Note that the car companies care about compatability, and there's a while ecosystem built around using cigarette lighters into cars. That's why they're all over minivans - they're no longer for cigarettes, they're now the standard car power plugs. :-)

That being said, cars are starting to get USB jacks. That's a good thing. But car technology changes slowly, for good reasons - if they put something into millions of consumer cars, people have to live with it for many years, so they are cautious about making changes.

I'd think that as a part of integrating iPhones/Lightning cables into cars, Apple would have to commit to supporting the technology for 5+ years, with backwards compatibility, so that people could plug their phones 5 years from now into the car that they buy now and have it "just work". Both the car companies and Apple care about that. Google too, most likely.

Except for power, of course. Plug in a phone into a cable and it's got high speed data and power. Connect a phone via bluetooth and it's got decent speed data but a power drain.

Apps that run on phones are fantastic for in-car use: Waze, NextExit, GasBuddy, etc.

They just want to be on a larger display, and integrated with the driver's controls.

Which, amazingly enough, is what CarPlay does.

For nuclear power, there's a similar argument to military - do you want your key infrastructure to be dependent on a supply chain that's not under your control? Imagine, for example, that key components come from China or Korea, and those countries decide to cut us off so we can't get repair parts. That could (eventually) force the US to have to either operate unsafely or shut down power plants.

This isn't theoretical - the US has done exactly this to cut off allies who became enemies, grounding their airplanes, shutting down their nuclear power plants, screwing up their telecommunications, etc.

I was replying to the comment that said "Let's not hamstring projects with a feel good but impractical 'Buy American' requirement. That's the main reason for military gear being so overpriced. If Korea, Japan or China can get components to us faster, more power to them.".

For civilian products, using whoever is most competitive can make sense (though our policy of paying companies to decimate their US manufacturing and engineering capability is stupid). But for military systems in particular, outsourcing to China or Korea is a fantastically bad idea. Even if it costs more, it's worth it to maintain control over the supply chain because the risk of not doing so is unacceptable.

Having your military supply chain depend on countries that you might be fighting against is a terrible plan. It's also in the national interest for the US to retain engineering and manufacturing capabilities. And, of course there's the possibility that they embed controls into the devices that they sell us, the way the NSA pre-hacked hardware being sold by US companies only in the other direction.

So really, it kinda does matter.

The parachute that brought the latest rover to Mars also disintegrated during testing. However NASA proceeded with the design knowing that the atmosphere on Mars is not nearly as dense as it is on Earth.

They got it working in testing after that initial failure - and even that failure provided extremely useful high-speed video of its deployment.

Note the colossal wind tunnel. This latest, flying saucer tested parachute is way larger than that Curiosity parachute - so they've figured out a whole new testing regime. One that helpfully more closely matches conditions in the Martian atmosphere, too.

There's plenty of legal p2p traffic. Almost all large videogames, particularly MMORPGs, use p2p to deliver installers and updates. At Pando, what we found was that not only were the economics much better, but the successful download percentage for managed p2p downloads, measured by people completing the download and getting the game installed and running, was MUCH higher for p2p downloads than HTTP downloads. It's pretty easy to see why - p2p protocols detect and correct for errors, and are quite persistent at getting people correct data, and are engineered to move huge files, while HTTP was never really intended to move single files the size of a modern MMORPG.

Perhaps I should explain - the Pando P2P network delivered guaranteed throughput by using both traditional CDN and P2P networking. For example, if a video stream has a target of 1 mbps rate, and it's getting 700 kbps from peers, it'll pull the remaining 300 kbps from a CDN. This reduces the CDN deliver volume (and cost) by 70%. And it turns out that when peers can pull from thousands of other peers, the comulative delivery rate can often exceed the traditional CDN delivery rate, and it's generally more resilient to networking issues, because an issue that might throttle or stall a single stream, such as a congested router, often won't affect other streams in parallel because they're independent routes.

The 80% was actually measured in a test run in partnership with a number of major ISPs (Verizon, Comcast, Telefonica, AT&T, etc.). We captured all p2p data transfer volumes, then analyzed them to analyze the network data flows, and compared a number of different peer allocation algorithms. Peer assignment that is aware of network topology did, in fact, reduce inter-ISP traffic volumes by 80%. And by rather more on the ISPs that provide symmetric bandwidth, because peers within an ISPs network can exchange data faster than with peers outside the ISP network.

More details at http://en.wikipedia.org/wiki/P... .

IMO, it's Verizon (finally) getting smart and taking advantage of their superior fiber network, giving their customers symmetric bandwidth that cable providers can't provide. Cable companies built a cheaper infrastructure, that physically can't provide as much uplink as downlink. So if Verizon can get people to value symmetric bandwidth, instead of just downlink, suddenly they have the winning network!

Start watching movies in HD (Apple TV, Netflix, Amazon Instant Video) and it'll consume any connection. Then have each of your kids watch their own video streams in their rooms...

I built that network (Pando Networks) a few years ago. The content companies were generally pretty slow to adopt p2p technology, but game companies are all over it. One pleasant aspect was that the advantage of p2p wasn't just economics, though those were great, it was performance. Because downloading from dozens of sources is much more resilient, and on good networks more performant, than downloading from one source. And, with an intelligent network, it could connect you with peers that are close to you in the network, reducing network congestion at the interconnects by 80%. When we ran a large scale test across all the major ISPs, we in fact saw that p2p clients were able to reduce inter-ISP data exchanges (for the p2p network) by 80%, simply through intelligent peer selection, which ISPs loved, and download performance was better, which downloaders loved.

And symmetric fiber networks are awesome at p2p.

Newton's philosophical point was that thinking that there must be a model that is absolutely "right" and all others are "wrong" isn't useful for evaluating scientific theories, because that kind of absolutist thinking kept humanity in the dark ages with competing cults of faith in ancient texts, and he proposed a more enlightened approach as being required to make progress. So, the more constructive question is not "who is right", it is whether a theory makes accurate predictions or not. If a theory makes accurate predictions, verified by independent testing, then it's useful. If a theory fails that test, it's not useful. But models are only useful in a specific domain, and only until there's some cast where it doesn't work, and a new or more refined theory takes its place and extends it. And, for the domain where it applies (i.e. the conditions we all live in normally), Newton's laws of motion work well and are quite useful.

It's true that Einstein's Theory of Relativity refined Newton's Laws of Motion to cover additional domains (e.g. near the speed of light). And numerous others have come up with additional refinements to address other specific domains. But none of those make Newton's Laws of Motion "wrong" in any useful sense. They build on Newton's Laws of Motion and extend them, which is (IMO) the opposite of disproving them!

The "debate about climate change" is about whether it's going on and is affected by human behavior.

In scientific terms there's no debate on that topic - it's settled.

In political terms, there's lots of debate, disconnected from scientific facts.

It all reminds me of the decades where medical science knew that smoking killed people, but cigarette companies tried to present the illusion of a debate, through paid fake "research" and massive marketing campaigns, "donations" to politicians, etc., and it took a long time for the political situation to acknowledge reality. This was a clever business tactic, because it let the cigarette companies sell billions more cigarettes, while the people that they killed didn't cost them anything. Not ethical, or good for the country as a whole, of course, but highly profitable for a few companies.