Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!


Forgot your password?
Check out the new SourceForge HTML5 internet speed test! No Flash necessary and runs on all devices. ×

Comment Re:Humans, not AI... (Score 1) 189

Any AI in the foreseeable future will be under control of human beings, either due to laws or financial ownership. I'm not the least worried about AI, but having watched this election, the humans in my country scare the shit out of me.

I agree that it seems implausible that an AI will "rebel" and set its own goals, that is still sci-fi. But ordinarily those scary men would have to enlist the help of many others, like the old DDR (East Germany) where almost a hundred thousand of a population of 16 million were STASI agents. Use a NSL, use a computer to transcribe it and analyze who is talking to who when and about what and for how long and you could have a quite Orwellian database with only a hundred people involved. Same thing with bank transactions, electronic tickets, number plate scanners, facial recognition, deep packet scanning and whatnot the automatic collection and processing enables a very small but organized and powerful fraction of the population to surveillance and control the rest.

I'm not really sure we could reverse that trend because just like we might discuss the pros and cons of everyone having tiny little digital cameras on them at all times it is still extremely unlikely to change. The average person is leaving more and more electronic traces and even those who try to avoid it is often found by metadata from their friends or they stand out by being the black hole that isn't sharing. Very often the system is rigged towards it, for example we didn't want bus drivers to get robbed so much so now it's a lot more expensive to pay cash than a prepaid subscription or cell phone payment, both of which link to a real identity here in Norway. People do what's cheapest so the few people who pay cash stand out. And really it's no problem until it is a problem, but then you're usually too late to change it.

Comment Re:Were the users randomized? (Score 1) 443

Not when you have a selection bias, it isn't. If your sample selection is consistently biased, no sample size will be large enough.

Agreed, but outside math class you have to look at the percentage and make an educated guess about how special they realistically could be. If you have a thousand employees your number one is probably a genius and your very worst a moron. The 10th from the top is also probably pretty smart and 10th from the bottom pretty stupid. The 50th smartest isn't aren't all that special though, if he can be more efficient with a Mac well it seems worth trying the top 100 or 200 too. It could of course theoretically be that performance drops off a cliff because it takes some minimum skill and understanding you just dipped below, but realistically if that happens it's probably the kind of thing only 1% or 0.1% of your employees grok. If 5% can use a Mac so can probably most of them.

Comment Re:Tired of this space obsession (Score 1) 87

The way I look at is if the reusable rocket guys get the cost of orbital rockets down to 1/10th of the cost that it is now, lots of options open up. If you can get 10 trips up for the cost of 1 now, suddenly assembling a Mars-distance ship in orbit and all the fuel and supplies to make it happen seems pretty plausible.

Assuming the rocket is really the blocking cost. A Falcon 9 expendable is around $62M * 130% = $80M for 22,800 kg to LEO. Even a hundred launches for 2,280,000 kg payload (Saturn V had 140,000 kg to LEO) would be "just" $8 billion and they've spent more on that developing the SLS before it's flown once. Granted it'd be some assembly required but >20 ton modules aren't trivially small either if we design good interlocks and docking in space is pretty routine now with the ISS. So for the sake of argument, say we're now down to $800 million instead that's basically pocket change in this context. We'll still need everything else, uniquely designed for this particular mission. I think a one-off mission will still be >10 billion. Musk is thinking further out than that though, he'll want to bring the price so far down we'll want to keep going. Not one mission but ten, hundred or a thousand. Good for him, but there's still some pretty big hurdles to cross... they need a customer who's willing to pay.

Comment Re:This is dumb (Score 1, Interesting) 194

That's what they said about iTunes, and Apple found a way. So I wouldn't count them out here...

Only because the music industry was totally oblivious to what would happen when they let Apple control the DRM, the same way IBM let Microsoft control the OS. They had to drop DRM because they were being buttfucked by Apple who used their market power to sell cheap music and expensive iPods and all their customers were locked in since FairPlay protected music wouldn't play anywhere else.

The motion picture studios have never been that stupid, even long before iTunes they controlled CSS on DVDs, they control AACS/BD+ on BluRays and AACS 2.0 on UHD BluRay and they have no reason to drop DRM. At worst even if it's broken they can still try using the DMCA and EUCD to make decryption tools illegal. And unlike the iPod that filled a need for a device people didn't have there's already tons of ways to play movies and series.

It's of course possible that Apple could find their Achilles heel but I don't think it's very likely, with or without Jobs. The industry would most certainly smell a trap, even if they couldn't figure out what it was. As much as I'd like them to just give up and have the convenience of a torrent site I think it's just very wishful thinking. That said, there are a lot of cable cutters so maybe...

Comment Re:So it appears . . . (Score 2) 177

there were two failures: the parachute release and the burn length. But both were likely set in the software on the lander, so I suspect parameters got borked somehow.

As in hardcoded on a timer? Unlikely. This is quite far into the descent and the parachute was probably supposed to jettison when a certain altitude/velocity was reached. That both the parachute and thusters was off suggests to me a sensor failure led the probe to think it was going much slower or flying much lower than reality. It would be odd for both systems to fail and at the same time be in good enough condition to send radio signals.

Comment Re:Remote exploit (Score 1) 71

Most attacks these days are a sequence of memory safety violation followed by memory disclosure followed by arbitrary code execution. ASLR is meant to make the memory disclosure part harder, but there are now half a dozen known attack techniques that allow ASLR to be bypassed. Off the shelf attack toolkits will include these mechanisms, so it's a mistake to assume that an attacker won't be able to bypass it. It increases the barrier to entry from script kiddie with 5-year-old toys to script kiddie with new toys.

Comment Re:self-driving or assisted driving ? (Score 2) 185

The point I was making, that if a human can interpret the visual information it's given, then a car with a bigger sensor set can in theory do it too. It's all about software at that point, but there's no limitation on hardware here that a human doesn't have.

Yeah but... Tesla's claim is like saying the brain consumes about 20W, the car can deliver 20W so it's "ready to support an artifical brain". While that might be technically correct it is also grossly misleading, in that we don't have and don't really expect to have an AI working at all or so well and certainly not within the constraints of a human body in the foreseeable future. Same thing with cameras, I expect the first real SDCs to use optical and radar and lidar and every other trick in the book to overcome the shortcomings of the brain behind it. Same way some talk of trying to simulate the brain with >10 megawatt computers.

Comment Re:Really? (Score 1) 244

You might be able to salvage ~20% of them with humans aboard. Might. Meanwhile, humans are a massive added source of additional risk to a mission; they dramatically increase spacecraft size, complexity / part count, consumables, and just in general make things far more difficult.

Having people on board cause their own problems, you wouldn't have to improvise CO2 filters on Apollo 13 if there was no one that needed oxygen so they might cause aborts too not just salvage them. Having humans on board will also limit your alternatives and lead to its own time constraints because you have to keep the crew alive. And most importantly, humans are not expendable and unmanned probes couldn't have any realistic return plan even if you wanted to so in most cases they still wouldn't be an option.

And it's not like NASA just send things out there with plan A and no plan B or a static programming they're stuck with. They try to find possible failure modes, emergency states, when the rovers get stuck on Mars they have huge team back on Earth trying to find out how to best wiggle free. That said, I think a research outpost on Mars could be useful. But as for the rocketry, most problems will be totally catastrophic no matter what. I'm sure there's a lot more practical issues to having an outpost a human could deal with, though I admit it's a bit circular because the reason to have an outpost is so people can be there. Otherwise you'd just send specialized probes to do their one thing.

Comment Re:Holy flamebait batman! (Score 1) 883

If you don't have a job, "relocation" is a bus ticket. But very few people move to improve their circumstances.

Not true. If you don't believe me, look at the statistics for worker mobility - they correlate strongly with wealth. Poor people are a lot more reliant on their support networks (family, friends, and so on). If they're in a poorly paying job, then they probably can't afford to take a month to look for a new one in the new location (especially with the real possibility that they won't find one). If they don't have a job, then there's a strong psychological pressure not to move to places with fewer jobs and there's likely to be a delay in receiving unemployment benefit as these things are typically administered locally.

In contrast, someone like a typical Slashdot poster can afford to stay in a hotel room for a week or two (or have an employer willing to pay the cost) while they look for somewhere to live and will typically be able to find a job before they start moving.

Oh, if we're willing to tax the first dollar of earnings (over the UBI), it's far more credible. But right now the majority pays effectively no income tax, so that would be a massive change.

UBI itself is a massive change, so it's weird to think that you'd introduce it without introducing massive changes. Most proposals for UBI have it replace the tax-free allowance. You might have a very small tax-free allowance on top of it, but generally the way of balancing the books involves paying tax on all earned income.

Comment Re:Including a Mac Pro tower, right? (Score 1) 142

Rendering, intermediates, etc all need bucketloads of *local* storage, i.e. you don't want to render on anything that has an IO bottleneck. Is thunderbolt really as fast as the internal bus?

Thunderbolt 1 ~= 2x PCIe 2.0 = ~1GB/s (signal is 4x but won't go full speed)
Thunderbolt 2 ~= 4x PCIe 2.0 = ~2GB/s
Thunderbolt 3 ~= 4x PCIe 3.0 = ~4GB/s

I guess if you RAID 0 you can go faster but if you have hardware and software that can create video at >4GB/s you're pretty special.

Comment Re:Including a Mac Pro tower, right? (Score 2) 142

I don't know why people continue to buy their stuff. Sure a regular boxy pc may not be as attractive, but it'll be a lot cheaper, perform better, be infinitely more upgradable, and can be modified to suit your personal demands.

Well you can take it out as same performance for less money or more performance for same money, I think for most it has "enough" though so it's really attractiveness vs price. As for upgrading most people are quite happy with laptops (stats from Norway age 16-65 says 30% use desktops, 80% laptops), I don't think they even consider it the same way I've never considered replacing the engine or gearbox on my car. It's not like any part will be outdated in two years anymore and if my needs radically change I'll sell it and buy a different car.

Sandy Bridge, Ivy Bridge, Broadwell, Haswell, Skylake it's not like CPUs are revolutionary improving. Graphics did a huge leap recently because they skipped the 20nm node but I'm pretty sure a GTX 1080 will be a pretty damn decent card five years from now. RAM, SSD/HDD, connectors like USB are quite stable too. It might not be the most economically efficient since I'd buy extra early for it to last, but I could certainly build a PC today I could weld shut and not open for 5-10 years barring hardware failure and not really miss anything. Which means it's again price aka money and so is service/replacement costs.

Basically if you have the money there aren't really that many other downsides. Much like despite what /. says the average consumer doesn't mind a non-replaceable battery.

Comment Re:Not a bad idea (Score 1) 46

Not a bad idea. This way a lot of people will just accept the ease of quickly getting a Samsung replacement, and not wander off and buy another brand.

It's mostly to lessen a PR nightmare, the Samsung Note 7 is now a prohibited item. So if you have been totally oblivious and try to bring one through security control you'll basically get the choice to not pass and miss some probably important and expensive flight/trip or to quite literally throw the phone away. These shops mean that if you had a good time buffer getting to the airport you can make a "Hail Mary" save by running off to the Samsung booth, get a replacement and still catch your plane. I doubt they'll get much intentional business, though I suppose some might for the convenience since they're going there anyway. I don't think that's why Samsung's doing it though.

Comment Re:Basic Scrutiny (Score 1) 883

Further risk-reduction is possible, for example by entering into a two-party agreement mediated by the Social Security Administration: the exact rent is direct-deposited to your landlord, and you get the remainder; if either party cancels this agreement, both receive immediate notice. As well, tenants can buy their way out of the cost-of-risk by placing a larger security deposit as an insurance fund against their own risk. All of this means we can rent stably to these people at a profit. (...) assuming everything has a 15-year life. The average life of a kitchen sink or cabinet is 50 years, while the faucet does last about 15; cheap countertops last over a decade and can be relaminated; and bathroom fixtures last 20-50 years at least.

I think you've only imagined this, not tried it. Many of the totally unemployable people with nothing but basic income would have drug problems, mental problems or otherwise be dysfunctional. Here in Norway they usually end up in publicly own housing because no matter how much they guarantee rent and deposit it's not worth the complaints, the eviction process, the claims process, the renovation time which translates to lost income and so on. I have a friend who did that mistake once, the apartment was totally trashed in less than a year, not just the furniture and appliances but like the floors, the walls, everything. I also have a relative that's not all there, he let the dogs pee and shit inside and since he lived alone it went so far the house was condemned and torn down. That's how bad it can get and even under normal circumstances they're not going to take care of it like it's their own. And if you can't prove they broke it, well...

I'd say cut all the lifetime estimates you would have had if it was yours in half and add the risk of disaster and I think you'll be closer to reality. Also if you have a cheap and not that great rental object I wouldn't bet on 100% coverage, if you got an attractive one you'll rent it again quick by lowering price if you have to but if there's an oversupply some of the less desirable ones will remain empty. There are those who manage to do it and make money on it, but usually they're also circling like hawks to spot potential troublemakers and nip it in the bud. It's easier once you reach the kind of prices only "normal" working people with money can afford.

Slashdot Top Deals

When all else fails, read the instructions.