To some extent pretty much every participant in armed conflict is a coward to one degree or another. Just look at how the British troops felt about the tactics employed by American revolutionaries.

Terrorists, and everyone else, usually do some sort of risk analysis when deciding what to do. While a full on assault of a military base might be more ballsy, it is certainly less effective than attacking a completely soft target full of unarmed civilians. Is it cowardly to minimize your risk while attempting to maximize the productivity of your actions? I guess so, but that is the way that everyone from the top General to the lowliest grunt operates. The only big difference is that typically the US military tries to limit it's scope to just targeting enemy combatants, whereas the terrorists give themselves no such constraints.

In my area you actually can't homeschool unless it is through a religous organization. And the public schools are pretty terrible. I know half a dozen teachers in the local area and everyone of them has done their damndest to make sure their kid isn't in the normal public schools. There are a couple good Elementary Schools but once you get to Middle School there simply aren't any good ones. There are some "Magnet" schools which are competetive to get into and anything under a B in a single class means losing your spot. We're basically planning to move once our children start to reach that age, because we can't afford private school

Arguably the same could be said for the small sub groups of the population that rely on herd immunity for their health.

Not entirely accurate. Aluminum is actually included most of the time because it helps to get a reaction out of your immune system. Doing this allows them to include a smaller amount of virus material because the aluminum will get the immune system moving. This shouldn't be a problem on an individual vaccination basis because the dosage is pretty low per vaccine. The possible problem though is that Pediatricians frequently can't count on seeing a child and administering vaccines on a regular basis, so they usually do a bunch of vaccines all in the same visit, which possibly exposes the child to much more aluminum in their system all at once than is healthy. And certainly more than is necessary because only one of the vaccines would actually need to contain the aluminum in order to get the immune system response. The simplest method to avoid this as a risk is to spread the vaccinations over the same time period with more frequent visits to the Pediatricians office. The disadvantage of course being that it's an inconveince for everyone involved. An aditional advantage though is that if a child has an adverse reaction it is much simpler to determine which vaccine was the problem.

I am by no means Anti-Vac but we have refused one so far, the Chicken Pox. Our reasoning is that the vacine is highly likely to actually cause a case of Chicken Pox, while it does not provide an actual immunity worth the term. What it does do is help make any succesive outbreak to be less severe. It requires 2 boosters or more so far, each of which can cause a fresh outbreak. It doesn't actually do anything to prevent Shingles, which is the real long term threat of Chicken Pox. Typically fatalities from Chicken Pox are limited to bacterial infections when the sores are not cared for. The biggest driver for developing a vaccine was to save working parents the time spent away from work caring for a sick child, which doesn't actually work out because with the vaccine and boosters you will probably have more outbreaks and so more sick time taken. And finally the big kicker is that because the immunity is much weaker from the vaccine than the regular Chicken Pox and requires booster shots as time goes on, we are likely to soon see a generation of young adults who don't actually have an immunity to Chicken Pox, that'll be lots of fun.

A big part of why teaching a child takes so long is because their hardware is under constant construction for decades. They also have a far from perfect memory of how that hardware performs. In theory an AI shouldn't have those problems. An AI should also be much faster at processing than a human, so at some point it should start developing far faster than a human could.

Besides which what is to stop an AI from following that law to the letter by putting every human into some kind of stasis and keeping them securely locked up until the heat death of the universe to protect them from themselves?

That is all very doubtful, although I would agree that you don't have free will, but then again nobody does, just the illusion/delusion.

Even from a very young age I've been able to recognize the thought process that led to various decisions in my life. I remember being asked for the first time what my favorite color was. The same question had just been asked of two of my brothers and they had picked the first two colors I would have likely choosen. I remember briefly wondering if I was supposed to like a different color, was it wrong to like the same color as one of my brothers, what if I didn't actually have a favorite color? We had just bumped over some railroad tracks and I remember the first thing on the other side was somebody's bright emerald green lawn, and I thought that green wasn't a bad color, it was rather pleasant and it hadn't been picked already. So I replied that green was my favorite color.

You very likely constantly evaluate the "correctness" of your thought or behaviour process. Did you ever try something and work at it to become better at it? Surely you didn't write your reply by hitting the keyboard with your fist or forehead. Your use of language also shows intelligence, no one naturally speaks a modern language, it is all learned and constantly tweaked to fit your surroundings or perception of how you want to present yourself to those around you. The fact that you have sought to improve your methods and performance such that you can participate in a conversation using a computer over a network of similiar devices with people possibly thousands of miles away and still recognize them as other individuals not unlike yourself would seem to indicate that you are self aware.

I think a big part of the problem is that we do have experience with what happens when you create a new intelligence and unleash it on the world. We've been doing it since before we were humans, their called children. Most of the time they turn out pretty decent especially when they are well socialized. The problem is that the first learning AI's we produce will very likely be sociopaths capable of learning at an insane pace. With children you can see behaviours and thought patterns starting to form over the course of time and work to adjust it. With an AI it could easily go from infant to omnicient teenager with such speed that there is no time to influence its development.

I'm not religously opposed to developing AI, but I definitely want to error on the side of extreme caution.

I'm not actually certain that the US recognizes any other nations laws as superceding theirs. If a US citizen commits some crime in another country they can still be held accountable for it here in a US court of law. The easiest example that I can think of is some guy who got married and went on a honeymoon in australia where he is accused of murdering his wife by disconnecting her oxygen supply while scuba diving. He was prosecuted both in Australia and here in the states. http://en.wikipedia.org/wiki/D...

The gambling example isn't really relevant because normally gambling laws don't make gambling illegal. Those laws typically outlaw business models that center on gambling. You gambling is incredibly small fries, and would be difficult to prosecute for because the evidence would be somewhere held by people over whom the courts have no leverage.

Microsoft is different here because even though the evidence is located elsewhere Microsoft still has plenty of assets here in the US that the courts can use as leverage.

In the case of that newlywed murder one of the interesting details of the case was that Australia wouldn't send him back to face trial here unless he would not be faced with the death penalty. The US was asking for him back, and I suppose there was plenty of things the state department could have done to leverage that decision. But it came down to the primary driver being the Alabama State AG and since he didn't have much leverage he had to consent to not push for the death penalty.

All of this isn't usually of any concern to vanilla citizen or even a vanilla criminal. But when you are an international corporation it's a very different ball game. What do you think has become of US internationals that had significant assets in Russia when all those sanctions started going into affect? Every large international company has armies of lawyers to help them make decisions and figure out if any given country is worth doing local business. What's funny is that even the US military has had to learn this lesson. When Saudi Arabia closed all the US bases there the DoD lost a lot of expensive infrastructure that had been built up over the years. So lesson learned, the DoD started building extremely cheap temporary facilities in partner countries where there wasn't a very long term expecation of residence.

Very good points.

I can't speak to how well protected those plans were. But it's usually painfully obvious to anyone working in those secure facilities just how unsecure they are. Worst of all you probably wouldn't even need to be secretive. Just getting people to challenge an individual doing something slightly out of the norm is difficult.

I generally like this approach for the DA, however I do have a few concerns.

1. From the wording it sounds like they might be prioritizing some people who are merely uncooperative witnesses.

2. They are using a list of "offenders" that includes people which do not have a conviction. Legally speaking that should mean that such a person is not an offender.

3. Given the issues of racial profiling, and how they gather stats this is likely only improving the criminal outcomes in parts of society that are already getting undue extra attention. That doesn't mean the system shouldn't be used but I would very much like to see them work a lot more on seeking crimes being commited by groups which are not currently profiled.

Agreed! My Daughter loves princess stuff but still has interest in all kinds of other "boy" stuff. My Son is obviously more physical than she is and he wants to either hit everything or use whatever it is to hit other things, but he also loves to play with baby dolls and his favorite cartoon is Strawberry Shortcake. I don't try and push my interests on my children but when they ask questions about why or how something works I make sure to give them answers that leave them with even more questions. I don't believe in pat or over simplified answers. Hopefully they will follow their own interests and ignore whatever stereotype society tells them they should be.

If you are only performing that algorithm once, and you don't need to do anything else at the same time then I suppose you are correct that more wouldn't necessarily help. Although that could still be wrong if the values of a and b are large enough that more cache or RAM would help. And there is also the issue of more RAM possibly allowing other important processes to function more quickly thus enabling the system to process your single addition algorithm in a more timely manner. Once you need to run the algorithm for more than one set of values though you move into territory where parallel processing can improve performance and having more RAM can increase performance by avoiding the need to access slower responding forms of data storage.

Subctraction as a result of adding CPU? I'm not sure how adding resources is supposed to change the actual functionality of your algorithm to do something else entirely. Besides that, you are wrong, that algorithm is already perfectly capable of accomplishing subtraction, as well as multiplication and if I am not mistaken division. All basic arithmetic is actually broken down into an addition equation or series of addition equations by your computer. For instance, subtraction is done through addition once one number is converted to a negative.

More electricity adding new functionality to an algorithm.... I'm really not sure what point you are trying to make that shouldn't already be painfully obvious. Adding a resource does not automatically add completely unrelated functionality or performance increase. It really seems like you are trying to make up nonsensical examples here to no real point. But it does sound fun so I'm going to take a shot at it myself: by giving the AI a ham sandwich it should immediately be able to convert your algorithm into a 5 year plan whereby a hormone free pig can be landed on mars through a green and energy nuetral form of locomotion on Mars, departing from the Earth five minutes before the ham sandwich was delivered.

I would submit that we have met the earlist preconditions for developing an AI superior to ourselves. That was accomplished when the first mechanical calculators were built, or if you prefer when we built the first computers. There is no mystic reason that humans and other animals exhibit intelligence, it is simply the end result of nuerons, among other cells, interacting in an amazingly complex fashion. At its simplest we just need to mechanically/digitally replicate the function of all of those organic parts, with each part interacting according to its own specific set of rules as influenced by the billions, or trillions of other parts. The problem currently with that approach is that we lack the knowledge necessary to code the function of each of those parts, and we aren't even sure we actually know what all of the parts are or how many of each is necessary.

I was skeptical at first but 7 Days to Die is definitely a great game, even in it's current unfinished state. My only complaint would be that I can't cheese much by using a wiki. There is a wiki for it but given the pace of development it is frequently out of date, and frankly just missing very pertinent information.

Then please inform me, in what situation is having more available resources not desirable?

The algorithm isn't necessarily improved by adding more resources. However the algorithm could be broken into more specialized algorithms and distributed to more hardware. That is after all what our brains seem to do, we have areas that specialize in certain tasks and the more neurons you have working towards a solution the better it seems to function. There is of course some point at which breaking the algorithm down further and distributing the work doesn't increase the speed of the solution, but that is a balancing act that I speculate will be always present.

Which brings up "speculation", yes, you guessed it, TFA and all of the commentary here is just speculation. How is that of any relevance to the conversation, you didn't think we were talking about history or something did you?

I would agree that there currently isn't any functional AI. The very best that I've heard of was software controlling some very basic robotics which trialed and errored it's way through some amazingly simple obstacles. Given that Dogs and Cats are self aware I would expect that any kind of useful AI would also become self aware.