yakult

Almost all social problems created from this change in immigration policy can be fixed with money. And we'll be swimming in money. And we can use it to fix problems unrelated to this policy.

They should just replace the whole shebang with 'pay us $x to get a visa, $y for PR, $z for citizenship'.

Company wants to hire from overseas? Fine, they can figure out how to split the cost between company and employee, with zero government involvement. Rich people from the US seeking doomsday shelter? A few million dollars, a clean criminal record and less than 48 hours in the queue. Much easier than the current system.

I wonder if Google will use this feature to accidentally-on-purpose step on other adblocking extensions' toes, tech wise. Will running uBlock on top of it hurt performance like running two antiviruses?

All of a sudden, Mozilla's decision to move towards Chrome's extension model looks suspicious.

End to the arms race? Hah. The next step is adversarial neural network generated images that fools the OCR while looking relatively normal to human eyes.

This war doesn't end until computers pass the Turing test and we build a complete simulation of the end-user into the adblocker/spam filter/marketing call blocker/girl-scout-cookie-seller-doorbell-deactivator.

The line is blurred here because it's describing human behavior, and we're certainly capable of intent. A little bit of it at a time, amortized over many generations and selected via evolution, but still intent.

Do you suppose these weapons are only useable against people you don't like? Do you suppose 'the other side' is powerless to retaliate, using the exact same tools? Trump is the president now, as you have said. Half of America voted for him, and I'm sure many of them wouldn't be sad for people on your side of the trench to suffer the same treatment.

On a tangential note, this constant demand to focus on other causes whenever somebody points out something's wrong is a big part why most activism fail. Every cause gets piggybacked on by a hundred 'greater' causes that it has to expend all its resources to support. Every member has to agree with all one hundred or they can gtfo. At the end of the day nobody gets anything done, but at least you can show your friends on Facebook how virtuous you are.

I'm arguing that it's a more sophisticated and on the whole more humane system of suppressing dissent, but we shouldn't deny that it exists and serves the same functions.

If the chilling effect is the same and the curbs to your freedom is the same and at the end of the day you still cannot say what you want for fear of reprisal, does it matter whether the coercion comes from someone directly on state payroll?

If the government outsources the jackbooting to private sector, does it stamp your face less hard? What if I design a governing structure that's distributed between a small elected 'government' branch and a large, permanent network of corporations?

You can criticise the president because it's a safe subject.

If you join the neo-nazis (or any similarly unpopular group) and posts that on facebook, you will lose your job, your customers will shun you, and if you are sufficiently famous the media will make sure it stays on your record forever. You will also have no platform and no way to gather like-minded people. Whatever social media you use will ban you or seek to limit your voice constantly and in imaginative ways (cf shadowban). Whatever hosting service you use to communicate with your group will shut you down unannounced when you are of sufficient size, forcing you to move, bleeding members all the while. Paypal will refuse to process your donations.

At some point somebody will dig through your past writings/interviews and misconstrue you as a pedophile.

At the end of the day you will be just another destitute crackpot that nobody pays attention to. Why bother with the jackboots?

Criticising Trump on HN is like criticizing US imperialism in China or criticizing Israeli settlements in Iran. It is a safe subject. Everybody agrees with you, including those who pays the people in jackboots. Obviously nobody will kick your door in.

What happens to people with the unpopular opinions? Well, Berkeley.

You can criticize the devil all you want in medieval europe, too. You can attempt to spread your religious beliefs without being kidnapped so long as your king and your bishop share the same beliefs.

Try copypasting a few lines from 4chan's /pol/ in other mediums and see how far your freedom gets you.

We can start with the prior that there will probably be a similar number of trues and falses. We are mapping 23 bits of state into 12 bits, we can leave out those states where T>>F or vice versa. If the test turns out to be one of those unexpected states, pick the answer that gives the best score.

Edit: Rereading, it looks like we're optimizing the worst case, rather than average. So we're looking for a 12:23 map where at most n bits are inaccurate, minimizing n. I'm sure there's a signals alg that does this...

My pet theory: much of 'gut feelings' and intuition boils down to having our internal black box predict the behavior of other people. When you are 'average', this often simplifies to reasoning about what you would do in the same circumstances, which is relatively easy. If you are 3 standard deviations out, you'll have to compensate that consciously, which is inaccurate and difficult and slow. This is like how UI designers have difficulty to discern which parts of their UI are hard to use, while the end user could do so effortlessly.

Smart people end up having to run SmartPerson + NormalPersonVM(pre-alpha) in parallel. Normal people just run NormalPerson natively. Then they complain about bugs and slowness ('smart people are out of touch with reality!', etc.)

I would like more clarification as to what this means. Does this mean the histories are purely client-side? If that's not the case, I don't see how they can make this claim.

I remember the exact moment when it died for me. It was when they added phone-home functionality and unique player ID when it moved from 2.x to 3.0.

When I called them out on this on the forums (this was still a fairly new and egregious thing back then) the devs told me to put up with it or go use something else. So I did.

Conventional chemical weapons are not cost-effective, because you have to hit a big area and cause big effects for it to be worthwhile, since both delivery and payload are expensive. With drones you can shoot/drop grenade-sized payloads that hits just a few people, barely more effective than an actual grenade. The goal is to sap morale and force lopsided reactions (such as forcing the opfor to wear chemsuits, or invest in cleanup teams).

It's mobile, for one. You can cover and threaten a big area with a drone, which means a tiny amount of chemicals go a long way, and that solves the other main issue with chemical weapons.

Main goal would be to sap morale: a few grenade-sized bombing runs and you force everybody to wear chemsuits under the desert sun.

One thing we might see more of in the future is drone-placed IEDs. Steal an Amazon delivery drone, mix up some home explosives and you're practically there. Drones remove one of the largest barriers to IED use, which is that people get shot when placing them. Drone + sniper rifle, sneakily deployed on tall buildings, would be effective for similar reasons.

Lastly, you know what suddenly becomes a lot more effective when you don't have to worry about getting killed by your own payload? Chemical weapons.

It's perhaps more accurate to say there are 'deep work' tasks, and there are 'interrupt driven' tasks, and any job is a combination of both, in various ratios. Often you can't even tell ahead of time whether a given job is going to be mostly one or the other.

Maybe what we need is a system where instant messages are allowed, but it imposes a cost of some kind to the person initiating it, so that it won't be used frivolously. Like the proof-of-work antispam captchas, except it should take ~30 seconds and be just once per conversation (defined as ending when 5 minutes go by without message; first replies are free).

We could also add a social cost by displaying how often each user has used this feature in the past 24 hours, next to their username.

Suggestion: get rid of the torrent files and just serve magnet urls in plain text without a clickable button. This way you get rid of the knowledge of who-downloaded-what, and also save some bandwidth.

I think the closest Chinese translation to IBM's 'Think' is actually Lenovo's Chinese name - 联想. In other words, IBM got scooped three decades ago.

There are other conclusions and interpretations you can draw from the two lines of reasoning. Much of the effectiveness of dog-whistle politics is that most of what's said remain non-explicit, so let me pop that:

The reasoning that there is a small group who would refuse to fund nonwhites/women, and that has an knock-on effect on everybody else suggests:

-that there is a 'culture of racism/mysogyny'

-that we need more diversity training

-that we need to move the overton window to exclude those people with the unacceptable views (and we should do so via diversity training)

-that we need to spend money to fund the people to make the above happen.

OTOH, the reasoning that many startups lead by nonwhites and/or women got seed-funding via affirmative action despite not being strong enough to get funded on a level playground, leading to lower average quality, leading to warier investors and difficulty attracting further funding suggests:

- that there is no 'culture of racism/mysogyny', just people being rational with their money.

- that we need to wind back the irrational AA money that is distorting the market and wasting its investors' money

- that 'diversity training' is not the answer and possibly counterproductive.

-that the overton window is either fine where it is, or could be moved a bit the other way.

It is unfortunate that there is so much attached here, and I'm fairly sure most of those things are not the intended interpretations by the posters, but you can see how this is a very baggage laiden discussion. Partially it's because it echoes many, many similar discussions elsewhere, and many of those previous discussions made many people very angry at each other.

The 2.0 version will then be employed by governments to silence dissent by bleeding the energy of movements before they happen.

It's a tool that cuts all ways. There's nothing in a 'positive' social movement - however defined - that makes it less vulnerable to this sort of baiting than a 'negative' one, because it's an exploit on a close-to-universal personality flaw.

There is danger in the middle ground as well, though. Many middle grounds are factually incorrect, for example between evolutionists and people who believe god put fossils underground to troll archaeologists.

Further, as a strategy it is exploitable: say you have opinion X and I want you at X+1. I can manipulate you by loudly proclaiming X+2.

tl;dr: there are no easy heuristics here. Treat the entire communication environment as hostile and manipulative, form your opinions with great care, and double check all your sources.

The important part is that they are going to annotate those first-party sources with 'fact checking', which is slanted left [0][1] in the US and has infinite regression problems. Also, they are doing this all with humans, and the overt perception of bias means that it will attract biased people over time.

It may eventually become an attack site that cherry-picks the past to provide ammunition, because that is a lower energy state, being easier to do than actual human annotation.

[0] http://thefederalist.com/2016/12/16/running-data-politifact-...

[1] https://thefederalist.com/2016/12/29/politifact-slants-truth...

'Truth' should go in scare quotes here. If they're going to dedicate an entire program to attacking someone, and they hold the reins in what goes in the archives and what does not, why would you expect the end result to be the full, unadulterated truth? Rather than out-of-context soundbites and a highlight reel of whatever makes Trump look bad?

When they launch something that overtly flaunts their political affiliation like this, my first thought is: what will they omit?

Are we going to get 'curated' history now? Will everything they archive going forward come with a third-party annotation by a 'non partisan' fact checking group, or disappear into the ether? The latter option is far cheaper and less prong to scrutiny, after all.

One primary use I have for IA is to look for things that have been censored off their original platforms. Will I only be able to find those that align with their views going forward?

What they don't tell you is that it's only 8 hours under ideal conditions, and when the battery is new. After a year or two of heavy use it might become 4 hours. That's when the 12+ hours really shine.

While I agree with your points, I would also add that most of the people who are in the middle in the second, bifurcated market would likely be in the low end or outside the market entirely in the first market, e.g. they won't even be able to afford the meat. So they're not really worse off, the goalpost has moved.

It is best to take the pessimistic view whenever it comes to how policies are applied and how technical solutions are used. Sure, it can be used constructively, but there will always be a few assholes using it to oppress their enemies and exploit it for material gain. Much like you can predict the success of a spam filter by thinking about how it will be exploited by the worst 1%, it is instructive to think about what this tool will do in the hands of ignorant, uncaring and corrupt assholes, because that's probably how it'll turn out.

And that's why we can't have nice things, which is the common thread between telemetry in win10, trigger warnings in colleges and censorship on social media.