Forgot your password?
typodupeerror

Comment: Re:How is that startling? (Score 3, Insightful) 349

by plover (#48478601) Attached to: Mathematicians Study Effects of Gerrymandering On 2012 Election

Gerrymandering has a long, proud tradition in U.S. politics. I wouldn't be surprised if it resulted in advantage to one side about half the time.

It seems that political power is self-limiting. One side will occupy Congress for a while, until the other side gets fed up and makes a switch. As long as one party doesn't control the whole Congress plus the presidency, gridlock keeps us safe from most of the excesses of either side. It's only when one side runs the whole show that it's time to worry.

Comment: I'm glad you spotted it (Score 1) 26

by plover (#48477229) Attached to: What is it like to be mentally ill?

Reading about how slowly it crept up on you helps me realize that "illness" is a really important analogy. There was no specific trauma (apart from your incident 30 years earlier) that triggered the need to seek help, no recognizable thing, just a steady worsening of your condition. I don't know that I would recognize pain like that that builds over such a long time. But if I do have self-destructive thoughts like that, I'll at least think about getting help earlier, rather than (maybe) spotting that "picking a date is really, really a sign".

Thanks

Comment: Re:First rule of computer security!!! (Score 1) 113

by plover (#48469229) Attached to: Auto Industry Teams Up With Military To Stop Car Hacking

The radio generally isn't on the same CAN bus as the ECM. The ECM is on the high speed bus which usually is reserved for engine and safety systems, like airbags and ABS. But as you noted, there are places where messages have to cross over: airbags need to be able to tell the entertainment system to have the connected cell phone dial 911. There are commercial CAN bus bridges available that provide this function, and they can be configured like a firewall to isolate all messages except those identified as needing to pass through.

Whether or not these bridges are actual security appliances is a different question. Who has the authority to alter those routing tables? Where is the password kept? How are they secured? CAN is a low level protocol that was never designed to be secured.

Comment: Re:First rule of computer security!!! (Score 1) 113

by plover (#48469083) Attached to: Auto Industry Teams Up With Military To Stop Car Hacking

The military is good at physical security. That's their mandate, after all. It seems logical to put them together.

However, they seem to suck at this aspect of it. There is no reason that an American vehicle (or weapons system) left in the hands of an Iraqi army battalion should ever be able to be commandeered by troops who switched allegiance to ISIL. There should be an American satellite link required for occasional checking-in, and the vehicle should be disabling itself if it's failing to check in, or if it's been added to the "captured vehicle list", or whatever. High-risk auto leasing operations are already doing something like this today, with a kind of inverted Lo-jack system.

And from a maintenance standpoint, this shouldn't be an issue. The machines already require sophisticated computer control to turn on and run. All it has to do is wipe out its program when the "blacklist" threshold is hit, whatever that may be. It's not like ISIL would be able to order a replacement aftermarket electronic control system for these from Alibaba. For that matter, the engines could include embedded charges (think exploding bolts) that would physically disable the machine on receipt of a suitably authenticated "hostile takeover" signal. Or they could simply continuously report their ID and coordinates, and a J-DAM could sort them out right quick.

Yes, I'd raise holy hell if my car's computer shut me down because the manufacturer added me to their blacklist. But this is like a commercial operation, where the assets don't belong to the drivers, they belong to the Army. And they never, ever belong to ISIL.

Until they get this right, why do we think they are going to get consumer car security right?

Comment: Re:Wake me when they solve the infrastructure prob (Score 1) 281

Infrastructure has to be built one sale at a time. Tesla is demonstrating one way to do it with their supercharger network, with trickle chargers in the home, and supercharging stations scattered around the country, trying to bridge gaps in coverage.

A hydrogen infrastructure will look different, because pressurized hydrogen isn't as ubiquitous as electricity. They might have better luck with a regional approach, selling commuter cars in one city, and building up an infrastructure there just to prove it can be done. This could go hand-in-glove with a partnership with a rental car company, where your car price comes with discounted rentals for cross country trips. They might even be able to start with some fleet approaches: delivery vans, local taxi services, city government inspectors, etc. Get a few vehicles out there first, then expand into the consumer market. Once the hydrogen delivery trucks start making rounds to carry fuel to the fleet terminals, it's not a stretch to get them delivering to consumer facing refueling stations.

Or maybe hydrogen delivery service stations could be provided in a novel format, like a standard shipping container. Build a tank and pump system into a steel box, and make arrangements with a company like BP to drop one in the parking lot of an existing refueling station whenever you sell a car that's not within 10 miles of an existing station. BP may like drilling for oil, but their primary business is selling vehicle fuel. This is an opportunity that doesn't bypass them, like home charging stations do.

The one thing that would be likely to fail would be to take billions of dollars of investment, and build a national network of thousands hydrogen refueling stations before the arrival of millions of hydrogen consumers.

Comment: Re:In a Self-Driving Future--- (Score 3, Insightful) 453

by plover (#48443623) Attached to: In a Self-Driving Future, We May Not Even Want To Own Cars

In suburban-heavy US metropolitan regions, Zipcars haven't made inroads yet because the sources and destinations of people are not close to each other. Suburbs are all houses (sources of people) but have no shops, factories, or businesses (no destinations). If my neighborhood was to have a successful zipcar garage that served everyone, it would have to contain as many cars as there are nearby residents, and it would still be emptied quite early in the mornings. The urban centers have few residents who would commute away from the city to work, and would not provide a demand for the tens of thousands of cars that would arrive every morning.

If the cars were self-driving, they'd be able to return to the suburbs to provide many trips per day. More trips per car means fewer cars are needed.

Comment: Re:Moat? Electric fence? (Score 3, Funny) 212

by plover (#48422125) Attached to: Congress Suggests Moat, Electronic Fence To Protect White House

"Hey, Joe, now that we've finished surrounding the Capitol building lawns with mines, we've still got a bunch of extra mines. What should we do with them?"

"They're not extra. They said ring the building, so the plans are to mine the walks and driveways, too. Maybe if they wrote the policy better, they'd have thought to add an access route."

Comment: Re:What? (Score 3, Insightful) 108

by plover (#48422093) Attached to: The Software Big Oil's PR Firm Uses To "Convert Average Citizens"

This is the same thing that every company big enough to do public relations at all does, except it's being described using inflammatory terminology.

That's what I was thinking. If they are getting real people to agree with their position and sign up with their on-line site, how would that make their individual choices illegitimate? How could that be painted as "astroturf" when it's clearly legitimate support?

Look at the other side. If I worked for a railroad that operated thousands of tanker cars that ship oil across the country, I might go to the stop-the-oil-pipeline.org site and pledge my support. As a railroad, I burn thousands of gallons of oil to ship millions of gallons of crude. I have no interest in protecting the environment, yet here I am, signing up. It's not because I'm an environmentalist, it's because I don't want the competition to take away my business. Where is the story claiming this makes the environmentalists an astroturf organization? There isn't one, because it's not.

Why isn't this story looking into the CRM software in use by the environmentalists? Perhaps their bias is a bit too evident.

Comment: Re:Impossible to build purely evil robots? (Score 1) 333

Isn't an atomic bomb just a very, very simple robot?

while (altitude() > TARGET_ALTITUDE)

        sleep(1);
explode();

And yes, it is impossible to determine if that algorithm will ever terminate.

A "good" compiler should throw an error and refuse to compile it, because the function's return can never be reached. An "evil" compiler will spit out an ignorable warning, but let you build your bomb. That implies we need to use evil compilers to program the Kill-O-Bots.

Comment: Re:By the same logic (Score 2) 333

So how many humans have to die before recognizing the AED is faulty? If it's a subtle fault, it might be delivering a barely ineffective treatment, and confused with an unsaveable patient. The THERAC 25 failure was a bit more dramatic, but it still killed many patients.

Would we accept the same levels of failure from the Kill-O-Bot 2000? We already fire missiles into crowds of people or convoys in order to take out a single high value target. If the Kill-O-Bot was more specific than a missile, but less than perfect, isn't it still a better choice?

Optimization hinders evolution.

Working...