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

I checked the list of pesticides in the article, and almost all of them were banned because of the effect on pollinators, not because of human health.

So using these pesticides only on products for export makes utterly no sense!


They were never used in EU, what happened was that EU exports the pesticides and then they are used in other countries and then those food products are imported into EU.

So EU makes pesticides that itself bans from being used on their own fields. Which isn't that weird, it isn't the chemical that is banned it is using it as a pesticide that is banned.


What you are describing is not hivemind, but rather paid participants. Companies pay for these "grassroots" recommendations, and Iran pays for those Jews posts.

It used to be more subtle with real people paid to post, but AI has made the quantity of it skyrocket, to the point where you can start to notice it, if you pay attention.

For example you'll see some comment about Jews, and very rapidly a bunch of upvotes. And you'll see a very similar comment elsewhere, with the same upvote pattern.

I've cut back quite a bit my participation in these types of sites once I realized just how many of the "people" I'm talking to are actually bots.

This talks about a company doing it: https://www.tomshardware.com/laptops/bots-targeting-the-r-ga...

This talks about Iran doing it: https://www.youtube.com/watch?v=Rz8whKktkQg


I don't know. I don't think it takes too many paid participants to sway a large group of non-paid participants who perpetuate the paid position.

Especially for product reviews, at the end of the day, the best product is the one you bought since most of them work well enough. I buy a new tire for my bike and buy the one reddit recommended and the next ride, buoyed by excitement for the new tire, go out and ride 1-2 mph faster than before, now all of a sudden I'm a convert. It's the best tire ever and I recommend it to all my friends.

Nevermind I don't have anything to compare it too.

This is super common in astrophotography community. You ask people what's the best camera or best mount and because they're so expensive most people only have had one, or maybe two and so everyone comes along to recommend their particular item because clearly it's better than the rest, when in fact, it's all about equal but nobody has compared. Part of that makes sense too, right? I buy a mount for my telescope from Software Bisque that's $14k and I decide to add another pier to my backyard observatory, $14k is a lot to gamble on and I know I'm happy with the mount I currently have, I'm just going to buy it again. I never tried iOptron's $7k alternative because if I hated it, I've wasted $7k


I think you have that backwards, WebRTC doesn't work, and STUN does.

I think you have it sideways. STUN [1] is the NAT traversal / "NAT hole punching" process that allows peers to discover their public IP addresses and establish direct P2P bidirectional UDP communication. WebRTC depends on STUN to establish P2P communication. You may be thinking of TURN [2] which amounts to routing traffic through an intermediary node that is visible to the two peers.

[1] https://en.wikipedia.org/wiki/STUN

[2] https://en.wikipedia.org/wiki/Traversal_Using_Relays_around_...



Perhaps the Los Angeles DA stopped focusing on low-level crimes so he could focus on the gangs running rampant within the LAPD: https://en.wikipedia.org/wiki/Gangs_in_the_Los_Angeles_Count...

The USA is consistently in the top 10 of nations with the highest incarceration rates. Prosecuting fewer "crimes" is a move in a better direction.

I live here! Could you have tried to at least find stories that supported your claim?


> Simply add random access times.

That doesn't work. Because the random times are uniformly distributed it's possible to remove it from the data by additional sampling. You do make it harder because you need a lot more data, but it's still possible to extract the signal, because the noise is uniform.


The interesting mitigation would be snapping I/O to a course clock.

You could then set it to hold the result until the next tick.

E.g. An I/O tick of 20ms, and it would only return on 20ms boundaries, then almost every SSD would look the same.

It would slow down the API a bit, but privacy has tradeoffs.


Probably still does not work. Assume a request takes X ms and let us look at what you will observe depending on where within a tick period it arrives.

If it arrives anywhere from 0 ms to (20 - X) ms after a tick, it will complete before the next tick, so the measured duration will be between X ms and 20 ms. If it arrives later in the tick period, it will miss the next tick and have to wait an additional tick period, so the measured duration will be between 20 ms and (20 + X) ms.

If you make N repetitions, you would normally see a spike of density 1 at X. With the 20 ms tick wait, you will see a uniform distribution of density 1/20 between X and (20 + X).

You would have to perform each request and then return the result exactly 20 ms after it was received in order to mask the request duration. But that just creates a new target, your timers and queues to delay the response. Or making the load so high, that requests take more than 20 ms.


The random times don't have to be uniformly distributed. Though it's enough for attackers to know the distribution to de-noisify it.

You joke, but they can partition off a small space, rent it out, and claim they are renting the entire thing to someone. It can be a housekeeper who actually has access to the entire thing.

They can also split the property into multiple independently owned properties that just so happen to be right near each other, each just under the limit.


If the housekeeper gets to live in a $5m penthouse downtown instead of a walkup with a 60+ minute commute, I'd file this as an unintended win for this policy.

Can't the attacker just man-in-the-middle to the real bank, and show the QR code to the phone?

Does the entire transaction take place on the phone? I don't think that's a good option.


Stocks are linked to entities that perform economic activity, i.e. make a profit, and share that profit. i.e. everyone can win.

Predictions markets are just bets: one person wins, one loses.


Options (especially 0DTE options) are a lot like sports betting.


And GameStop would probably like a minute (or maybe not) to discuss the tenuous connection between some underlying economic valuations and stock price.

Yes, the stock market is underpinned by real economic activity. But, be real, there's a lot of essentially betting involved too.


Cars are a very unimportant part of changing to clean energy.

The most important part is the generation. Making specific types of cars required right now is VERY premature, and will just cause backlash.

Let's focus on just one (main) thing: Clean generation of electricity. The rest will come in due course.


EVs are 3 to 4 times more efficient than ICE vehicles. So just converting to EVs is a huge efficiency gain. In fact, if we electrified industrial and transportation using existing technology we could reduce energy demand by up to 40%. And as you recommended, power generation from solar (where/when feasible) removes the need for ongoing fossil fuel purchases.


That 40% number is impossible. First of all transportation is 30% of energy, which already makes 40% impossible, second if it's 4 times more efficient then it would still use 8% to 10% of energy, making your actual savings 20% of energy at the MOST.

But your 3 to 4 times number is also not real, because the actual number is 2 to 3 - and that's measured at the outlet, if you measure starting from primary power generation they are about 2 times as efficient, not 4.

So I stand by what I said: Electric vehicles are not what matters for clean energy, what matters is power generation.


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

Search: