I kinda have to disagree. You're quite correct in your analysis, except the whole bit about your terabyte-years being the cheapest which was the point if your post.

More accurate may be that your "terabyte-warranteed years" rate is the cheapest, but in terms of actual usage, many people may disagree with you. I haven't had a Seagate drive fail since 2001. I think the oldest I have in a system somewhere is 2004, but that's besides the point - that drive is priced out "terabyte-years" where years = 10. I have at least a dozen drives with 2-year warranties that are still running error-free after 5 years.

Therefore, I can't agree with your conclusion that paying 50% more for a longer warranty is worth any more at all. Most consumer drives simply don't fail very often anymore,

No.

An operating system should be just that. If you want a version of Windows bundled with Office, great, but the OS should be exactly the same in both cases.

As for "useless services running in the background", while many might not be necessary, most users won't know which usage cases would require which services. You can open up a list of all services and decide which you don't want to start, already, but presenting users with this choice is pointless. If you're thinking "for the games, just remove all the ones that gamers don't need", then just stop. Gamers use a lot of services, even ones you'd never think games would use.

Do you know which services are used by all game DRMs out there? Which services are required by emulators? Which services are required by virtualisation? Which services are required for syncing all this with your 3D TV? Which services are required to handle the decryption of your blu-ray movies? Which services are used to handle the authentication of that? I could go on and on, but I'm working so must cut this short.

Thinking that a "gaming OS" would just be "OpenGL drivers and something to let you add hardware" is very short-sighted. Our modern systems are fairly complex as are our modern games. The tools used by developers require far more of the OS than you can imagine, and if you think Windows loads a bunch of useless services, then I suggest firing up SteamOSwhich *is* a dedicated gaming OS and look at just how many processes it requires.

If the 'oblivious' drivers are driving at the speed limit, however, the cars wouldn't tell each other to "move over", and you'd still be behind them. You reek of self-importance.

No, not really - and as I was writing it I thought "I bet someone's gonna bring Moore's Law into this and then I'm going to have to explain". So I'll explain - the 50,000 years was a figure thrown out there. Really, as long at time taken > life expectancy, OP won't be able to find a result. The actual time to perform that many encryption cycles would be in the millions of years. If Moore's Law progresses over time that would certainly be brought down, but not within OPs lifetime. Then you've got to compare the data set. Nevermind that physically storing that many 32-bit strings would take more atoms than exist on our planet. The point was simply that OPs suggestion was ridiculous.

And when you're done in 50000 years with our current supercomputers, let us know the results. The number of possible combinations is a bit over 170141183460469231731687303715884105728. Good luck with your bubble-sort.

Yeah, my thoughts entirely. Was a little surprised this was making "news" since I've been using sat-based wifi whilst flying over the Pacific for a few years now on several carriers.

It's just a joke.

Funny.. I've had a sore neck and within 15 minutes I'd been seen by a doctor and was being wheeled in for a CT scan. Maybe call an ambulance next time you're in Australia and have a medical emergency?

On my debit card I get the regular VISA protection which I've had to call on a couple of times to get fraudulent charges removed. My partner has had the bank call her to ask if her debit card was really being used in Manilla. Generally I guess the protection you get varies between banks and providers, but I've been happy with the service I've received. One of the limitations of a debit card is that you can't perform "I got the service I paid for but decided I didn't like it so want my money back" charge-backs.

Rewards, now there's the first real reason I've heard that it can be useful to have a credit card. :) I'm not denying there are any positives to CCs, was simply calling out OP on his inability to name them, and his labelling of debit cards as "evil and useless" as somewhat ridiculous considering the alternatives. Most people can't pay off their CC each month and end up in debt. CCs let people spend beyond their means - it's the whole idea of them. Unfortunately, also, many folks DO pay fees on their credit cards. Also, many people are stupid. Just saying, debit cards aren't, per se. :)

The opposite tends to be true.

If you have money to pay off the CC, then the credit card is unnecessary and stupid. Why would you need to maintain a line of credit just to pay it off with cash you already have? This is where a debit card's perfect. It's a credit card you can use everywhere and costs you less to run. While most online vendors may charge you a credit card fee even for using a debit card, almost no retailer ever will. To them, it's just an EFTPOS card. You gain the convenience of a credit card with the general fee-freeness of an EFTPOS card.

Where the evil comes in is that the whole premise of the credit card is the financial institutions hope that you *can't* pay it off by its due date, at which stage they get to start charging you interest.

If you have the cash to pay off a credit card, then I can't see any good reason to have a credit card instead of a debit card. To pay more fees? How's that a bonus for you?

Right, but it's only just gained keyboard support, so up until now it was a fairly boring text-demo.

> perhaps short of every participating computer having a quantum component that stops working as soon as you observe it.
Shh.. don't give them any ideas. ;)

Nonsense.. not only does it never mention 9 or 5, it splits the day into 6 even segments of 4 hours. If what you mean is "it *presumes everybody works consistent hours", that's not necessarily true either. Who says when you're working is when your "peak" productivity occurs? Note - it doesn't ask which time period you're most productive over, just at which point productivity peaks. If you can't work out an average for this, then just go with one of the joke options.

They don't even need to do that - in most devices the CMOS is a square and it's simply software which dictates whether the output is portrait or landscape. You could simply force it to capture widescreen even when held in portrait mode. Probably the reason they don't do this is it would confuse the folks who.. don't understand this stuff. "I'm holding it vertically, why isn't it recording vertically?" Actually - it's really just usability, but perhaps there should be an option on most of these devices "Always capture widescreen video".

Now.. some devices do have slightly wider than tall CMOS sensors, such as the iPhone 5s slightly landscape sensor, but a minor down-sampling of video resolution (since, if it's in portrait mode, the sensor is portrait so not quite as wide) would have little effect - especially on the majority of camera phones which don't record 1080p in the first place, so a "full-width" 720p widescreen video could be captured in portrait mode anyhow.

While this would the expected situation, the evidence demonstrates that it isn't.

http://www.zdnet.com/six-open-source-security-myths-debunked-and-eight-real-challenges-to-consider-7000014225/
http://www.theregister.co.uk/2004/03/05/does_open_source_software_enhance/
etc..

You can search this on your own. The general consensus is that the "many eyes" theory is flawed, and outside a few exceptions where a particular product has been security hardened beyond usual standards, most experts agree open source software in general tends to be no more or less secure than proprietary software. On the flip-side however, it is true that when the source code for a closed-source product does get compromised, we do generally get a new flood of exploits.