I think you just answered your own question there. Because not everyone can do it.

Of course, the fact that you did something that anybody can do but almost nobody does can be cool too.

I make stuff all the time. Just the other day I made a belt sheath that holds a pair of kitchen shears and a pair of forceps for when I'm fishing from my canoe. I made the sheath out of duct tape, which took me about five minutes. Anybody could have done it, which I think makes it a cool project. But suppose I'd sewed the sheath out of fish leather that I'd tanned myself. Not anyone could do that, and that would be cool too, but in a different way.

Well maybe Putin's going around cutting everyone's hawsers. That would leave them towing the line.

Well, this is what you signed up for.

Google's ToS explicitly states you can't use it for anything illegal, and their privacy policy states they can poke around in anything you send through them to make sure you're complying with the ToS. Furthermore their privacy policy allows them to disclose your data to third parties in order to "protect against harm to the rights, property or safety of Google, our users or the public as required or permitted by law."

So Google in this case was doing nothing that the user in question didn't grant permission for. Like most Google service users he didn't bother to read and think about all the documents he was supposedly agreeing to when he signed up. But it's not that hard to do. I did it, and I periodically check for changes.

How hard can it be to replace Ballmer? Very, very hard. But I think that's not he question you meant to ask. That question, I believe is this: How hard can it be to do *better* than Ballmer?

Long, long ago, early in my career, I spent about fifteen years in the non-profit sector.

You don't ignore office politics, but you don't take sides either unless there is a crisis brewing -- something illegal, highly unethical, or financially dangerous. When you work in IT, you're in a "support" position, rather than a "line" position. Your job is to support. So when there's a big pissing match between two line functions, your job is to support *both* sides.

Often this means documenting business processes that sort of evolved via the lava flow antipattern; 50ish is the size where things start to get out of hand, because it's the size where the amateurishly hacked-together processes that keep the organization running start to break down because everyone can't be aware of everything that's going on in detail, in real-time. Make it your business to understand what business systems (not necessarily computer systems) *accomplish*. That puts you in a position to offer a third way, the one that emerges as obvious to everyone once somebody has figured out what's actually going on.

It's supposedly hard to implement changes in non-profits because of the consensus-driven decision making processes, but I found that I could make that process work for me. Lack of understanding is a vacuum; presented with a clear picture people usually line up behind the obvious solution quickly. But you do have to do your homework. Never surprise anyone with anything in a meeting. Bring people up to speed with things you're going to say about their work *before* the meeting so they don't feel blind-sided.

In a crisis be prepared to do the right thing. If you're in a non-profit they're paying you below market rates, so you can do better elsewhere. There is no call for getting yourself sucked into something that offends your self-respect. I resigned one job because my superior (the COO) was doing things that were financially reckless and improper (spending without proper authorization). I informed the CEO in my exit interview. That was my solution to the problem of not getting drawn into a persistent pattern of dysfunction.

When you handle sensitive information, just ask yourself what is the professional thing to do? Be discreet. Resist the temptation to peek at data, and when you *do* accidentally learn something you're not supposed to know, disclose that to the responsible parties. Be trustworthy, and present a trustworthy face.

Finally, don't let them pay you far below the market rate for your services, and expect a really good benefits package, including 1.5x to 2x the vacation you'd get in a for-profit. Insist on the respect due a professional. Non-profits are full of young people who haven't learned that the IT guy isn't there to be kicked around when they're frustrated, and the fact that you're in a support position rather than a more glamorous line position doesn't make your work any less important.

Implementation makes a difference. Early versions of NT were quite good, but unpopular because you needed 16MB of RAM (if I recall correctly) to run them in an era when a high end personal computer shipped with 4MB of RAM. Over the years they tried to hold the line, at one point getting the minimum down to 12MB of RAM, but perhaps not coincidentally stability got really bad.

With slave labor, no less.

Yes people are limited by their culture and time, but not *that* limited. Braun deserves condemnation for using slave labor in WW2.

All the arguments based on classical economic theory only work if the assumptions of classical economics hold, particularly the assumption that there is a free market.

Amazon is arguing for its freedom to set prices it charges in its ebook store; that would be no concern of the publishers if we lived in a world where ebook users could simply buy books in non-proprietary formats from any Internet storefront they wanted. But we don't live in such a world. We live in a world where most ebook readers are controlled by Amazon and inextricably linked to its store. It wouldn't have been hard for Amazon to build the Kindle that way. Define some public book trading protocols, bootstrap the standard by building those protocols into the Kindle and Amazon's online store, and instantly the world is a better place for everyone except printers and bricks-and-mortar bookstores with no Internet presence. But Amazon didn't do that, because the Kindle is designed to tie the user to Amazon, the way the iPad is designed to tie the user to Apple.

So what we're looking at is a maneuver by Amazon to corner the market on books *in general* by killing off the traditional paper book trade. Preserving the ability to buy most books from someone other than Amazon seems like a legitimate reason to me.

Nobody's book is so incredible that people would buy them no matter what the price. If my only way to get Shakespeare was to pay a ten thousand dollar license fee I'd find a way to do without.

You really shouldn't assume that anyone who disagrees with you does so because they're stupid. Publishers know their marginal and fixed costs and certainly have a pretty good idea of the price elasticity of their books. The situation is more complicated than you know.

You can't compare Hachette to Valve, because Valve owns the whole Steam ecosystem, and delivers its services to users' commodity PC hardware with no intermediaries (other than Internet service). In the case of Hachette v. Amazon, we're looking at a situation where Amazon owns the point of sale, and has more control over the users' devices than the user himself has. And yes, you can read ebooks on a PC but few people will want to do that. And yes you can download ebooks in non-proprietary formats like epub from sources other than Amazon, convert the format to .mobi, and use file transfer to move the converted file onto the kindle; but that's a significant barrier for most people.

So what we're looking at is a move by Amazon to take control of the book market in a way it cannot as long as paperback and hardback sales remain strong. Amazon *looks* like a friend of the consumer because they're calling for lower prices. If they get what they want, then ebooks may well make a significant market share headway against paper books.

You might think that's fine, but it's not *generic* formats and *commodity* hardware we're talking about. It's formats and hardware controlled by an inextricably linked to *one* company. And that may mean lower prices today, but what will it mean ten years down the pike when Amazon corners the market on books?

I'd be interested in your source for this particular tidbit, particularly how it shows blacks are any worse than whites in this regard.

I went to high school with a lot of tough white guys from South Boston and Charlestown in Boston, back when Whitey Bulger was still a big deal in Southie. Let me tell you most of them didn't see intellect as their path into the middle class. A few did, but not many. I've also worked with PhD scientists who were black and came from urban black neighborhoods. You get a mix of attitudes everywhere, whether it's in a black ghetto or white ghetto or a middle-class white neighborhood, but usually being academically advanced doesn't make you popular unless you live in town with a big Jewish population.

Speaking of Whitey, his people used to spread the myth that he kept drugs off the street in Southie. In fact he was kicking the Italian mob out of Southie so he could have the drug trade all to himself. Whitey wasn't a hero, he was a parasite. So why did people believe the lie? Wishful thinking. The people who got education and became professionals moved out of the neighborhood, so the one example of guys who rose in life that you saw every day were the mob. And you had to hope they were good lads at heart, because they had the neighborhood by the balls.

There's often a "we're all in this together" thing going on in poor, downtrodden neighborhoods. Part of that is a resentment of anyone who acts like their above the rest, and that includes people who flaunt their education or sophistication. But that's because intellectual accomplishments don't seem to be within the reach of everybody. You don't get that attitude in cultures which believe in self-improvement.

So let's *not* talk race. Let's talk education and economic opportunity. If people have a way up, see that way, and believe they can do it, they will rise.

Of course. But putting your servers behind a separate firewall isn't the same as putting them on the same network as the clients with *no* firewall.

In any case, we're talking about an in-store POS system with TWO clients. We're not talking about an Internet facing server that has to handle thousands of connections per hour. Even if the server had FIFTY client terminals the impact on performance would be nil.

Well, I think the first two films are a mixed bag. I rather liked getting meet Radagast, and to see what Gandalf was up to in Dol Guldur.

A screenplay adapted from a book has to stand on its own as well as live up to the book. Where the movies have fallen down is living up to the book. The consensus of my writer friends is that the screenwriting team (Walsh, Boyes, Jackson and del Tormo) doesn't trust Bilbo to carry the story, which deeply undercuts the themes of THE HOBBIT. Lack of respect for THE HOBBIT novel is pretty common among LotR fans. They often dismiss it as "just" a fairy-tale -- an attitude which would have disgusted Tolkien himself. It would have been better if writing this screenplay had been entrusted to someone who loved THE HOBBIT for itself, and understood it better.

Surprisingly, I thought the non-canonical character Tauriel was one of the best parts of the movies. Yes, she was there to give the story a so-called "strong female character", but that's a silly objection. Writers always put characters in stories for some reason; the question is whether they fit in and come to life. I think adding a strong non-canonical character is better than giving so much screentime to a weak but canonical one: Legolas. No disrespect to Orlando Bloom, but the writers dont' give him much to work with. The part could have been played by the CGI model they used in the action scenes.

One of the reasons I'm accepting of the whole Tauriel subplot is that it carries a deeply Tolkienian theme: the love between mortal and elvenkind. That was a profound part of Tolkien's personal mythology. On the gravestone he shares with his wife Edith, he added "Luthien" to her name and "Beren" to his. So I don't view weaving that theme into a dramatic treatment of the HOBBIT story as disrespectful to the author at all.

Here's the interesting thing, though. Let's agree that bottom-feeding is the natural ground state of a corporation; what that means is that nearly everybody is doing it, which means bottom-feeders on average don't make any more profit than you'd get investing in a mutual fund, but involve a lot more risk.

If you want to make *more* profit, you have to be about something *in addition to* profit. A great company has an identity which has value. This, by the way, is how Carly Fiorina ruined HP. She re-imagined HP as something more generic.

Air gapping the sensitive information is one of those things that looks easy on paper but runs afoul of the fact that people don't like to work that way. It's inefficient. It's not like people have *two* jobs, one sensitive the other not. They have one job in which sensitive bits are intertwined with regular bits, so in practiced people tend to cheat and do *some* sensitive work on the non-sensitive network.

Even if the users are unrealistically conscientious about never doing anything sensitive on their non-secure PCs, this intertwining of sensitive with ordinary information means that you can probably deduce a lot from apparently innocuous data. Imagine you discover the following information from poking around on an engineer's "non-sensitive" computer:

(1) He is exchanging email with certain university researchers setting up face to face meetings.
(2) He has downloaded datasheets for several families of exotic electronic components.
(3) He has telephone appointments in his calendar with salesmen from Unobtanium Corp.
(4) His browsing history shows he's been reading up on certain mathematical topics on Wolfram Alpha.

Now put this all together and another expert in his field might be able to deduce a lot more than you'd expect than if you looked at any one of these factors. The interconnected nature of an engineer's work means that if you remove all the sensitive bits it leaves a hole of a characteristic shape.

Until someone install something else on the network segment. Like a wireless access point. Or until malware takes over one of the trusted hosts.

Security vulnerabilities always involve violations of some assumptions you make, e.g. that anything coming from a certain set of hosts is benign, or that if a process on a server opens up an IP port it's *supposed* to do that. You want the security of a system to depend on as few assumptions as possible. If it does no harm in day to day operations and offers protection when your assumptions fail, why *not* run a software firewall?