None of this addresses how he managed to hop from the entertainment system network to the flight system network, which many people have claimed are air gapped from each other

Not quite air-gapped, bridged one way. Otherwise how do you think the flight page on the entertainment system gets its data form?

The aircraft has two networks. The inflight system is Ethernet based, traditional IP and everything. Inflight WiFi is usually a separate network from this, maybe, which leads to its own satellite transponder and antenna array on the aircraft.

The other network is the one all the avionics talk via. On modern aircraft, it's Ethernet-like. It's not quite ethernet, more slotted and with QoS guarantees and priorities. Basically it has real-time extensions added to it. They are not compatible with each other. It is NOT IP based at all, relying on proprietary protocols and addressing. There is a bridge device that allows data from the avionics network to be passed to the inflight network, but not the other way around. The bridge does not allow communications the other way because it lacks the ability to transmit on that network.

On older planes, the network isn't Ethernet based at all, it's a completely proprietary protocol, and again, the bridge is one-way because they lack the ability to transmit.

The easiest way for a passenger to take over the plane electronically is to get through the floor. The cabling for both networks usually runs close to each other.

Mini-ITX is absolutely colossal compared to the NUC. Even the Mac Mini is gigantic in comparison

Not so much once you consider the Mac Mini has the power supply built in, versus the power brick that comes with the NUC. The NUC is physically smaller, (though taller) but coupled with the power supply the size almost the same.

I don't use it on my phone, didn't use it on my Disney pass, and would not use it for coffee either. None of these organisations have either the security awareness of credit card companies nor the statutory framework requiring them to cover losses where you are not at fault. I like to limit my exposure to the amount I add on

More correctly, I don't see the point

I mean, instead of Starbucks charging you $5 a day on your credit card, you have them charge $25 every 5 days? Doesn't seem to beneficial for me.

It's auto-top-up, so it's not like you can't spend $25 in a day and have it charged once instead of 5 times.

So what's the benefit, other than the company only paying the credit card per-transaction fee once, versus 5 times? And giving them access to your billing information.

Especially since Starbucks takes credit already - is there some sort of benefit to using a gift card that refills itself over your credit card?

If governments didn't back student loans, there's no way in hell banks would give them

And this is the real crux of the issue - you have to remember a student loan is for a rather large amount of unsecured credit. Anywhere from luxury car loan to low end house mortgage. And at least those two are secured loans because there is an asset associated with it.

Banks aren't stupid. They know a student loan has poor ROI - the interest rate is legislated, you're easily loaning anywhere from $10K to a good chunk of a million dollars and payback only starts years in the future. Oh yeah, and in the end, there's no asset they could repossess and sell to recoup losses.

it's why the government has to give guarantees on the loans as well as limit methods of discharge - there is otherwise no sane monetary reason why any financial institution will loan money for education.

Personally I find it amazing that we still need "labels" and "publishers" for anything nowadays. There's a huge global distribution network out there available for almost no money, there are facebook and twitter addicts that will gladly do massive "word of mouth" advertising for you for free. I think some artists need to realize at some point that in a world where pictures of a blue/black or white/gold dress can take the whole world by storm almost overnight, maybe the reason no one wants their music is actually because it sucks.

Well, there's a filtering effect that labels have so in general, the approved music will appeal to a great number of people.

Because if you really looked at it, most of anything is crap. There's tons of bands out there - it's practically a rite of passage growing up that some teens will get together can make what they consider to be music. So much so that there IS a lot of crud. Sure the best indie musicians can rival or exceed that of the manufactured stuff, but finding that gem requires wading through a an ocean of sludge.

And far too often the good is often heavily buried. Marketing is hard, and there's probably a number of good musicians who left the scene because their work died in obscurity.

Sure you can promote yourself on twitter or facebook, but you're just going to appear as mere noise in the crowd. "Going viral" is extremely difficult, happens rarely, and is fickle and unpredictable, and once your 15 minutes are up, you'll be forgotten all over again.

At least Steam rides its own platform rather than demanding the introduction of proprietary extensions into a tool meant to browse the open web. Don't get me wrong, I'm a Netflix subscriber and I'd like to get my content easily, but at some point I have to wonder why they don't develop their own desktop apps to support the client side.

Then you get people complaining they should be using the browser instead of their own apps which only work on Windows and OS X, versus a web app which works on Linux as well.

You already get people complaining on mobile that "apps" should go away and be "web pages" instead, and now you're advocating the reverse - that what WAS web pages will now be apps.

Well geez, what happens is you go to Netflix, select your movie, and then it launches the Netflix app, like what iTunes does now. So the web becomes just a portal for apps and to do anything requires installing bunches of apps.

Want to download music? Here it is on iTunes, now install the iTunes app to purchase and download. Want to download movies? Well we have the iTunes app, the Google Play app, the Xbox app, ... just a portal for apps.

I mean, on Windows 8, Netflix has an app. I don't think they have one for OS X, but who knows.

Still, if Netflix did restrict their service to apps, you'll find a bunch of Linux users suddenly complaining that it doesn't work anymore. And probably a bunch of people whose friends or children upgraded them to Linux and are now unable to enjoy their Netflix.

There is no good solution. Mozilla's solution is probably the best - sure it's "unpure" and "not ideal", but it's all about compromise and realizing that users will do what they want to do. If Netflix doesn't work on Firefox, no amount of "DRM is bad" philosophy will let them watch movies. They'll take the path of least resistance, Google 'how to get Netflix to work on firefox" and see the solution is "Install Chrome" or "Use IE" or "Use Safari".

It's all about picking your battles. No point in winning the battle by excluding DRM only to lose the war by being marginalized.

Or, and I'm just spitballing here, people could be adults and not think they're being cool and edgy by putting a dick somewhere on a map. Personal responsibility and all that.

Most likely it's because of those well-publicized dick images on the map. And the famous Android-peeing-on-Apple image. And a few others out there.

Quite likely what Google wanted was a way for its maps to stay updated, and what it ended up with was a virtual bathroom wall, with the exception that not only is it useless, it makes Google's map data even worse than useless if Google tries to rely on it for navigation or routing purposes.

"Turn left onto Penis way, then right onto Vagina Road".

Then there might be a concerted effort to "google-bomb" Google Maps and force a bunch of people down side streets that were reclassified as thoroughfares and all that.

The funny thing is - OpenStreetMap doesn't seem to suffer from this problem, despite it crowdsourcing all the map data.

I'm not sure why a simulator would ever want to bash people that hard. You'd think it'd be almost more jarring to have the simulation just stop completely -- lights go on, screen dark.

Easy, to simulate the event. If you're supposed to do something just before crashing, it would help if you simulated enough of the action so you can prepare for that as well.

Simulators are designed to train so that in emergencies, responses are practically reflex. If you're about to crash into the wall, you need to let go of the wheel or you risk getting sore thumbs from the simulator. Do that a few times and the driver will automatically let go even in a real crash (where thumbs may be severed instead of merely hurt).

Likewise in the helicopter sim - if you're going to land hard, make sure your tongue isn't touching your teeth. And other preparations you may make. It probably won't hurt as bad, but a little of it means if you're in the situation, your reflexes will make sure your tongue isn't where it might get cut.

One can be "in the right" and still not have done the right thing. For example, if the light is green I'm in the right not to slow down for the intersection. But that doesn't mean I shouldn't take precautions to check if someone is coming the other way. If I had I might have avoided the accident that was not assigned to my "fault".

Depending on where you are, even if you had the green, you can be assigned partial fault if you hit the idiot running the red (turning right on red, while legal, is technically running the red).

Sure it probably won't be 50-50, but maybe 25% your fault (you should've slowed down when you noticed the car pulling out) and 75% his fault. Even 10-90 is possible. The only way it would be completely the other car's fault is if they pulled out and there was no way you could realistically slow down or avoid the accident.

I don't think they really can. Companies set their prices higher because of simple supply and demand. Australians still consume these products at the higher prices, even though those prices are way above the minimum wage/cost of living differences between the US and Australia. It would require the Australian government to introduce specific price gouging legislation, aimed directly at media companies, and include wording comparing AU to other world prices for the product.

I thought Australia was already doing this by encouraging people to access stuff like Netflix using VPNs and such, even going so far as to have Australia Post set up a virtual sorting center in the US so you can order goods and have them shipped within the US, then bulk-shipped to Australia.

Of course, the problem might really be that the laws are such that to remain profitable, you have to change higher prices. I mean, think of a simple law like mandating that consumer products get 2 years of warranty. Pretty innocent, except it really means you're agreeing to the extended warranty - what may cover 90 days in the US with a 25% extended warranty to 2 years means that warranty price is built into the Australian price.

I mean, compare prices of stuff like Apple products - they actually turn out to be fairly comparable after you account for warranty (the EU and Australian models build in the price of AppleCare), taxes/VAT, and currency differences. Within a couple of hundred dollars in general due to currency fluctuations, but hey. (I pick Apple because they've been fined in the EU for selling AppleCare unnecessarily).

The real problem would be repealing the legislation - consumer groups rightly will protest, and some businesses will rather pocket the cash as extra profit. Though maybe ads of "NEW LOWER PRICE!" can help push the savings down

By all means allow multiple levels of service. Let customers flag some traffic at their routers as high-priority which gets better latency guarantees, of course at a higher price. Then users playing games could choose to have super-low-latency connections for the stuff that matters.

Wasn't this part of IPv6 - that QoS was built into the protocol? So yes, you can mark traffic as high priority and be charged for it as appropriate? It's handled in the routers so it seems like a perfect opportunity to monetize and speed up adoption of IPv6.

Of course, then the next malware hack would be to flag ALL your packets as high priority so you pay more...

This is not related to the SSD. If your cpus are pegged then it's something outside the disk driver. If it's system time it could be two things: (1) Either the compilers are getting into a system call loop of some sort or (2) The filesystem is doing something that is causing lock contention or other problems.

Well, it could be more than two things, but it is highly unlikely to be the SSD.

One thing I've noticed with fast storage devices is that sometimes housekeeping operations by filesystems can stall out the whole system because the housekeeping operations assume the disk I/O will block when, in many cases, the disk I/O completes instantly and essentially does not block, causing the kernel thread to eat more cpu than intended.

True, however, it seems to be caused by the SSD. As in the same machine with SSD and HDD, the SSD will cause the issue, the HDD will not.

And that's the real level of granularity I have into the problem.

I do note it only happens when there is a lot of I/O going on - even the simple act of tarballing a big build directory stalls out (I was actually trying to avoid this issue with the 840EVO by simply refreshing my build tree by tarring up the build onto the HDD, then deleting the SSD, doing a TRIM, then untarring).

The problem is it's not consistent at all. A similar PC (same model, different SSD and HDD) using an 840 Pro (and now 850 Pro as an upgrade) never suffered from the problem.

And given no one else seems to have found the issue with Linux, I'd hesitate posting to the LKML - the 840 Evo's have been out for ages, and if there was a real problem, it would've been reported.

It's just strange when you look at the CPU graphs in Gkrellm and it goes from blue (user) to all orange (system) time and even it stalls out. Like the kernel goes into some sort of introspective state where it contemplates the universe and ignores everything else.

Like I said, it may not be the SSD, but the SSD seems to be an important contributor to the problem.

As to consistent hardware, you're saying this like this is hard or special or something. PC game makers do just fine with variable hardware. Yes, consoles have consistent hardware but that doesn't mean much. That just means you have one version of the operating system with one set of drivers that are slightly better debugged than what the PC people deal with. So what.

It does generally make a big deal because the few AAA games around often break on "different" hardware which can involve merely owning an AMD video card on an NVidia game. Or vice-versa.

Luckily, the number of AAA games on PC is diminishing, and the indie market is exploding, where instead of driving each card to the edge (and thus causing all the problems), indie devs generally code for a common baseline, even Intel graphics.

And you have to admit that Windows does an impressive job at smoothing out the differences. Because back in the days of DOS, things were way more "exciting" in terms of handling differences. Nowadays, Windows presents a generally consistent API set so it doesn't matter what sound card, monitor, etc., you have.

Though, the BIG reason for consoles is easy - piracy. With PC piracy rates above 90%, developers look to consoles because of the vastly lower piracy rates. So when they make a game, the ROI is in consoles and if you make enough money, the PC port will hopefully pay for itself.

Why do you think everyone practically uses Steam? It's a cheap and easy DRM system that comes for practically "free" and offers enough friction that those who buy it will buy it, while those who pirate will pirate, and in some case, it's actually possible to track pirates. Doesn't do much to stem the tide of piracy, but the PC port of most games is a write-off anyways.

Communism fails when anonymous assholes can take advantage of you. Warning signs is when you feel someone is taking advantage of you, but don't know who.

Capitalism fails when rich assholes can take advantage of you. Warning signs is when you feel someone is taking advantage of you, but you have to cooperate with them anyways.

The reason we're stuck with sub-par economic systems is basically because human thinking is small.

Communism, capitalism, they work. In small groups. The problem is humans generally believe that if it works in small groups successfully, it will work in large groups just as well.

Think of it this way - there's a reason why we have two schools of economics - microeconomics, which deals with the economy on a small (personal) level, and macroeconomics, which deals with the economic on a large (city, state, country) level.

What applies to one system doesn't generally apply directly to another. It's basically the reason why we're in what we're in - we keep electing politicians who say things that DO make sense on a small scale, but do not scale and end up going horrendously wrong at the large scale.

Perhaps we need to scale it up from micropolitics that works at a family or village level to macropolitics that applies at a city/state/country level.

This is false. A pilot with a private (non-commercial) license may fly a passenger who reimburses them for the expenses of the flight, including plane rental and fuel. It doesn't become commercial until they make a profit.K/blockquote>

FALSE.

They may only reimburse you THEIR SHARE of the costs.

The FAA is very clear on this. There are several conditions that must be true for it to be OK to carry a passenger and get paid without a commercial license.

FIRST, the passenger's trip must be incidental to the flight. I.e., you the pilot would be going there anyways. So if you were flying off to see a ball game, and a friend asked if he could join you as he was going to a wedding, just fine. However, it is NOT fine if your friend asks if you would fly him for a wedding, and in the meantime, you discover you can go see the game. Your decision to make the trip must be yours and you would've completed the trip without the passenger.

SECOND, the passenger may only pay for their share of the expenses of the flight. So if you were flying as per above, your passenger may not pay the entire cost of the fuel. And of course, he cannot pay you for your time. If you rented the plane, he can pay for his share of the plane rental, but you have to be out of pocket as well. So if you're sharing your own plane, you're entitled to get back from him half the fuel, and half of the general running costs of the plane.

The FAA has made these clarifications even clearer because of aviation inspired Uber-type companies. Of course, this makes it truly a ride-sharing enterprise since they're paying their way, and you're paying yours.

Profit is not the only thing that makes a flight commercial. And yes, you need to keep good records because the FAA has busted pilots on this.

Of course, getting your commercial license isn't that much harder over a private's (especially if you stick with single engine land - you don't need multiengine to get a commercial, though if you do upgrade, you will need to take the commercial multi-engine), so if you really want to skate the line, perhaps it's best to just go for it. Then it's all free, legal and in the clear and you won't worry about the FAA. Especially about the first point where the trip is incidental and it boils down to subtlety. Did you really want to go there or did you make a special trip?

It's a lot like the ham radio "non-commercial" rule.