From the article most of the spending is on things that are beneficial to society as a whole, not just NSA. These include K-12 funding for science fairs, math clubs, and STEM summer camps. Unless the NSA is influencing these in harmful ways, such as pushing ideology beyond the normal "if you do well in school, you could do cool spy work for us" recruiting I don't see a problem with taking their money. Same for the research grants and conferences, which all result in publicly published fundamental research, that help the entire cryptographic and big data communities as a whole. The only program I would have a problem with are any classified research and the sabbaticals to do classified work at the NSA.

No, you could use a conductive rail, like a subway, and rack and pinion system to move the elevator. The rack and rail would add a fair bit more total weight to the building compared to a cable. But more importantly, the motors would have to be much much more powerful! Modern elevator systems have a counter-weight balanced on the other side of that cable, which means the motor only has to overcome friction and the small difference in weight between the elevator and counterweight (which varies depending on current payload). The motor on an elevator like Noah is suggesting would have to provide enough force to counteract the entire weight of the elevator + payload + motor + friction, which is at least an order of magnitude more than a traditional elevator.

If I worked for Wikileaks, I think I'd be encrypting everything especially if it involved using a Google server.

No, good scientists understand significant digits. As far as geological epoch go, the time elapsed between the start of the industrial revolution and the start of the nuclear age is insignificant. Furthermore, while the technology began at the industrial revolution, the impact of that technology didn't have global environmental scale until later on. We don't mark the other geological boundaries at the point where precursors to change appeared, we mark them when change became significant. If you look at graphs of human energy or CO2 output, the knee in the curve does occur at around the mid 1900s. The fact that there happens to be an easily observable geological marker that occurred at that time makes it a convenient dividing point, and as good as any other of the arbitrary dates picked to divide otherwise well-distinguishable geological epochs.

If anything, I would argue they risk jumping the gun too early, not setting the date too late, as there may very well be a much bigger global change in the next 10's thousands of years of which the last millennium will just be regarded as a precursor to.

Tax preparation software is not a good candidate for open source software. You need domain experts (accountants and lawyers) to be involved to validate the interpretation of the Tax Code; open source projects have a difficult time attracting these sort of contributors. The law changes every year and if you don't keep on top of the changes becomes worse than useless; it becomes a liability. You have solid deadlines; you can't just release when it is ready.

That's OK, the conversion is really easy. 1 USCU = 10,000 CCU. Canada's central bank hasn't gone crazy lately with printing of the Canadian Coolness Units, and the dip in crude hasn't seemed to impact it yet, so I think that it's still 10,000-to-one.

Again though, I recommend using the SI Coolness Unit - the Fonzie.

Thanks, I should have been using the established coolness unit of a Fonzie.

Because it is cool. You're measuring ROI in United States Dollars, when you should be measuring it in United States Coolness Units.

Seriously, this is the argument that people use on me with trying to convince me to buy a hybrid, or more fuel efficient vehicle. My car is horribly inefficient (seven seater SUV) but I either need something that big to haul around 4'x8' construction materials, I ride my bicycle, or I drive it like once a month out of town for a few hundred miles for work. It's entirely paid off, and the (relatively high for me) purchasing gasoline part of owning a car (unit cost per mile driven) is insignificant compared to the free/already paid for fixed costs of owning a car.

An ex-girlfriend and I had this discussion, and eventually it came down to the don't you want a nicer car to drive around? argument. No, I don't want one, if I have to pay for it. Having a cool car isn't that important to me. I have a different girlfriend now...

There is no financial, or logical, reason to automate a home to save electricity in your case, unless you want to be cool. If you want to show all your friends how "green" you're being (despite all the manufacturing, shipping, and other environmental costs used in producing the crap you're busy buying), write blog posts about your home automation project, take a bunch of pictures and post them to instagram, then it makes sense. OR If you plan on living in your apartment for more than 200 months (16 years) then you'd eventually break even on the project cost...

H. Beam Piper wrote about this in 1952, in his book Null ABC. The author detailed how literacy in schools continued to decline, as more and more educational gadgets became available, until society was divided between "literates" and "illiterates." The illiterates controlled the vast majority of business, but literacy was still required to practice law, and serve in the judicial branch of government.

Check out a physical version of the book here, an audio link here, a free eBook version here and a free audio book (that is probably the same as the paid one I linked to you above) here.

I really enjoyed the audio version I listened to. It was extremely entertaining, and a scathing social commentary on the future of public education as H. Beam Piper (correctly) envisioned it.

One important point that others above have alluded to but haven't outright stated:

While the exponential scaling of rocket equation is an important limiting issue when building larger and larger rockets, for any given rocket (or rocket configuration) the payload capacity is fixed. If you have a payload that is too large for a Falcon 1Pegasus, but doesn't need the full capacity of a Falcon 9, all that extra capacity goes to waste. It costs essentially the same amount to launch a Falcon 9 at 60% capacity as it does to launch it at 90% capacity. You can share payload with multiple customers, but that limits which orbits they can use.

Space X can calculate how much weight the recovery system and fuel requires and how much money they can save by reusing the first stage, and give a discount to customers who give up that additional payload capacity. If there is a market for those lower cost launches, then great. If not, then keep treating the 1st stage as disposable.

Running a business like this takes a lot of work, and for it to succeed well enough to actually get working rockets off the ground you need to attract top-notch engineers who believe that working for you isn't just a waste of their time (more than a billionaire's plaything), and management that can create the right environment for them to succeed without blowing through your money for nothing. It is much less expensive, less risky and less time consuming to just pay Russia for a thrill ride than to create your own rocket company. So I can understand why most would choose to go that route, and leave the latter for those who genuinely want to shake up the market.

That said, the total payload mass that the ship could support is roughly the same whether it is inside the airship or outside in a gondola, and the more space you want to make available for use, the more mass you would have to dedicate to structure rather than payload. So it would be less cramped than a tiny capsule, but you would still need large expanses of mostly empty space to provide the needed buoyancy.

In practice, it might be better to have a balloon filled with a less dense gas to decrease the total volume needed to support the desired payload, and then have an attached air-filled "gondola" that is nearly as large as the balloon.

I would highly recommend you take as much as possible at community colleges, paying as you go. The universities in my state (New Mexico) accept community college credits very, very well. Slightly before you've exhausted the community college course load, apply to, and get accepted into, a bachelor's program in some sort of engineering (not all science degrees are equally marketable.) After you're accepted, and have completed a year or two's worth of marketable engineering courses at the community college, you should be able to get an engineering internship and continue to pay cash for classes. These student, engineering, jobs (in my state) pay more than English degree professional jobs do. I've seen this approach work with computer science students.

My state has extremely inexpensive, or free, tuition for residents and access to a huge amount of engineering resources (two national labs + tons of military bases + the initial stages of a tech start up scene) as well as dirt cheap cost of living. I realize this approach might not work well in other states, but that's the approach I talk with people about. I'm working with a guy that studied music, but is getting into web development. His goal is to get accepted into a master's program, and spend an extra 2-3 semesters in it taking undergrad courses. If he can get funding (as a research assistant, or teaching assistant) that will be a great approach too.