croon

17 ms·

241.
by croon 11mo ago
The Wachowskis coined "red pill", that's not how it's used though.
242.
by croon 11mo ago
No. But do I believe he directs engineers to lobotomise grok everytime it frames reality in a way he dislikes? Absolutely. And do I believe that grok is used to output grokipedia? Also yes.
243.
by croon 11mo ago
I have a story quite similar to yours, both timeframe, KDE and fedora, but I did instead switch to CachyOS instead. But regarding your Nvidia issue, I did get it working on Fedora first, but it was remarkably cumbersome (I put the blame squ
244.
by croon 11mo ago
To be fair, you're either paying for hours or for output, because I assure you you are not paying staff accurately for their output. You can of course sack someone who outputs notoriously little, but if you get output exceeding your av
245.
by croon 11mo ago
Its' entire (stated) raison d'etre is that wikipedia is biased and wrong, so let's swap the question and instead ask for evidence of wikipedia sticking to wrong edits. If not, then grokipedia must exist for some other reason.
246.
by croon 11mo ago
Ironwood is 192GB, Blackwell is 96GB, right? Or am i missing something?
247.
by croon 11mo ago
I find it curious that the "Twitter Files" were never actually made available anywhere. Anyhow, I checked your link, and the source for your statement, is simply an article containing another statement, namely this: > “The same
248.
by croon 11mo ago
There is discussion on whether wer/e usage in werewolf has further/other roots, but assuming consensus landing on my GP, then yes!
249.
by croon 11mo ago
Curiously "man" is etymologically gender neutral, where wo/wif(man) denoted the female prefix for man, and the male dropped what would otherwise be were(man), seen in for example werewolf, which is gendered.
250.
by croon 11mo ago
To be fair, ignoring the drama is just adapting to changes, which is crucial in this field. Our old repo defaulted to master, our new ones defaulted to main. No time was spent on bike shedding.
251.
by croon 11mo ago
Do you verify regularly against real documentation/outside sources/subject authorities/question the output? I do, and regularly get wrong information from premier LLMs. I still use it for information retrieval because interro
252.
by croon 11mo ago
In your case you wouldn't have needed a popup/bar. In all other cases, a "Decline All" option should be a the most prominent option (or defaulted to would be fine). The current implementations are either non-compliant (i
253.
by croon 11mo ago
Frankly, the cookie banners are an example of bad enforcement. Most of the annoying ones are actually non-compliant with the regulation. I'd say that regulation is mostly fine as well.
254.
by croon 11mo ago
I'm not convinced. Perfect enforcement would be a great signal exposing bad law much more clearly, so it can be rewritten/scrapped.
255.
by croon 11mo ago
Ok, maybe you're right, but can you categorize those and show that link? Yes, the UPF classification used here likely flags some false positives and perhaps misses others, but if the methodology is otherwise sound and the link is there
256.
by croon 11mo ago
I'm as big a skeptic as anyone, but having it integrated into the OS could be very helpful to a lot of people. I myself will of course run for the hills due to the considerable trade-offs of that decision (always-online, tracking, surv
257.
by croon 11mo ago
Windows is a niche too.
258.
by croon 11mo ago
I think what you're describing is more commonly included under epistemology under philosophy, and I agree that it would be a useful background in each of those domains, but for some reason in the last few decades we have downgraded the
259.
by croon 11mo ago
> Yeah, but that’s why chess is strictly played in person for most high-level events. So is e-sports, for much of the same reasons; audience event, even playing field, auditors/judges etc.
260.
by croon 11mo ago
While I do occasionally miss it there as well, I think the main difference is that I very rarely use desktop applications for information gathering. I never "read" a desktop application, whereas that is mostly what I use a browser
261.
by croon 1y ago
Assuming they mean type 1 yields a better dialogue.
262.
by croon 1y ago
Is it possible you're conflating LLM skepticism with overall AI (which is a label applied to something every decade or so) skepticism? I don't believe LLMs with DLC will reach AGI, but I assume it will be happen at some point in t
263.
by croon 1y ago
I'm trying to be as charitable as I possibly can be, but it looks like you're arguing that brutal prisoner torture is deserved if the victim is (portrayed as) heinous enough. Is this correct or was your argument something else? I
264.
by croon 1y ago
If it's a majority of topic experts, I think it is. I work with many (might even include myself) and we disagree constantly. If we do agree on something, I'm fairly confident it's accurate and trustworthy.
265.
by croon 1y ago
Because every principled politician steps down over a mistake, so the only remaining ones are unprincipled (or the elusive spotless) politicians.
266.
by croon 1y ago
An illuminating question would be if they're sending their own kids to college or not.
267.
by croon 1y ago
I think the point is that an actual discussion having to take place is enough to worry about.
268.
by croon 1y ago
I think so, because it's generally a frustratingly frequent point having to be made to people still convinced that the covid vaccine(s, of different varieties) was somehow worse than just getting covid, which we were pretty much all gu
269.
by croon 1y ago
> there is no compounding in wide spread infection where you're guaranteed to meet a carrier several times a day, and thus there is no practical difference between 25% and 38%. You'll get it either way - say you meet 10 carrier
270.
by croon 1y ago
> they do. Covid vaccine doesn't significantly decrease infection rate [1] (that, if you think a bit about how immune system works and how covid infects, in particular is why viral load is the same in vaccinated and unvaccinated). T

More ›