You don't AI to implement an automated system. As you said, it all works fine with the debit card system. The problems are not so much with ACH itself as they are with the implementation (manual processing and infrequent polling).

The advantage of sticking with ACH is that it's already widely implemented. Rolling out truly automated backend handling and increased polling rate can be done gradually, bank by bank, while not interrupting the operation of banks that haven't upgraded yet.

This whole thing reads like, "We need a new system of transportation! The automobile sucks because we've all agreed to only use it once a day."

Perhaps I'm missing something, but why isn't ACH ("Automated", by the way, even though I like "Atomated" better!) up for this task? Even if the upgraded ACH isn't instantaneous, it could at least be faster. Increase the polling rate and the transaction handling and the whole system is faster.

We already have a system in place to handle money transfers. It could use some tweaking, sure, but kludging a replacement based on debit cards isn't the right way to get to a better system.

The fact that we're not perfect is not a reason to avoid striving for perfection. Understanding the mechanism of human nature, with all of its irrational twists, is part of the path to bettering ourselves.

The fact that most people don't even accept that an emotional attachment to a position that lacks evidence is an issue indicates that we don't even understand our own motivations and thought processes very well. It's alright that we're not perfectly logical and our irrational behavior probably benefits us in significant ways, but it's important that we be able to recognize where the motivation for our (lack of) reasoning originates.

Traffic deaths aren't random, even if they aren't intentional. Nearly every traffic death can be traced to a specific and often avoidable cause. Addressing a minute fraction of those causes will have a dramatic effect on the number of people who die in the US every year.

On the other hand, if your opponent's most successful attack ever can't be distinguished from year to year variations in the death rate of Americans, spending any significant energy fighting him is a waste. We could have a 9/11 attack every single day for hundreds of years and still not deplete the American population. This is an ant-bite of a threat and deserves an ant-bite appropriate response.

Fatigue has set in. You've been here long enough to know that we have made a fuss throughout the years. Nothing at all has ever come of it, so we gave up complaining. Relatively simple it is, too. Many of the gripes about SlashCode of old have been fixed over at SoylentNews.

Anyway, asking for improvements now is dangerous... we might end up with Beta!

That's true, but that's a special case of the more general:

The only difference between almost any person involved in governing and a totalitarian is that the former says, "You are free to do whatever you want, as long as what you want is what I think you should want."

Despite the lofty goals claimed by almost any person of any party, whether running for office or just voting, the main reason that people get involved in government is to assert control over others. There are positive and negative outcomes of their actions, but every single one of these people think that things would be better if only they were king. The only tool that government has is coercion; political differences come down to how that tool is to be applied.

Most of the rest of those who actually want to reduce the power of government either still want the government involved where "what I think you should want" is concerned, or have other non-governmental means to effect coercion.

If you see any political party in the US as not fitting into that statement, it's just because "what [you] think [they] should want" and "what [they] think you should want" are aligned. Your liberal adversaries see themselves as just as rational and correct as you see yourself.

You can! Every passenger's running phone, and some computers that are awake, are sent through the baggage scanners at every airport. Even more impressive, the computers that run them are next to the poorly built and maintained scanners all day every day.

Someone dying, even voluntarily, on a mission would cost NASA way more than $375 million. Between administrative, investigative, and PR costs, that is not a risk worth taking. Astronaut deaths cost significant money for a long time afterward in budget considerations alone.

I installed solar panels on my house for exactly this reason. It's really just a whole house UPS, but once the batteries, chargers, and inverters are installed, adding panels is a minor extra cost. No more power outages. Saving money (and not giving as much to the power company) is a bonus.

I'm not sure that intelligence, beyond a certain baseline, really enters into the formula for creating long lived social structures. For the individual, there is no tangible benefit to creating structures that will last more than a few generations. After a few generations, as you point out, they will require the efforts of other people to keep alive, so there is little that the originator can do to ensure that the structure survives.

The most rational course of action for all individuals involved in a society is to maximize their personal benefit and plan to pass that benefit on to their offspring. Tangling with those who are successful at maximizing their benefit has real negative consequences, so the most rational action for the downtrodden is to take as much of the scraps as possible or jealously guard what you have while staying out of the eye of others.

The ambition to create structures that will outlast you and your offspring comes from irrational motivations. Historical drivers for this are abstract concepts like duty, fairness, or religion. There is no tangible benefit to creating these structures and there is often great personal cost. As there is no assurance that the structures will even survive, no matter how well you craft them, it's hard to say that building them is an intelligent choice.

I think that improving the lot of humanity as a whole is a noble goal, but I don't think that it has anything to do with intelligence or rationality. History is littered with people who have tried, and sometimes succeeded, to do so at great personal cost. Often their success only lasted a few generations, if that, before being undone by others. What real benefit to we get from creating these structures and how does that benefit weigh against the costs required to build them?

Those negative characteristics are only negative in the context of forming long lasting social orders (really only in the context of forming long lasting egalitarian social orders -- dynastic empires last longer than most of our social structures). They are extremely positive in the context of the affected individuals and in the case of avarice, they are beneficial to the affected's offspring (and their offspring, and so on). There is extreme benefit to be had from sabotaging the social order and norms.

We humans are intelligent enough to form effective and equitable social structures, we just don't have the collective stomach for removing the saboteurs from our society. So we/they continue to undermine every system we devise.

Of course it sends the keys to WhatsApp! If you install the client on a second phone, it just works, right?

So they're either:
1) generating a new key on each device and encrypting all incoming messages to every client's public key (or just encrypting the session key, a la PGP. -- While this isn't sending the key back to the mothership, new keys can be added at will, so copying traffic is easy.)

2) generating one key per account and shuffling it to newly installed clients through their server (possibly encrypted with the user's password... which they already know)

3) generating a key from the user's password directly with PBKDF2 or the like (a la SpiderOak, but (like SpiderOak) the client is closed source and they already know your password or could get it easily).

4) randomly assigning a symmetric key to each session and communicating it in-band to the clients involved in the chat.

Personally, I think 3 or 4 are the most likely because the infrastructure is the easiest and it still carries "end-to-end encryption" buzzword compliance.

The single hardest part of properly using encryption is key management. It's also the most vulnerable aspect of even weak crypto. Anything that simplifies this for end users, without requiring anything of them, is likely making serious security/convenience compromises.

[I'm still a big fan of hardware tokens for key storage and decryption. It greatly simplifies user key management while giving the user something familiar to associate their "key" with. It's not perfectly secure, but having to compromise a smartcard secure element requires more of the adversary.]

In the end, I wouldn't be surprised if any case the US had at all for extradition is ruined by all of the misdeeds they've done in their attempt to 'get' him. They're really overplaying their hand here (as the DoJ has a tendency to do) and it's going to end up biting them (as has happened several times in the past).

I'm in the same boat and I've found that just sending all of my domain's email through Comcast's servers works well enough. I hate doing this on principle, but it has saved me so much hassle that it's not worth fighting.

Depending on your MTA, the configuration will be different, but the arrangement is generally referred to a using a Smart Host. Basically, your MTA directly connects to the ISP's SMTP server and sends the mail from there. Comcast requires authentication to use their servers, but they don't do anything funky to the mail they pass on. All of the headers remain intact except for the DKIM-Signature, which is replaced(?) when Comcast signs the message. I've never had a bounced message that I rerouted through their servers and they support TLS and IPv6, so it's not the worst setup.

I'm sure that if you share your MTA details, someone can help you with the configuration.

White is a more valid descriptor than Caucasian since most "white people" aren't from the Caucasus region. It's fascinating (in a disgusting sort of way) that people have latched onto that term to describe white people. The term "Caucasian" itself is pretty heinous, being coined by Christoph Meiners as part of his theory of polygenism, where he described black people as basically being subhuman animals.

I think "white", "black", "red" or whatever is the least racist form of descriptor since it makes no assumptions of identity, culture, or heritage. The color of our skin is certainly useful as a means of physically describing someone: the tall black man or the brown-haired white woman. Anything more than that presumes to make important inferences about a person based on the color of their skin (read: prejudice).