I thought the deal was that the fraud was paid for by the least secure element in the chain. That is, if you're a merchant and you're still using the old swipe system after a certain date, you'll be on the hook for any fraud. However, if you've switched to the new systems that are theoretically more secure (and they are, provided that the data isn't pre-stolen) the bank is once again taking on the responsibility for the fraud. (Of course, merchant prices go up to cover the bank's costs when the bank screws up too, so...)

One for for the Advantage. I've been using this model of keyboard for around 15 years now. When I wear one keyboard out, I buy a new one. I have them at home and at work.

If you work in emacs (or really anything that requires a lot of modifier keys), it's unbeatable.

But honestly, beyond the curve, THE killer feature is that backspace and return are thumb activated. Your thumb simply doesn't get tired the same way your pinky does. Considering how much I have to hit those two keys during the day, there's really no way that any normal keyboard will ever be better than the Kinesis.

Sure, the Rolex is obsolete right now. :)

The Rolex works, but as a timekeeping device. A pretty poor one, really, since even the best mechanical movements are no match for a digital watch and will drift over time. Besides, why do I need to strap a thing to my wrist to tell the time? I can see three clocks on my desktop as I type this (the Windows start bar one, and two emacs buffers--I have a clock in the modeline).

The Apple watch is marginally more useful, and arguably easier on the eyes in certain configurations. (There are luxury watches that are the most obnoxious affront to good taste. Most of them, really.)

Anyway, I don't want one, but it's really no worse than spending money on a Rolex.

Not only that, they're probably not naive when it comes to this sort of thing. They know that if they don't want to be liable, they have to operate at arm's length from the data. Not only will they be able to tell their customers that they don't snoop, they're never on the hook legally for what their customers are doing because they're not involved.

There are about 300 spills a year from pipelines. You can do the googling yourself, but that's how many spills were considered 'significant' in the USA in 2014.

I tried to find a source that isn't obviously left-wing so you wouldn't reject it out of hand: http://www.forbes.com/sites/ja...

Admittedly, every single method of transporting oil is bad, and all of them lead to environmental problems and costs. So while pipelines may be 'best'--for some meagre definition of 'best'--there is an absolute amount of damage that's done. Conventional bombs are 'safer' if you drop them on my house than nuclear bombs, but it's a distinction without a difference when it comes to whether or not my house is still there.

Maybe the maintenance is as good as we could hope (I doubt it; we're talking about fallible humans and corporations with a bottom line), but this isn't exactly a solved engineering problem.

Nobody really knows how bad a bitumen spill would be. There's a lot of extra chemicals in there to keep it diluted and flowing.

In the end, one catastrophe a year is too many, even if it seems like 20 years later maybe things would be okay. I'm dubious about that number; 25 years after the Exxon Valdez spill, Prince Edward Sound is *mostly* recovered but still has problems. 20-30 years is a big chunk of time to humans, and we still have to live in this world.

No, not really. Almost everyone outside of Alberta is either ambivalent or pretty much against the pipeline. The tar sands are a disaster enough without giving more incentives to them to mine more bitumen.

I'm delighted to hear that he vetoed the pipeline. Of all the ways to produce oil, the tar sands are among the worst. They use an enormous amount of (fresh) water and energy. There are significant pipeline breaches every few weeks. Some you hear about, some you don't. A pipeline that long, well, I don't give it much of a chance of running for its intended lifetime without a leak. As a former Albertan, I find the idea of a bitumen slurry being dumped on the prairies really sad. I can't imagine it would be any better once it crosses the border.

I think this is great news.

No, actually, I'm not. I'm assuming that people would put that $10 towards a specific album and thus, a specific artist. If I stream dozens or hundreds of artists a month, the $10 is really diluted, so even if I spent most of my time listening to one artist, they necessarily won't have made as much money off of me as they would've if I'd bought their album.

Well, I mean, to a certain extent I agree. But you and I are having this conversation, which means that we've thought about it. Spotify and services like it mask the conversation from the public. They think that the artists are getting paid an amount commensurate with their skill, and roughly corresponding to the amount that the public enjoys the artist's music. If you ask someone what they think a band gets paid from a streaming service, I'm sure they'll tell you a figure that's much higher than it actually is. To an extent, it even removes the ability of people to vote with their dollars. When $10/month is being split between dozens of artists, even the person that comes out on top is poor. In the end, it may end up backfiring on the streaming services. If bands decide they can make more money doing releases on BandCamp and Spotify is undermining their album sales, maybe they'll pull out of the streaming services to sell fewer albums, but still make more money doing it.

But on a more philosophical level, I think it's sad that art is reduced to a mere calculation and balance sheet. I paid $10 for some albums and I've gotten joy out of them that can't be defined in dollar terms. In a very fundamental way, a lot of music shapes the way we interact with the world. I mean that very literally; music changes and shapes your brain as you listen to it. Emotion and music are tightly interwoven. It's a surprise when you meet someone that takes no enjoyment in music at all.

So while there's a fundamental truth to what you say, driving potential artists away from what could be a workable career doesn't really benefit us. If streaming revenues were enough to make up a decent, middle-ish class living, I don't think I'd argue with you. But these streaming revenues aren't enough to do anything other than have making music as a hobby, and really undermine important album sales. I'm sure there will be lots of artists that continue to make music and live in poverty--that's been true through the ages--but the promise of the age we live in was that MORE artists would be able to reach more people and more EASILY make a living.

While true, it's irrelevant for two reasons:

1) The radio doesn't play the songs you want on demand; and
2) The radio is effectively a way to drive album sales. Spotify is a REPLACEMENT for the album.

It used to be that it was worth it to play your songs on the radio (even at a loss) because people that liked your one song might want to hear the 10 hours you wrote--that would never be heard on the radio--and spend $10 on the album.

Now you pay Spotify $10/month for unlimited access to the entire album. To the entirety of the artist's catalogue. To the entirety of all the included artists' catalogues.

This is obviously and trivially less money than any one of those artists would make previously from you if you liked their music. Perhaps the argument could be made that more people are listening and giving a tiny amount of money to each artist, but I rather think that given the stats I've seen, this isn't even close to true.

This is much different from the time when people were pirating albums, since many fans would go out and buy an album that they downloaded because they wanted to support the artist. Now people feel that because they're paying $10 to Spotify or Rdio that they ARE supporting the artist. They're not going to pay for a subscription AND an album. That's exactly the opposite of the point of these services.

They need a new model. Streaming on its own for $10/month is clearly not enough money to go around. Spotify has infrastructure costs and has been bleeding money (I think they had a break-even or profitable quarter just recently?). Meanwhile, they also need to distribute the remainder of the already paltry $10 between a zillion artists. It makes no sense.

And slashdotters are always eager to comment without reading the source material. There's a big difference between saying that Google is doomed (that's not what this article says) and Google may not be able to stay at the top (that's much closer to what this article says).

Thompson's original article predicts no doom or gloom for Google.

This ad, ultimately, does all those things. I mean, it's cat food. It's there to feed your cat. In each video, there's a moment where the cats eat the cat food very happily, but the ad doesn't spend a lot of time screaming 'BUY FRISKIES' at you. It just shows some happy cats eating food. The benefits are that your cat is happy. It will cost about as much as cat food costs, one presumes, since they're not advertising it as up-scale gold cat food.

I'm not going to buy Friskies cat food. I buy food from the vet, and I know perfectly well that Friskies is garbage. But advertising allows content to get made--it's the basis of the last 50 years of TV production. So if they're going to make entertainment that's entertaining first and a small ad spot second, I'm okay with it.

But regardless of your personal views on the matter, that's a big pot of money sitting there that's ready for taking. In the context of the original story, it means that Google is passing up a lot of money that SOMEONE will grab.

Honestly, I don't think they are, when they're good.

For instance, the 'Dear Kitten' videos are brilliant. I love kittens. I love the way zFrank makes videos. So he made some videos for BuzzFeed and Friskies. I watch them gladly and of my own volition. Friskies only just barely puts their mark on the video, and I'm compelled to watch it and share it because the content itself is so worth watching. That's the kind of ad Thompson is talking about.

BuzzFeed is the king of that kind of advertising. A lot of their content is annoying, but it gets shared, and it makes them money.

Thompson's argument is mainly that that's the future of advertising, and Google's ability to capture those advertising dollars is incredibly limited. They can't make a Friskies video like that--that runs counter to all the ways they do business. It doesn't mean Google won't continue to be huge and rich, just that over time, other companies will surpass them.

There isn't any gloom and doom in Thompson's article. Not really. He's not arguing that Google is on the ropes, merely that it won't be able to hold its position at the top. There's a big gulf in between.

And he's probably right. Not because Google is doing badly, but because the size of the advertising market is *enormous*, and Google's current strategy limits its ability to capture more than a certain amount of that market. Over time, Google may find itself swamped by other companies (Facebook, for instance). This isn't because Google is bad, it's because other ways of making money in advertising have broader reach and make more money. I haven't seen any evidence that Google is willing to cannibalize its own business in the short run so that it can make a longer-term bet--that's an Apple strategy, and not many companies do it. Apple didn't care that iPhone sales would tank iPod sales, even when the iPod was a huge money-maker. They knew that they were trading one for the other, and it would be a good long-run bet.

This isn't to say that Google won't do that, or that Google is necessarily forced into what Thompson feels is its destiny, merely that that's the trajectory they're broadcasting right now.

Actually, if you go to Thompson's site and actually read the source article, he says very clearly that he doesn't think Google is going anywhere. The thesis isn't that Google is DOOMED, it's that Google's time AT THE TOP is limited. Those are two vastly different things.

Ben Thompson is a really thoughtful writer, and this game of telephone that we're seeing by Manjoo writing about Thompson's article has clearly caused confusion.

This isn't a doomer article, honestly.

Keeping in mind that I'm NOT a programmer that has spent a lot of time thinking about this...

But pre-distributing tasks just sounds like a queue becoming many little queues without any particular advantage. Moreover, if you're not sure when a processor is going to be done its work, pre-assigning work may lead to bigger bottlenecks and starvation. If you try to pre-analyse the tasks so you know how much work is needed, you're effectively wasting cycles on a problem that doesn't need to be solved--it's almost pure overhead, since the analysis could be part of the solution.

I'd have to read the paper to see exactly how they're getting such a huge speedup here. The graph is nice (and I believe it) but the explanation in the article isn't very satisfying.