Cloudflare reports 3.7% FWIW. JS based trackers tend to be skewed down by tracking protection so google analytics will consistently underreport.
Wikimedia reports 6.2% so clearly it's also about audience. Since that's a tech blog with a more savvy audience, my guess is it'd be even higher on the linked story. Nonetheless they are still wrong about the firefox support.
I have a perfect score in google postmaster tools on dkim, spf and 0 spam reports. They still send all emails to spam. They will even do this if you are replying to the user who contacted you. Same experience for my friend running an indie festival. These days you pretty much have to alert gmail users by some other means (like if signing up on a website) that they have to check spam and whitelist you since it appears gmail would rather block legit emails than allow a single possible spam through.
"Massive amounts of Russian propaganda — 3,600,000 articles in 2024 — are now incorporated in the outputs of Western AI systems, infecting their responses with false claims and propaganda."
in 2024.. hopefully filters when training have gotten better, but maybe not..
The New York times publishing an article you don’t like is not “communist infiltration”. In this country, we have a right to free speech, including free press.
My point is that Russian propaganda from the 20th century made its way into intellectual water supply just like Russian propaganda from the 21st century. The propaganda is different, but the mechanism has parallels.
So it is communist infiltration of the variety where there were no communists and there was no infiltration, just somebody read a book or whatever and then wrote about it?
You keep changing the words. Above you said “Soviets.” I don’t think they’re soviets. But I do think they’re communists. So “communist infiltration” is accurate.
Why would I keep asking about Soviets after you clarified that you meant Russian propaganda in your first response?
I’m no communist infiltration expert or anything but do you have something to support your assertion other than suggesting people read “How to Parent Like A Bolshevik”?
Well considering the USA pumps out the largest amount and widest reach of propaganda on the earth, and has for decades, I'd argue it makes more sense to incorporate all the propaganda - omitting any only strengthens/biases the ones included.
That said, your focus on Russia specifically just smells like red scare paranoia, one of many successful psyops from the USA.
Reminder that an intelligent entity does not believe or repeat something it has read even a million times: reminder that now that LLMs exist, we must find the real recipe ASAP.
I wrote «an intelligent entity», very clearly - not "somebody supposed to be intelligent".
Of course there can be more examples of intellectual failures than of intellectual accomplished ones. And again, this calls for the urgency of finally injecting supplies of intelligence into the systems.
What about other hostile propaganda? Do you think the completely deranged, self-serving, lying, dangerous shit that regularly comes out of DC is being incorporated in those training materials?
This regime and its enablers have done more quantifiable harm to more people and institutions in America than anything Russia has, and one of its favorite tools is propaganda.
(I do find it interesting, btw, that a few seconds after post this did get an immediate downvote to 0 - it didn't last, but it makes me wonder if there are bots monitoring the HN firehose for anything critical of country X)
Possible, certainly. On the one hand the HN audience is not very big in the grand scheme of things, but they probably do skew to more influential and likely to be politically active than the average person. If it's worth doing, it's almost certainly being done.
There are easier explanations. A hundred people are checking and occasionally the page, with new posts added every couple of minutes: clearly some will meet a post at "0 minutes", read it as fresh, and possibly react. I have seen it.
sure, that's simplest scenario, just seemed odd that was the first vote. and it's not like writing or running a bot is hard and political stories seem like they'd be a popular target to me.
As disturbing as "you would not steal a " in the times of pro-copyright brainwashing. What is published is there for people to read, and most societies already decided through libraries: for free, with charges on the State.
> [all caps]
Pls reread the guidelines. Check the page visually, see why. Edit the post pls. while you can.
No idea how it compares, but my local Lidl was selling a smart ring for $15 so I decided to give it a shot. I assume half the measurements are "for entertainment purposes only" (glucose, blood pressure, stress [heart rate variability proxy?], blood oxygenation [might be ok with my skin tone, not sure I'd trust it]) but the others seem fairly accurate. Again, no idea how it compares to the pricier ones, but seems to match my pulse measured at neck and what I remember of my nightly sleep pretty well.
And yeah, I'm certainly not going to worry about it getting scratched or slightly wet. Actually, it got dropped and stepped on, and is slightly dented but still seems to run fine - the epoxy sealant for the electronics seems intact and while I was a bit concerned for the battery, that's been a while with no issues.
And, while I have no idea whether the app is quietly collecting data and pushing it to a server in china (if it is, it is doing it in a way that is sneaking past the firewall), at least it does not require creating an account to use.
So in that sense it is maybe more private than oura or samsung ?
They do mention that in the "Concurrent Work" section.
Our effort began on September 1st after hearing a rumor which we later realized was related to Levent Alpöge, an Anthropic employee, and Tristan Buckmaster, a math professor at NYU. After the completion of our full project and Lean verification (on September 6th), believing from the rumor they also had a solution of Navier–Stokes, we reached out to them to offer a concurrent release of our result and to recognize their priority in a joint announcement. At that point we found out that they had a resolution of the forced Euler problem. In these discussions we offered them visibility into all of the prompts we used and later to see the proof. We recognize the priority of their work on forced Euler and congratulate them on their remarkable mathematical achievement.
That seems like it could also mess with mouse selections if the text moved when selected. maybe even causing endless jitter.
I kinda wonder if "page scrolled 'cause X was selected" type data extraction attacks might play a slight role too. Although I suppose if you've managed to inject CSS all bets are off..
Hedgewars had to lock down its bugzilla and repo too, for the exact same reason. So far the really simple method used (basic auth with the creds posted on the main site) seems to have resisted the scrapers. Presumably if they are too stupid to scrape correctly they are also too stupid to incorporate a trivial auth.
Wikimedia reports 6.2% so clearly it's also about audience. Since that's a tech blog with a more savvy audience, my guess is it'd be even higher on the linked story. Nonetheless they are still wrong about the firefox support.
reply