Hacker Newsnew | past | comments | ask | show | jobs | submit | jml7c5's commentslogin

(2019)

A bit unclear - it says

> Weller’s largest study yet is due to be published later in 2019

But in the header, it also says

> Updated May 31, 2024

I'd still love to know whether there has been further research in the meantime.


And https://web.archive.org/web/20260000000000*/https://www.outs... seems to suggest 2021. I guess we'll put that above, although we had 2019 in the 2022 thread (https://news.ycombinator.com/item?id=48715020) :)

It was published early 2019, maybe late 2018.

Searching for the title of the article ("Is sunscreen the new margarine?") makes it easy to pinpoint: there is a open letter responding to the article, dated January 15, 2019. https://www.asds.net/Portals/0/PDF/LetterToTheEditor-Outside...


Ok, let's go with 2019!

Was this AI Dungeon? If I recall correctly, it was popular as a way to access GPT-3 which wasn't available publicly.


I don't recall. I am pretty sure it was a link I clicked from HN article or comment though.


Also I remember that the link was temporary...like it was inactive a day later...so it may have been a little preview with gpt-3 for HN-ers....


It looks like I misremembered a bit. The first iteration that blew up was AI Dungeon 2 (probably this link: https://news.ycombinator.com/item?id=21717022 ), which he apparently had to shut down after a day or two because it ended up costing $20,000 for weird GCS reasons.

It was briefly an easy way to access GPT-3 before the public release, but that was later and apparently not as important as I thought.


I bet that was it. Thanks!


>If there had been no major regressions, do you think anyone would have complained?

If there had been the same regressions but no AI, do you think there would be a 300 post issue full of people complaining? People are holding the update to a higher standard just because they used AI.


Previous discussion from two months ago, when this was announced: https://news.ycombinator.com/item?id=47561678


Damn, I’ve only checked the last month for duplicates.


Oh, I wouldn't worry about posting a dupe. Announcement vs release is different enough, and people are clearly interested in it. I didn't link it as any sort of admonition; there was just some good discussion there.


I loved that demo, but the problem with "monitor as a window into the world" is that monitors are relatively small and people don't sit very close to them. The FOV you obtain with most setups is disappointingly small. You need to be relatively close to a large display for it to work well. It's one of the reasons why the idea never took off in the first place, I think.


The problem is that this is either unsupported by evidence or a meaninglessly shallow claim. After all, almost every herbal remedy does something, but it doesn't mean it's actually therapeutic for some given condition.


The problem is, you’re out of date regarding the recent efforts put into studying and documenting long-known herbs.

Curcumin is a polyphenol that operates at the molecular level to disrupt multiple inflammatory cascades. It achieves this by simultaneously blocking the transcription of inflammatory genes and interrupting the enzymes responsible for generating pain and swelling.

https://pmc.ncbi.nlm.nih.gov/articles/PMC10111629/

I’ve got a couple other examples I’ve come across recently that I can additionally share.


That paper is co-authored by Bharat B. Aggarwal, who has been found to produce fraudulent research [1], and many of whose papers have been retracted. This is the guy mentioned in the New Scientist article!

Curcumin has been extensively studied, and a common observation is that it is fantastically bioactive in vitro, but tends to have zero meaningful properties when introduced into the biology of a real human being. Researchers have categorized it as an IMPS (invalid metabolic panacea), i.e. a drug whose chemical properties are an illusion, and has ended up becoming a "black hole" for scientific funding [2] [3].

The part about how it "disrupts multiple inflammatory cascades" and so on sounds terrific until you realize these are behaviours observed in vitro. The fact is that curcumin is unstable and highly reactive, so it gets torn apart and neutralized early during digestion, leading to insanely low bioavailability. Tons of compounds are anti-inflammatory in vitro. Very few actually are in the human body.

[1] https://reeserichardson.blog/2024/01/30/the-king-of-curcumin...

[2] https://pubmed.ncbi.nlm.nih.gov/26505758/

[3] https://pubmed.ncbi.nlm.nih.gov/28074653/


Are you sure that means this particular paper is bad?

Fair point, though. Good news, lots of others:

https://pmc.ncbi.nlm.nih.gov/articles/PMC9881416/#S9

https://lpi.oregonstate.edu/mic/dietary-factors/phytochemica...

Also curious how many of those detracting have observed the use of capsules,

or the difference in a person not making proper gut acids / liver function.


Thanks. I was confused for a bit, given these days you can do

    echo "-1000" > /proc/<pid>/oom_score_adj
to disable OOM killing for a process.

https://github.com/torvalds/linux/blob/master/include/uapi/l...


There's also /proc/sys/vm/panic_on_oom and /proc/sys/vm/oom_kill_allocating_task for other behaviours suggested in the comments.


Probably a copy-paste mistake while quoting the source. If anything, it makes me less inclined to believe it's AI generated.


>that doesn't smell like AI

Note that it is AI-generated. It appears to be based on a human-written outline: https://github.com/Explosion-Scratch/locusts/blob/main/artic...


No, it's not AI generated. I wrote the outline then the article. Parts of the code for the site itself (code only) are AI generated but the article I wrote by hand over the course of a day and a half or so! (Referencing my outline)


Ah, my mistake! I saw the references to writing style and assumed. I'm sorry. Unfortunately I can't delete or edit my comment any more.


Thanks for the heads-up. I skimmed through looking for the answer to the title and my radar didn't go off immediately. I'm happy the humans involved are realizing they shouldn't let the AI phone it in.


>In Florida a data center drew water in an unauthorized way and the surrounding community had issues with brown water and had to go into water rationing.

Can you cite a source for this? I can't find it anywhere.


The other comment is correct. I meant QTS in Georgia


I thought that might be what you were referring to. As far as I know the issues there were construction causing turbidity, and water used for construction accidentally going metered. Neither are really relevant to it being a datacenter. Apparently it'll use closed-loop cooling, so it's not going to draw much water in operation either.

It does seem like a canard.

https://dustinrhone.substack.com/p/on-fayetteville-georgia


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: