graph visualization tool I've used that stays robust and usable at such scales (at least ~10M, but possibly a lot more).
535.
by oersted 2y ago
Well, you do need to trust that the crypto system is working as advertised. Crypto is not immune to bugs. Precisely why untrustworthy auditing, like certik, is a big problem.
536.
by oersted 2y ago
Indeed, trust is about knowing that a counterparty has a strong incentive to act with goodwill. For important things, the incentive is that they will be punished by the justice system if they don't act with goodwill. Without a real-wor
537.
by oersted 2y ago
You do have a point there, KYC/AML does tend to make a few large exchanges a trust bottleneck. But still, would you bet a billion dollars that whatever crypto system you are using has no bugs, vulnerabilities or backdoors? Are you goin
538.
by oersted 2y ago
True, but let's be realistic, there is no practical way to get crypto outside of exchanges for a normal user, and that's a serious trust bottleneck. And if you are looking to use crypto as real currency, then you'd expect the
539.
by oersted 2y ago
Yes I believe the direct translation is simply "imp".
540.
by oersted 2y ago
I know this is rather obvious and has been repeated to death but: There's just absolutely no point to crypto if you end up having to trust institutions that are fundamentally less trustworthy than mainstream financial institutions and

More ›