'Anonymised' is mostly a weasel word. It isn't always impossible; but the more interesting the dataset is, the more likely it is that there's a clever re-identification attack with good odds of success. If you are serious about preventing those, you tend to have to nuke the data so hard that they aren't of much interest anymore.

Unless robustly demonstrated to the contrary, it's an essentially worthless claim.

If you don't trust an OS vendor, isn't using a network monitoring tool on a different host entirely, with physical access to the wire, pretty much the only way to go? If they were so motivated, the OS would basically be a rootkit with device drivers and a userspace API...

TFA specifically notes that the behavior described was observed with all visible 'privacy' settings adjusted. Presumably the story is even cheerier if those aren't switched off.

If /. was put in charge or anybody who thinks the way an average /. user does, then all sorts of things will qualify as trolling. All of a sudden difference of opinion qualifies as 'trolling' on a government level and is punishable by jail time. Vast majority may find it wonderful, that opinions of a minority are qualified that way, however everybody is in minority opinion on some things.

I'd go with 'no' and 'no'. Yes, the end goal is to discover the cause, the mechanism, and the effect as precisely as possible; but the universe of possibilities is absurdly gigantic, easily larger than you could ever afford to study.

So what do you do? You start by trying to cut the search space into more manageable chunks with this sort of study, which doesn't provide the level of precision you ultimately want; but can (relatively) cheaply and easily provide some leads on what is worth looking at in greater detail and what isn't.

Here is my take on it, I have a number of people from well known schools, same number of people from schools that nobody heard from and in some cases people from unknown schools who dropped out and started working for me when I offered them a job. I have a number of people that had no formal higher education at all and a couple of guys that didn't know much about computers before they started here.

AFAIC I care about the attitude, I care that the person can work within a team, that I can work with the person, whether they are eager to learn. This is a starting job for all of the people I hire, very few of them worked in the field before, I do not pay much but that is also part of the equation. The people that do not have student debts do not have the same problem as those with debt, they don't need to try and get a highly paid position right away and so they can afford to work with me, where they are gaining more than enough experience so that eventually they are propelled to better paying jobs.

I will say this: I have about equal number of good coders, whether they had any formal computer science training or not, but I go through many people to find good ones but it does not take much time at all to know who is who. Just in the last 2 months I interviewed about 15 people, 3 of them ended up with me, 2 are going to be excellent developers. Out of the 15, 4 decided it was too hard in the first 3 days. 2 decided that they made a mistake and shouldn't be in this field right during our meeting.

In the interview all I do is I show them what we do, how we work, ask them what they like to do and explain the structure here: you are studying here before I put you on an actual paying project. I teach you what you need to know and while I am teaching you, I am not paying you anything. Normally takes 2-3 weeks for a developer to go through training and start being productive. This is my way of doing stuff, I do not care where you studied, what you think you know, I only care to figure out if you are OK for the team and if you can learn and have the right attitude for this company.

They do do that already. Their prices are what they are and not 2 times what they are now for a reason as well. Of-course actually building Apple hardware products is a costly endeavour, nobody else can really afford to do what they do to achieve the save level of user friendliness and beauty. I don't buy Apple products (not because of price, I just do not like their software), but plenty of people do. Their products are priced to satisfy demand of a specific set of population, however if their revenues did not cover their costs and did not make them good profits, they would have lowered prices by lowering the cost of production and very likely by sacrificing build quality to gain market share, however their customers are used to certain degree of quality and shine, and so Apple spends more than anybody else delivering just that.

If Apple could actually cut their production costs in half, while delivering the same quality of product, their prices would go down to increase market share, but they are in business of providing top quality systems (even with some problems, like the new phones that probably bend much easier than some others).

If you in fact believe that you could provide similar quality of the product as Apple for half the price, you yourself could do what you are talking about, but apparently it is not as easy as you say, since other companies do not provide the same, shall we say 'shine' in their hardware so far, while being able to charge much less than Apple.

It's amazing how quickly that happens. I work in an industry where we have a very cyclical business climate, so we have frequent layoffs. It usually keeps the engineering staff pretty top-notch. We haven't had a down cycle since the 2008 crash, so the cruft has certainly built up. I can only imagine what happens at a place like Google where the only turnover is people quitting!

I feel the need to plug the Pandas module for Python. It does a lot of R-like operations on huge datasets. It takes care of time-series alignment and has many other nicey-nices. Basically almost everything you think you need to invent to manipulate your dataset is probably already implemented in Pandas.

Or a gigantic application, like Photoshop or MATLAB, that sits in your "Applications" folder. Have you guys never heard of the term "killer app"??? Am I that old?

I think "app" came naturally to long-time Mac people, since there has long been an "Applications" folder which a lot of (most?) people just called the "Apps" folder. This lingo followed the jump to mobile on the iPhone and Windows people were introduced to the term for the first time. MATLAB has long been an "app" for Mac folks.

I get my view of how the lower class is treated by the cops, from watching the TV show "Cops".

Understandable, then why you hold your views of the poor.

Generally the poor are poor because they make bad choices.

Some people, true, but not generally. If you're raised by poor parents, you're up against a very big wall. It's hard for them to buy school supplies, and if you're raised by a single parent it's harder for them to help you because they're probably working two jobs. Having a single parent is the result of a poor choice, but it wasn't your poor choice.

And a poor kid can forget ever attending college.

People who have bills to pay and have to go to work in the morning don't have time to be out causing trouble.

Most of the US's poor work, and are not criminals. But the cops still treat them as criminals; hell, society itself does.

Those folks working at McDonald's, WalMart, the corner convenience store are all on food stamps. That friendly face you see at the checkout counter is the face of poverty, not the idiots you see while watching "Cops" (that show is government propaganda, BTW).

So we are a classless society, in that you're not stuck in one socioeconomic class no matter how hard you work.

Rags to riches is extremely rare and takes a hell of a lot more than hard work. My late uncle was one of the exceptions. Creativity and hand-eye coordination runs in the family, and a stroke of bad luck was the best thing that happened to Dan.

He was injured in WWII in the navy, and became friends with a fellow patient in the hospital who had lost a leg in the war. The army gave him a prosthetic, and Dan saw it and said "I can make a better leg than that" when his friend showed him the new leg, and he did. They went into the prosthetics business, and all it took to sell one was for Dan's partner to talk to a recent amputee, who would invariably say "What could YOU possibly know about it?" All he had to do was pull up his pants leg and it was an instant sale, because you would never know that he was missing a limb.

But there were so many lucky breaks, including genetics, that his rags to riches story (both sets of my grandparents were poor all their lives) would not have happened had a single thing, especially meeting his future partner, who was a born salesman; that's something that doesn't run in the family. We couldn't sell a ten cent hamburger to a starving man.

The fact is, if you're born poor you're almost certain to die poor, and if you're born rich you're almost certain to die rich. If you're born middle class there's no telling; you could die rich, middle class, or poor.

If this society is as classless as you say, then why did no one spend a single day in jail over the banking crimes that brought down the economy? Why did no one from Sony go to prison over XCP? Why was OJ Simpson "not guilty"? Do you think a poor black man would have been treated the same? Hell, a poor white man would have gone to prison under those circumstances.

What are you talking about? We had good 128-bit encryption back in the Clinton years; that's what the whole "weak vs strong" encryption issue was about (only 40-bit crypto was allowed to be exported).

I actually agree on the "old style" Thinkpads, but recent models look more and more like Lenovo's other stuff so I don't know how confidant I am. We all have Thinkpads at work and even the last generation was uneven, with the cheaper models acting... well, cheap. The old-school stuff was very high-quality.