Yes but X99 and DDR4 blows any chance of doing Haswell-E on a budget. I need a new PC and is considering either 4790K or 5960X, the former is fine now while the latter is going all out on new tech which I hope will last longer. Eight cores crushes the mainstream chip in multithreading. Eight RAM slots in case I want to double up, of a type that will exist long and improve much. Plenty PCIe lanes. Slightly weak single threaded performance at stock but considerable overclocking potential. With 10% performance improvement per generation it'll take ages until I need an upgrade again. On the other hand, a 4790K might last me long too.

They measured multiple things! The statement "We depend too much on science and not enough on faith" was measuring attitudes about science, and neither the article nor the report present it as an example of scientific literacy. Here is what the article stated as proof of scientific literacy from the article:

The executive summary of the report goes on to list some tests as an additional assessment:

Commenting to undo accidental moderation. But since I have to say something anyways...
It makes since that they would draw 9-5 on the graph, for easy comparison and that they would label it the standard workday, since that is what is traditionally been considered as such. But I have no clue how they could look at that graph and come to the conclusion that most people still work from 9-5, as the article text claims.

While I completely agree, subjectively, I also understand enough psychology and statistics to know that a) the feeling of control is mostly emotional, not rational. It's why your mother in the passenger seat is scared in situations where you as the driver are completely cool - you are in control, she is not. That she's more easily agitated only makes it more visible. It's a well-documented fact that experiencing the same situation once in control, or even just seemingly in control, and once not in control is experienced very differently.

Statistics, on the other hand, show that even at this early stage, autonomous vehicles have a better-than-average track record. So while you may feel less safe, the numbers say that you are actually more safe.

Which is why autonomous or semi-autonomous (assisted driving) vehicles do not rely on one input source alone. V2V is not intended to replace all the sensors and stuff, it's one more input source.

Humans are really, really bad at paying attention to monotonous tasks for extended periods of time. The sooner our cars drive themselves completely, the better.

...it could be that drivers systematically overestimate their abilities and underestimate the dangers. Given that we've evolved to live at walking and running speed, moving only our own bodies, it's not a big surprise that our brains don't give us the correct clues at 180 km/h or even 50 km/h when driving a one ton metal thing.

Subjective driver comfort is not something I would use as a measurement for safety.

Yeah, because V2V has about 300 m range. Posting my location to people within view range is really a massive "privacy concern".

We complain about patent trolls getting trivial patents for non-inventions by taking something totally normal and adding "with a computer" to it, but sometimes we do the same. Licence-plate reading cameras are a privacy concern because they can enter your location into a global database in near real-time. Telling people electronically what they could see with their own eyes? Hardly a privacy problem. If we were talking about a system to intercept these signals and update some global database, yes - but that is just the license-plate-reading-camera problem with a different technology. The problem in either case is not having a license plate or having V2V, but the people turning local information into global information.

And other than license plates, it's easy to solve it. Your car could automatically generate a new random ID for itself every time it stops for more than a minute, for example. Pseudonymity is quite cute when you understand it.

Since the idea is that this universe is a simulation, who says it is a simulation of reality? Maybe we are some kids crazy fantasy world in which the container has to be larger then its contents! FREAKY!

The trick to thinking outside the box, is to stop thinking the box is real.

IF this is a simulated world, there is no reason to assume the rules in the simulation are the same as the ones of the world in which the simulation is running.

Are you fucking kidding me? This is so plainly in the "if it's not illegal, it ought to be" category that it's really difficult to think of a more clear example.

It's a direct attack on a competitors system, intended to deprive them of their ability to deliver their service. In IT security terms we'd call it a DOS.

If this rumoured playbook exists, someone ought to go to jail for it. To me it's bright as daylight and even asking the question seems stupid.

Vote me down, I can't do math... 600 MHz-900 MHz/6 MHz bands is 100-150 channels, not 1000-1500 so 4-6 Gbit total for the loop...

No. At least not one that makes sense for storing one or two copies of a consumer hard drive. And you're stuck with a huge investment in one generation of tapes, unlike HDDs where you can gradually buy bigger and better drives. I'd rather see hard drives get cheaper and tape not than nothing getting cheaper at all. What's the real practical downsides to HDDs for the average person anyway? They're standard and can be hooked up to any computer (real fun if your tape drive dies on you or is lost/stolen). They're random access. Without a tape robot it's not more convenient. Without a environmental controlled tape vault I wouldn't trust their longevity claims.

Personally I think the ideal consumer backup solution is three hard drives, one offline next to your computer and one online hooked up via high speed Internet. Anything that nukes your files can't get to your offline copy even if the online copy is hacked or accidentally sync'd, anything that destroys all local copies like theft or a fire can't get to your online copy. One drive goes bad and you should still have two good copies though RAID1 on your main computer would be nice, just to avoid the downtime.

And for what it's worth, most consumer data isn't really worth backing up as they're just a cache to the Internet. I just checked and my total personal stuff (photos, videos, documents, source code, whatever) is 370GB, while I got 10TB+ of other things. And a lot of that which goes under personal is actually "backed up" in that friends or family got copies too, so strictly speaking I could do with even less. I actually see they have 512GB thumb drives now (at insane prices), actually my whole backup could fit in that now.

Well, from the looks of it a coax cable can carry anywhere from 1000-1500 6MHz channels @ 42.88 Mbit/s so 42-63 Gbit/s, subtract TV channels (200 @ 10 Mbit? = 2 Gbit/s), divide by number of subscribers sharing the rest. It shouldn't take that much money to cut a loop in half though, just pick a midpoint and run two coax cables straight to the central office. Considering how rapidly things progress with competition I really doubt there's any technical difficulty in delivering more.

Works for the US military, with more tanks you can acquire more oil...

Sure, I assume that all cars will have something like that. Heck, since the car will be doing navigation it will likely have found a gas/charging station and pulled over long before it even got to that. But regardless they will never be perfect. What if it sprung a leak and couldn't pull over in time because it judged that there was no suitable shoulder (mountain road, narrow bridge), and this info wasn't in it's database to enable it to plan ahead?

We have been mass producing cars for over 100 years, and by all reasonable measures they have never been as reliable as they are today. Yet they still break down on occasion. Self driving cars will have all the same mechanical and electrical problems that we have today, with software problem on top of that. You can mitigate some of these hardware problems with additional sensors, and fault-tolerant design of the driving computer, but only to the point where the sensors and software are significantly more reliable than the hardware they are monitoring, and only for the situations that are programed for.

There always will be situations where things break down in unexpected ways that the car isn't capable of handling on it's own. And based on the historical rate of reliability improvement, those situations won't be uncommon for quite some time.

May the computer totally fail to realize that the bridge is about to give out or the building about to collapse or an avalanche about to hit or a dam about to burst or you're driving right into a rioting mob or some other disastrous event? Even if I assume that the car will never, ever throw the controls to me and expect me to take over doesn't exclude the possibility that I want to take immediate physical control to avoid some kind of danger that goes above and beyond a computer's understanding of traffic rules.

They may never be removed. Everyone is focused on the split-second decision scenario when talking about this issue, and on that I agree that humans will cause more problems than they solve. But there are many more situations where manual override is needed and beneficial. What happens when the car runs out of gas/charge and you need to push it to the side of the road out of traffic. Or the computer is malfunctioning somehow (software bug, squirrel chewed halfway through a wire, dead battery/alternator). Or when I need to move the car somewhere off-road that the AI refuses to recognize as a valid driving path. There are plenty of not so time critical scenarios where some sort of manual override is needed and those aren't going to go away even when we trust the software to do all the driving. Once we admit that they don't have to be intuitive for split-second reactions, then they don't have to retain the traditional layout, nor be designed for comfortable everyday use, but some sort of steering, brake control, and neutral gear select will always be needed.