Follow Slashdot blog updates by subscribing to our blog RSS feed

 



Forgot your password?
typodupeerror
×

Comment Re:Stupid question from crypto-newb (Score 2) 45

Determining whether two boolean circuits are equivalent is a famously difficult problem to solve. In fact, it can be reduced to SAT (http://en.wikipedia.org/wiki/Boolean_satisfiability_problem) which was the very first algorithm that was shown to be NP-complete, which in general means it is impractical to solve on real computers.

Comment Re:check your assumption at the door (Score 1) 103

Not really. Your lifting power can never be more than the weight of air it displaces and hydrogen is already a whole lot less dense than air. If you do the math a complete vacuum will only lift about 7% more than hydrogen. Even then I don't think there is any technology that will give us light containers that can withstand vacuum pressures of any usable size.

Comment Re:Silly question: (Score 1) 169

No, from the outside it will look like the cube just falls in, depending on the size of the black hole and the angle of approach, we may see it undergo spagettification first.

However, you can use a black hole for data retention as a delay line. There is a distance above the black hole that is called the "photon sphere" which is the point that the orbital speed exactly equals the speed of light, meaning that photons injected at the proper angle will actually be in orbit around the black hole.

So, you can use a laser beam to spit out your data on an almost but not quite orbital path, sending the data around and picking it up after it orbits, picking out what you want and re-transmitting everything else. The latency would be high, but the storage space would be incredible. Imagine you set the angle so the light orbits enough time to travel a light week before you capture it again, todays optical interconnects work at 100Gb/s, a week is roughly 600 megaseconds so you get 60 petabytes of storage more or less, per frequency you use, and not to mention you can send data in both directions, and have delay lines longer than a light week. You could have a whole storage hierarchy with delay lines of a few light seconds to centuries around the same black hole to balance latency/storage space.

Comment Re:Mis-calibrate everything FTW! (Score 1) 226

Yeah, I figured it was probably sarcasm, but it actually isn't that far off from the anti-science or just plain science-ignorant positions that some very vocal people tend to take. (oh sci.physics.relativity, I mourne for you.). So I figured on the off chance that I can make at least one anti-science individual reconsider their views, it was worth replying too.

Trust me, I am much happier that you were posting than sarcastically than if you were a kook who actually believed it and wanted to argue. :)

Comment Re:Mis-calibrate everything FTW! (Score 1) 226

I probably deserve a "Whoosh!" for this but I'll bite anyway.

Such a scheme would fail at the reproducibility part of the review process. You have to describe your process in the paper such that someone else can reproduce your results, if they build a correct machine that isn't mis-calibrated and then get a different result, it will then call your paper into question. Pull these shenanigans enough and people will stop publishing you and take a long, hard look at whatever university gave you your degree. Universities are very motivated to weed out the bad seeds, sometimes someone slips through the cracks.

Of course, you may still have a valid paper if your machine was mis-calibrated and failed in a new way that no one expected before and the value of dissimating that information so others learn from your mistake is worth it.

Comment Re:Do they account for hypothesis-mining? (Score 1) 226

> The real test is to come up with the hypothesis first, then collect the data.

That is exactly what they were doing. Testing the hypothesis that the standard model accurately describes nature. They found it didn't, hence the need to explore it and come up with new hypothesis's to test.

1) You start out observing something the current theory can't explain.
2) Come up with a new theory that accurately predicts all experimental results so far, the newly observed effect, and that also predicts something new that has not been tested yet.
3) Test the new thing that the new theory predicted. If you do observe the new effect, it lends credence to the theory.

Wash, Rinse, Repeat.

They are claiming to be on step one, with an inkling of step 2 being worked on. not step 3 to which your specific criticism would apply.

On a tangent, the most commonly overlooked part of the process among cranks is the consistency part of step 2, namely that your new theory must accurately predict everything that has already been observed. I don't think it is a simple oversight, there is some metal block among cranks that keeps them from appreciating it, hence the propensity to claim they can prove Einstein wrong. Which of course doesn't make sense, he was already proven right. That doesn't mean that relativity is the final answer, it just means that it successfully predicted observed effects that the old theories didn't, and was consistent with all observed data about the world so far. (Well it breaks down at the quantum level, but so did newtons laws, so it was still a strictly better theory in that it predicted more things correctly, but still not everything)

Comment Re:That system had better be bullet-proof (Score 1) 60

Many subway/public transport systems already have swipe readers, such as TAP in los angeles. It just requires carrying around a special TAP card and opening a TAP account rather than being able to use your phone and an independent billing method. Most stops have automated payment kiosks only and no one gets stranded, they just aren't very attractive vandalism targets and there are a lot of CCTV cameras at the stops.

Comment Re:Where's the -1 Hate button when you need it? (Score 1) 218

No, a continuum just doesn't work in practice. people vote the minimum if they think something currently has too high of a rating, or the highest if they think it is too low, in order to have the most "impact" on the score.

This is why I always thought they should explicitly phrase things like "This movie currently has a score of 7.3, do you think this is overrated, underrated, or just right?" to more clearly align things with how people think.

Comment Number One way to reduce drunk driving. (Score 1) 343

_Make the trains run past last call_ Serious. (not an issue everywhere, but in los angeles it is a huge one).

For the most part, people don't want to drive drunk. A $1.50 train ride out easily turns into a $150.00 cab ride back if you are just a few minutes late for the last train and there are huge swaths of LA that cabs literally just won't go at night. People are scared of getting stuck in the city, staying overnight on the streets isn't fun because you can't afford a cab ride back and the trains stopped. Once people experience it, they drive from then on, even though they probably wouldn't otherwise, just in case.

They don't even have to do a full schedule, just every half hour or 45 minutes would suffice. sigh.

Slashdot Top Deals

To the systems programmer, users and applications serve only to provide a test load.

Working...