He may well believe that past results are no indication of future results, there's one overwhelmingly important fact that comes to mind: noone will be able to buy the stuff made in the robot factories if we're all unemployed or minimum wage serfs.

And if noone can buy the stuff, the owners aren't going to get rich selling the stuff. Which means THEY won't be able to buy stuff either....

No, but the rich folks will just own it all.

See Manna where robots take over. If you don't have a job, you get shuffled off to a jail-like facility where your basic needs are provided for at the whim of the owners who stuff you into a way-to-small box because they really want to sell that real estate you're "living" on with millions of other people.

So, the gubmint agency that built the Internet... "owned" by the same gubmint that built NSA wants to build the new TOR to increase privacy?

Sounds trustworthy to me.

You missed the fact the government created TOR in the first place... (TOR was created by the US Navy).

TOR is basically a US government project. Which is why it's funny when everyone says to go use it to protect your privacy. After all, doesn't the NSA run huge farms of exit nodes which can capture a good chunk of traffic?

I don't think they're getting that much for the ads. After all, netflix manages to offer ad-free stuff for $8/month, same as Hulu+. It's probably closer to the difference between $8/month and $12.

I think the ultimate reason Netflix is creating it's own content is that the more content it controls, the more influence it has over the other media copyright holders. If Netflix can legitimately argue that if copyright holder X doesn't play ball, that it's average subscriber won't sign up to site Y for $Z revenue because the subscribers will simply watch something else, such as one of Netflix's exclusive shows, then they're leaving money on the table, and they don't like doing that.

Sort of like a backwards HBO. HBO does great shows, but are really exclusive about them. If you want to see Netflix's shows, you have to sign up, but it's not nearly as expensive as a cable package + HBO.

Hulu however has current TV a day later. Netflix doesn't get it for months. Hulu's also sponsored by the content networks who are trying to basically regain their ad revenue.

Netflix is creating original content because it can - its business model depends on subscribers and growing that subscriber base. Showing unlimited movies that have been out for months, or TV seasons that everyone else has seen months ago doesn't grow subscribers, especially since OTA TV will get them for free too, just on a less convenient schedule.

Instead, Netflix has to basically create content or inherit content that the networks can't justify carrying so subscribers have something new to watch.

Hulu, Amazon, iTunes, etc., they get the latest TV within hours of airing which is why they generally cost more because their first-run. Netflix picks up the rest.

In the lifecycle of a movie, it first comes out in theatres. Then it comes out as a digital rental (CinemaNow, Vudu, etc). Then it comes out as purchase - either digital (iTunes, Ultraviolet, etc) or physical (DVD/Blu-Ray). Then general rentals, then Netflix, and finally, regular free TV. This takes around a year or two to fully execute.

The lifecycle for a TV program is first airing, then digital sales (Amazon, iTunes), and Hulu. Then months later, season box sets on DVD, and Netflix.

If you're not fussy about waiting, Netflix is a great service. Most people though can't wait that long for their TV, so there are options. And Netflix knows once they have the people who don't care or who don't mind waiting, their subscriber base is saturated. They need to have new content to attract new subscribers who may not watch much of the catalog, but will catch the exclusives and pay for it.

I'm a huge IPv6 fan and I don't game. What IPv6 does is recreates the pre-NAT world of easy communications between systems. Going back to the symmetrical world where everything on the internet is a server simplifies commuting immensely. That's why I want IPv6.

Hah.

Easy communications? Fat chance. Because there'll be firewalls in the way leading to plenty of issues - enough so that assuming you can connect between two hosts is not a safe assumption.

In fact, you'll return to the early NAT days when they were rare, and spend hours trying to figure out why your VOIP app works half the time, but when someone calls in, you can't talk, at all because someone has a firewall in the way and it's blocking the connection.

These days, NAT is pretty much understood and it's easy to detect and work around. Tomorrow, with IPv6, people are going to forget their NAT lessons and we'll go through the same pain that we had a decade and a half ago.

And let's not forget the nice corporate firewalls that already exist today and filter everything that's not HTTP, HTTPS, FTP or SMTP. Just silently dropped. Those will be really fun to diagnose.

And work firewall-less? This is the modern internet, and remote vulnerabilities, spoofs, amplification attacks and others are just sitting there waiting to be discovered.

The myth of apparent reachability died ages ago. Along with the ability to plug a home PC straight into the internet without a firewall.

An off-the-shelf SATA 840 EVO SDD hits 98,000 read IOPS, and all those tasks you mention added together wouldn't hit more than 1% of that. They're the very definition of network bound operations. The average email in my IMAP spool right now is 43KB and would take 11 4KB operations to completely read from or write to storage. Browsers site there idle 99.9% of the time. IRC? Not that I've ever seen.

1% of 98,000 IOPS is 980 IOPS. The best hard drives still only manage around 100 IOPS (10ms seek+latency).

Application loads generally produce a lot of IOPS, especially since they load dozens of libraries and configuration files and other things, and the CPUs can be so quick they're lapping it up as fast as possible. Do it on a hard drive and reading 1000 4k files scattered about takes 10 seconds. On an SSD, that takes 1. It goes from "ugh, takes forever" to practically instantaneous.

And that's where the real value is - it increases interactivity because disk I/O suddenly gets very cheap. Especially in Windows where the document model is application centric - you open a document, Word opens, you work on it, then close the document that also quits Word. Then you open the next document and Word reopens. If you're lucky, the disk cache has it so it loads quickly. If not, thrash heaven.

And I've seen people double-click something 3-4 times because it took so long the first time they didn't realize it was just grinding away. Or they needed to open two word docs, an excel spreadsheet and a few other things at the same time.

If SSDs were completely bunk things, then the apparent snappiness that it gets people must be fake, too, right?

Actually, yes. These loans can't even be dismissed via bankruptcy and many government benefits are tied to the repayment of them.

Because no lender in their right mind would loan students money to begin with.

Student loans are unsecured - the student doesn't put up anything and can borrow easily $100K. And if the student can't find a job, it's not repaid. No lender would take that kind of risk, especially at a measly 7% interest rate - typical unsecured loans are closer to 20-30% (you know them as credit cards).

So the government has to put in some incentives otherwise financial institutions will not provide the loans as they're too risky. By making it non-dischargeable, the banks know that declaring bankruptcy will not make it go away for pennies on the dollar with no recourse (assuming they can get anything - secured creditors are paid first and often the money isn't sufficient to cover those so those take pennies on the dollar. Unsecured creditors typically get zilch).

It's just like if your company goes bankrupt - as an employee, your unpaid wages and benefits are unsecured, so you line up with the rest of the unsecured creditors to hope to get some money.

Back to Tesla, those DOE loans went to a lot of people. Few companies every paid even a portion back. Tesla managed to pay it back, early, with interest.

and fuck the attendant, who is probably complicit in the scheme.

Actually, the attendant pointed out the camera several times - and the fact that she was being recorded.

Even more than that, she identified herself as a celebrity and that she would post some crap about it all over TV.

So no sympathies at all. Even less so when you're trying to exert influence over others by using your authority.

And definitely no sympathy for her situation, because it was pointed out that everything she was doing and say was being recorded.

Perhaps her car was towed illegally. That doesn't excuse you for harassing the attendant and trying to use your position of power to influence them. Especially as the attendant repeatedly notifies her that she's under surveillance.

Anyhow, the response is not to be "nice" or "polite", it's to be "diplomatic". There's a time and a place for everything, and a time and a place where something is inappropriate. I don't know if it's social media or what, but it seems to have created a pile of self-entitled people who can muster up twitter to "shame" people and companies. Oh no, my package hasn't arrived yet? Tweet the company is awful and doesn't keep promises. Voila, package is now at door! Company didn't give me another free sample? Tweet, and there it is!

Unfortunately, the power company is still expected to make sure that the power comes in at the right voltage and frequency. And with control on only part of the inputs, that's a lot harder. The fewer inputs they control, the harder...

Theoretically, you can design a control system that'll handle the problem. But, so far, noone has bothered to, because noone's had a need to. As solar becomes more common that'll change, and the problems will go away.

Already exists, and Germany and California require all solar installations use smart inverters if they're going to hook to the grid. A smart inverter is a regular inverter except it's controllable as necessary.

They're able to do frequency detection (if the frequency drops, more power is needed on the grid, and if it rises, less power is needed and it should cut back. Even more advanced ones can invert with reactance.

Of course, it all means you can't dump all the power you generate into the grid - and if you're residential, it can also mean there's not enough transmission capacity to go from your house to the business and industrial district where it's needed most.

There's also something called "net metering" where the power company sells you electricity at retail price, and buys it at wholesale price.

Either way, if you're blessed with it, the general best idea is to just use it all - if you're in a sunny and hot location, well, use it to run the A/C when your system is producing maximum power (and maybe even use cold storage), because selling it back to the grid doesn't generally pay off,,,

Is it really common practice now to have laid off workers escorted out by security?

Depends on location. In the US, it's extremely common, potentially due to their more violent nature and the second amendment.

In a lot other countries, layoffs can take different schemes - they may provide notice of layoff - as in you're going to get a severance and all that, but it's a 2-week notice, and no, they're not going to buy you out, you're going to work those two weeks. Seems incredible, but a lot of companies do it because they want an orderly transition, and they do trust their employees enough to not be burning bridges. Some even go out of their way to help them find a new job (instead of just giving them the number of the employment agency) - and that includes counseling services. Heck, even benefits often continue for a few months after the layoffs (health insurance).

I don't know what it is about the US - perhaps their proclivity towards violence leads to basically shoving them out the door after the meeting is over - if you need any personal belongings, they'll fetch it for you and pack up the rest of your stuff.

Yes, but it took two whole years before the fix came out. And the fix was made within a day of the exploit being released.

Yes, I can understand 90 days being a bit tight if you're talking fundamental software like operating systems (which require a lot of testing, staging, and you lose some to Patch Tuesday), especially since root causing and fixing can require a bit of time. But two years is a bit on the long side.

More like the guy got ignored and once he released the code, the "OH SH*T" came out.

This is one of those struggles between what's right and what's reasonable... 90 days is a bit quick for something big like an operating system where a change can break everything, but it's also on the long side for something that only breaks something really minor, like Minecraft.

I seriously doubt it. I don't see how location reporting for a payment transaction in which location data is irrelevant could possibly pass Google's privacy policy review process. Collection of data not relevant to the transaction is not generally allowed[*], and if the data in question is personally identifiable (mappable to some specific individual), then a really compelling reason for collection is required, as well as tight internal controls on how the data is managed and who has access. I don't see what could possibly justify it in this case, and I can see a lot of risk in collecting it.

There's a lot of information involved in payment systems you think are irrelevant, but are passed on in order to judge if the payment is potentially fraudulent or not.

That includes things like location data (are you making a transaction where you normally make transactions, or in a new location (this is often the reason why credit cards are blocked when on vacaation), are transactions being made in impossible schedules - e.g., you used your credit card in Seattle, then in New York an hour later?). Online sellers routinely pass your shipping information to the card processor for the same reason - and can even ask for enhanced scrutiny. This verifies that the shipping address is known to the processor (either you put it on your credit card or it's been used constantly).

You may be confident that Google isn't collecting the data, but it's made available as part of the standard fraud checks. It's also why there's slightly more pushback against Google Wallet (where Google does get all the information involved in the transaction) than Apple Pay (where after setup, Apple is out of the picture and the information is shared between you, the retailer, the processor and your bank).

Oh for fuck's sake.. it's very simple: Avionics need to be on a physically separate network from everything else, preferably encrypted. If there was 'air gap hacking' going on or even possible, wouldn't we have seen it long before now? Wouldn't an intelligent, capable, well-organized, well-thought-out terrorist (yes, Virginia, they do exist) have found a way to sneak the equipment necessary aboard a flight and implemented his hack, taken control of the plane?

Exactly.

And yes, it's possible to "break through" the airgap - cellphones are known to cause EMI issues with certain equipment on certain planes (e.g., lose GPS lock, increase INS errors, cause drift in the heading indicators, etc).

If you really wanted to cause problems, I'd go with a broadband transmitter that causes EMI in the airgapped control network more so than trying to hack it through in-flight WiFi.

The big problem with Wikipedia is that in spite of what the publicity says, it is only a small number of people who contribute, and a surprisingly large number of those people have an agenda for what they edit.

imo, with Wikipedia, truth is not the goal. A certain point of view is the goal.

No, the big problem with Wikipedia is politics. Wikipedia is the reinvention of communism, and it's proceeding just like Animal Farm and other Communist nations down the path to failure.

Heck, it's already at the "Everyone is equal, but some are more equal than others" stage.

That's the main problem - you have editors and higher ups who now patrol their part of Wikipedia who are not interested in the truth, correctness or other aspects - just in having little power struggles. Heck, for a time there were massive parts being deleted for arbitrary reasons (usually along the lines of "this content is not suitable for Wikipedia" despite having plenty of similar content around). And these days, well, edit-reversions by the same power-mad editors have basically rendered any reason to edit it moot.

I mean, there's a small amount of contributors because everyone else got driven away. Try to fix a mistake and you'll et into an edit war with an editor who thinks their interpretation is completely correct even if it's obviously wrong.

Yes, it's an encyclopedia anyone can edit. Except that if you do so, chances are someone will revert it in a few minutes because they don't agree with what you edited, even if all you did was fix an error. "Everyone has equal edit rights, but some people have more equal edit rights".

The study of Wikipedia itself is quite fascinating, no many times you get to see political ideology put into play and see the results. Usually you end up with people getting hurt or humanitarian crises if you try to experiment.

OS X does not support NVME, so no, there are no NVME drives being supplied to Apple. And finally these NVME drives are just entering production and are not available in either the OEM or Retail channels yet.

The new MacBook supports NVMe on OS 10.10.3, so support is rapidly coming...

Their reasons involve context switching and interprocess communications. Context switching has got to happen (unless they run IE in kernel space) so just get it over with. Interproces communication has always been a weakness in Microsoft systems. Since day one. Multitasking OSs are here, folks. Get over DOS.

The bug here affects the HTTP server side, not IE.

And in HTTP servers, there are LOT of context switches - in basic static file handling mode, you read a file (syscall to read file), then you write it to a socket (syscall to write to socket). in effect, a webserver is just copying from two file handles, and incurring a kernel-usermode transistion twice every round.

Add in a moderately busy webserver and you could be spending significant amounts of time just switching between modes.

Using larger buffers helps, but if your site consists of lots of little files, it's still the bottleneck.

Linux has similar functionality - see sendfile(2) and splice(2), among other commands to actually manipulate in-kernel memory buffers.

In fact, doing it in the kernel has an added bonus - if you support zero-copy, no copies are made rather than potentially having to copy to/from userspace (more overhead).

Of course, in the Linux model, all the processing happens in user made and only the tedious file copying is accelerated which ups security.