Author: matt buchanan

Ahead of a March 2 event and TiVo HDs suddenly going extinct, two listings for a TiVo Premiere HD and Premiere XL HD have popped up in Best Buy‘s system.

The Premiere XL is THX certified and records up to 150 hours of HD content. It’s $500. The regular Premiere records a third as much, 45 hours of HD content, and it’s $300. They’re expected to land at Best Buy on March 27. Of course, prices and dates and recording capacities we expect will be by far the least interesting aspect of these things. We hope. [Engadget]

The Bloom Box is the latest energy miracle that sounds too good to be true: Debuting with a wide-eyed segment on 60 Minutes, it promises to be clean, cheap and backyard-friendly, the solution to our energy problems. What is it?

The heart of the box is a fuel cell. Though Bloom Energy's CEO K.R. Sridhar—a former NASA scientist—says it's a new kind of fuel cell. And though it's cleaner than any combustion engine out there, it still relies on fossil fuels and biofuels—not just hydrogen, like some other kinds of fuel cells do. Nevertheless, the folks at Bloom are doing something that could help make reduced emissions a reality for big businesses first, and then later, for homes.

Here's everything you need to know about fuel cells and the Bloom Box >

Join the conversation about this story »

See Also:

• Fuel Cells Will Have To Slum It With Forklifts Now That The World Loves Electric Cars
• Helen Mirren And Husband Test Drive The Chevrolet Equinox Fuel Cell
• Pentagon: Global Warming Will Lead To Global Conflict And Danger

The Cloak camera bag, presented without comment. (Mostly.) [Cloakbags via Craziest Gadgets]

Let’s not turn everything with a screen into an ebook reader, pleeeease? The DSi XL is launching in North America on March 28, and Nintendo’s putting that gorgeous 256×192 display to use with 100 Classic Books June 14. Dear Christ.

The 100-book package features works from Shakespeare and Mark Twain and it’s 20 bucks. Mercifully, you can adjust the text size, meaning if you want it to be readable, you’ll be able to scale it up to something like four words per screen. I’m sorry, but reading on a backlit screen with a resolution of 256×192 spread across 4.2 inches sounds like the quickest way to a migraine ever. Actually, that could be the best excuse for getting out of school ever. “My eyes are bleeding from reading so much!” [Kotaku, Bloomberg]

The Bloom Box is the latest energy miracle that sounds too good to be true: Debuting with a wide-eyed segment on 60 Minutes, it promises to be clean, cheap and backyard-friendly, the solution to our energy problems. What is it?

The heart of the box is a fuel cell. Though Bloom Energy‘s CEO K.R. Sridhar—a former NASA scientist—says it’s a new kind of fuel cell. And though it’s cleaner than any combustion engine out there, it still relies on fossil fuels and biofuels—not just hydrogen, like some other kinds of fuel cells do. Nevertheless, the folks at Bloom are doing something that could help make reduced emissions a reality for big businesses first, and then later, for homes.

To get a good grip on why we should care about this thing, let’s first look at the basics of fuel cell technology.

Fuel Cell Basics

Like a battery, a fuel cell is an electrochemical cell, basically meaning it derives electricity from chemical reactions. Sandwiched between two electrodes—an anode and a cathode—is an ion-conducting material called an electrolyte. Fuel flows in one side, over the anode. An oxidant flows into the other side, over the cathode. What happens, very basically, is that the fuel and the oxidant react, like strangers locking eyes across a room. The metaphorical sparks that fly from that encounter are actual electrons, which flow into the fuel cell’s circuit. Bingo, electricity. As with any molecular reaction, the recombination of atoms produces some waste as well—like water or carbon dioxide. So while it’s cleaner, there’s definitely a byproduct.

To be clear, a fuel cell’s not like a battery; it’s like a power plant. Once it converts fuel to energy, it sends that energy out the door. And as such, it requires some peripheral way to physically storing the fuel ingredients, and some way to capture produced electricity—such as a battery.

There are a several different kinds of fuel cells—unsurprisingly, since they were invented in the 1830s. Generally, they are categorized based on what their electrolyte is made out of, but sometimes they’re referred to by their fuel and oxidant, which varies too. You’re probably most familiar with “hydrogen fuel cells,” like for cars and small electronics. These are in fact proton exchange membrane fuel cells, which happen to use hydrogen as a fuel and oxygen as an oxidant. (The PEM fuel cell is what is specifically diagrammed above.)

Solid Oxide Fuel Cells

Bloom Energy’s Energy Servers are of the solid oxide variety of fuel cell. There’s two ways to do up an SOFC: A tubular design, which you can see above, or a planar design, which is what Bloom uses, as you can see below, since it allows them to be stacked into very neat boxes.

A solid oxide fuel cell is made out of all solid state materials—that is, every major component is made out of ceramic-like stuff. Bloom Energy claims their fuel cells are made out of “sand” baked into ceramic squares, and that’s just what an SOFC is. The exact material is a slightly secret sauce as are the black and green “inks” that coat the ceramic plates. Bloom’s got a pretty nice little Flash animation showing the basic process.

The major thing about an SOFC versus other fuel cells is that the material composition means they can run crazy hot—up to 1800ºF, says the US Department of Energy—and have to, since the ceramic materials don’t become active until they reach a certain temperature. Only at this temperature can they perform the chemical reactions with the fuel and oxidant we talked about above. The problem with the high operating temperatures is that traditionally it has lead to higher maintenance costs. You know, stuff breaks down. The goal for this technology is to have an “uptime” of 99.99%, as cited by cited by Scott Samuelsen, who’s the director of the National Fuel Cell Research Center at the University of California-Irvine. Bloom’s own trial at Google cites a 98% uptime.

The types of fuel cells you hear more about—the “methanol” ones that can already power laptops—do their business at a much lower temperature. Toshiba has one that typically runs at 120º to 200ºF. Though Bloom’s is obviously not a tech that could be a laptop’s power source, the Bloom Box’s higher operating temperature is a big advantage over “legacy” fuel cell technology. Bloom Energy VP of Marketing and Products Stu Aaron told me it gives them “fuel flexibility.” They can use biogases from land waste or fossil fuels like propane—so far in demos it’s been an even split between biogases and natural gas—whereas low-temp fuel cells require hydrogen in a much purer state that has to basically be refined or extracted via chemical processes.

While some other SOFCs use the hot exhaust generated by the reaction kind of like a cogen—a means of capturing heat emitted by a power generator, so that it too can be converted to electricity—Bloom’s Energy Servers simply recycle the heat within the cell, since the temperature generated by the reaction is almost exactly the heat needed for the reaction to happen. The rated efficiency spec for their current energy server is greater than 50%, compared to around 10% to 15% for solar (though University of Delaware-led researchers did recently hit a world’s record of 42.8% for solar).

Again, to be clear, the energy generated isn’t emission-free: These servers generate a small amount of CO2 when converting natural gas or bio-gas. It is less than what would get released if the same fuel was combusted, however. Customers can pick which of the two kinds of fuel they’d like to use; the trade off is between “optimizing for cost or carbon reduction,” depending on the company’s priorities, says Aaron.

Electricity In Bloom

Right now, the only box that Bloom is selling is a 100-kilowatt-hour energy server, which you can check out there. Inside are thousands of solid oxide fuel cells—each one able to power a light bulb. The cells are arranged in stacks, which are aggregated into modules, and so on, with a common fuel input. Right now, they’re just for corporations—like Google and Coca-Cola—and run about $700,000 to $800,000 each. The goal’s to get them down to three grand, where they’d be suitable for home use. That may still sound expensive, but they pay for themselves in 3-5 years, says Aaron, with an energy cost of 8-9 cents per kW hour vs. the 13-14 cents it typically costs in California. (It saved eBay $100,000 on their power bill.)

But cost is where the real skepticism comes in. Fuel cells aren’t a voodoo technology. They work. They produce energy. What analysts, and others, are wondering is whether Bloom’s really cracked the secret to making them cheap, at least some day. The critic that CBS trotted out on 60 Minutes, Green Tech Media’s Michael Kanellos, says that while there’s a 20 percent chance we’ll have a fuel cell box in our basements in 10 years, but “it’s going to say GE.” Which is fine with me, actually, because that means another season of 30 Rock jokes.

Still something you wanna know? Send questions about fuel cells, terrorist cells or Boom Blox here with “Giz Explains” in the subject line.

Days after The Great Purge of all apps considered sexy (unless you happened to be a major publisher), developers have spotted a new category for submitting apps to the App Store: Explicit.

The new category hasn’t shown up in the App Store yet, but what it potentially means is obvious: A place for explicit applications to live. We’ve been waiting for such a place since the App Store opened, actually getting excited when Parental Controls made their way into iPhone 3.0, hoping it would release a pent-up flood of apps like the long-lost South Park app, or Playboy for those so-inclined. An explicit category suggests that it could finally happen.

It’d be a smart move on Apple’s part. Phil Schiller can say we removed all of those overtly sexual apps because they were offensive, but we also value the ability of our adult customers to choose the kind of content they consume, so we’ve made a special walled-off section of the App Store for that more explicit content. I doubt that even with an explicit category we’d see anything more hardcore than R-rated, though. But still, it’d largely remove Apple from the sticky situation of being an arbiter of taste, an inconsistent censor.

Or you know, maybe it’s none of that. [Cult of Mac]

Flash 10.1 Beta 3 spreads the hardware-accelerated H.264 video love to more graphics chips, specifically some decidedly less beefy netbook chips: Intel’s GMA 500 series and Broadcom’s Crystal HD video decoder. Stick with 720p Flash video, and you’ll be dandy.

The Broadcom Crystal HD chip is a video decoder chip you’ll find in newer Pine Trail notebooks, like HP’s Mini 210, while the GMA 500 chip is in a ton of notebooks, from Dell’s Mini 10 to Sony’s Vaio netbooks and others. Congrats if you made it this far without immediately flipping to Hulu to check out how much peaceably your machine runs with Flash 10.1. [Adobe via Netbooked via Engadget]

Really? Really Sony BMG? After over 30 million views you just now decided to make YouTube pull Rick Astley’s “Never Gonna Give You Up” for a terms of use violation? If you were going to try to stop rickrolling, couldn’t you have at least done it a couple years ago?

Do you know what’s going to happen now? People are going to be rickrolling other people as much as possible using illicit streams as a protest all day. Thanks, assholes. Thanks.

Update: Via Ken Fisher, here’s why Sony BMG pulled the original video.

Update 2: It’s back, it was all a big mistake. [Neowin]

During the Frankly Speaking podcast, the hosts, who are Microsoft Australia Developer Evangelists spilled some details on how limited the hardware for Windows Phone 7…phones are going to be. There are precisely three chassis variants.

Chassis 1 is the phone we all saw on launch day—a huge touchscreen with a minimum 1GHz processor (read: Snapdragon) and a dedicated GPU, along with all of the other requirements we’ve talked about. They’ll be the first out of the gate. Chassis 2 is the keyboard variant, with sliding keyboards and touchscreens, which are apparently more Palm Treo-like. What Chassis 3 is like is still a mystery, though.

While the three supported chassis are obviously pretty locked down, it’ll be interesting to see how much of their own spin manufacturers will be able to rub on Windows Phone 7, given precisely how much they can’t touch. [Frankly Speaking via ZDNet]

Even as US authorities are getting pretty damn sure who’s behind the high-level hacking attempts from the other month, and that they were launched from the Shanghai Jiaotong University and Lanxiang Vocational School, China’s all “Nuh uh, eff you guys.”

I mean, that really is the essence of their rebuttal. Chinese Foreign Ministry spokesman Qin Gang said that “Reports that these attacks came from Chinese schools are totally groundless and the accusation of Chinese government involvement is also irresponsible and driven by ulterior motives.” And that’s that, I’m sure. [The Hill]

The problem with elven cameras like Canon’s S90 is that they’re awkward to hold sometimes. Richard Franiec’s custom grip—CNC-machined from a block of aircraft-grade aluminum and black-anodized to match—adds just enough extra surface area to fix that.

It stays just about as pocketable since the grip sits 1.5mm below the lens, and is attached to the camera with very high bond adhesive tape (meaning you can take it off without screwing up the camera’s finish). It’s 33 bucks, which isn’t a bad price to pay to make one of the best point-and-shoots on the planet even more usable. [Lensmate via Gadget Lab, Richard Franiec’s Picasa]

Wired’s Steven Levy takes us inside the “algorithm that rules the web“—Google’s search algorithm, of course—and if you use Google, it’s kind of a must-read. PageRank? That’s so 1997.

It’s known that Google constantly updates the algorithm, with 550 improvements this year—to deliver smarter results and weed out the crap—but there are a few major updates in its history that have significantly altered Google’s search, distilled in a helpful chart in the Wired piece. For instance, in 2001, they completely rewrote the algorithm; in 2003, they added local connectivity analysis; in 2005, results got personal; and most recently, they’ve added in real-time search for Twitter and blog posts.

The sum of everything Google’s worked on—the quest to understand what you mean, not what you say—can be boiled down to this:

This is the hard-won realization from inside the Google search engine, culled from the data generated by billions of searches: a rock is a rock. It’s also a stone, and it could be a boulder. Spell it “rokc” and it’s still a rock. But put “little” in front of it and it’s the capital of Arkansas. Which is not an ark. Unless Noah is around. “The holy grail of search is to understand what the user wants,” Singhal says. “Then you are not matching words; you are actually trying to match meaning.”

Oh, and by the way, you’re a guinea pig every time you search for something, if you hadn’t guessed as much already. Google engineer Patrick Riley tells Levy, “On most Google queries, you’re actually in multiple control or experimental groups simultaneously.” It lets them constantly experiment on a smaller scale—even if they’re only conducting a particular experiment on .001 percent of queries, that’s a lot of data.

Be sure to check out the whole piece, it’s ridiculously fascinating, and borders on self-knowledge, given how much we all use Google (sorry, Bing). [Wired, Sweet graphic by Wired’s Mauricio Alejo]

Christ, this is getting complicated. Mirrorless cameras with interchangeable lenses that promise near-DSLR quality without DSLR bulk are verifiably exploding. After the now-established Micro Four Thirds format from Panasonic and Olympus, Samsung jumped in with its Hybrid NX camera. Yesterday, Sony revealed its own spin, the Alpha concept.

And now, there’s Sigma, known for its prickly but highly capable compacts powered by its oversized Foveon image sensors, is going to make their own system with Foveon. More interesting, maybe, is that they’re thinking about making lenses for the other guys too. Oh well, who really wants simplicity? [DP Review]

There’s something I love about CompactFlash cards—speed, and toughness—versus SD. The CompactFlash 5.0 spec takes them to a new level: 144 petabytes of possible storage, and exponentially faster transfer speeds (from 128KB per transfer unit to 32MB).

I have to say, I can’t wait to get one of these inside a 1080p video-shooting DSLR. And jeez, the CF Association’s website is so damn 1998. [CompactFlash Association via DP Review]

Windows Phone 7 is a new beginning for Microsoft, and at the same time, an ending. The epoch of the “slap our software on any old hardware” open platform is dead.

There’s a spectrum of hardware and software integration. At one end, you have the likes of Apple, RIM and Nintendo who create software and design the hardware that it runs on. It’s controlled and tightly integrated top-to-bottom. At the other end, you have the classic Microsoft model—they just create the software, and a hardware company like Dell or HTC or Joe’s Mom buys a license to install it on their machine, which they sell to you. (FWIW, Microsoft would argue they’re in the middle, with open source, that is, “unstructured openness,” down on the other, wild ‘n’ crazy end.) In the center, you have a mix—there’s still a split between software and hardware, but one side dictates more stringently what’s required of the other side, or they work more closely together, so it’s sorta integrated, but sorta not.

The Philosophy of Sorta Open vs. Sorta Closed

The integrated philosophy is summed up pretty nicely by the legendary Alan Kay, “People who are really serious about software should make their own hardware.” It’s about a better experience. Granted, today that mostly means “design their own hardware,” since very few companies actually make the hardware they sell. Take a MacBook or iPhone—sure, Apple made it pretty, but it’s actually manufactured by a company like Foxconn to Apple’s specifications. HTC and Asus, on the other hand, do design and build their own hardware, not just for themselves, but for other companies. (For instance, HTC built Sony Ericsson’s Xperia X1 and Palm’s Treo Pro.)

The other side is typically couched as a kind of openness offering choice which drives competition, and therefore, pushes innovation, as Steve Ballmer puts it: “Openness is critical because it is the foundation of choice for all of our customers…choice, which will drive competition, which is ultimately the engine of innovation and progress.” The other argument is that it creates a bigger platform for more innovation to happen on, since more stuff’s running the same software. It’s the benefit of there being hundreds of different PCs that run Windows, versus a handful that run OS X: Sheer numbers.

As for the nasty things they say about each other, the top-to-bottom guys say that the hardware-software split leads to a crappier product, because one single company’s not in charge of the experience, making sure every little bit works. Like how multitouch trackpads universally blow goats on Windows laptops. Who’s fault is that? Microsoft’s? The guys who built the laptop? Advocates of choice say that top-to-bottom integration kills innovation and hardware diversity, all the while making systems way more expensive. If you want a laptop that runs OS X, I hope you like chiclet keyboards and paying out your gnads.

Those are the basics. Microsoft, throughout its history, has mostly made software for other people to stick on their hardware. Apple has, one dark period aside, basically always designed the hardware for its software, and sold them together. Yin and yang.

The Coming Change

The Entertainment & Devices division of Microsoft, with its “Chief Experience Officer” J. Allard, is different from the rest of the company. It made the Xbox. The Xbox had—waitaminute—Microsoft software running on Microsoft hardware, which you bought together as a package. Why? Because a gaming console wouldn’t work very well as an open system, sold like a desktop computer. People buying a gaming console expect a single, integrated experience that just works. This is a historical truth: Since the NES, Nintendo, Sony and anyone else entering the business who you’ve actually heard of will only build closed boxes.

E&D also made the Zune. Why? Well, Because Microsoft’s open hardware approach bombed in the portable media player business. Miserably. The PlaysForSure ecosystem was totally schizo—effectively a multi-layered DRM released by a group whose responsibility was media formats and players for the PC. Microsoft handed out DRM, codecs and syncing software, and a partner would (pay to) make the media player, typically with third-party firmware in the middle. The players never “played for sure.” They worked, but only if you were lucky and managed to sacrifice the proper number of goats under the correct cycle of the moon on the first Saturday after the second Thursday of the month. At the same time, the iPod’s top-to-bottom, seamless ecosystem proved itself: It owns 70 percent of the MP3 player market. Microsoft realized the only way to compete was to make the software and the hardware—alienating all of their so-called “hardware partners” in the process. So, Zune. Which single-handedly slew the undead remnants of PlaysForSure and its ilk, when it wasn’t compatible with Microsoft’s own ecosystem.

But these were exceptions. They’re consumer products. Entertainment experiences. Niche products, in Ballmerspeak. Not computers. Windows Mobile started life as Pocket PC because it’s a computer you shove in your pocket. So Microsoft played it like it played the computers on your desk, an approach that worked pretty gosh darn well for ’em there.

The Long Death Spiral of Windows Mobile

You can’t really exaggerate how PC-minded Microsoft’s approach to mobile was. The ecosystem was wild and messy, getting a little more organized with the Pocket PC 2000 OS. Pocket PCs actually did adhere to a generic set of hardware specifications put out by Microsoft (not terribly unlike their Project Origami for UMPCs some years later), but there were tons of devices from tons of manufacturers, along with multiple editions of the Pocket PC software—like the Phone Edition, which tacked phone powers onto PocketPC’s PDA core. With Windows Mobile 6, Microsoft stopped calling the devices Pocket PCs. And you know where things went from there.

Smartphones—of which about 180 million were sold last year according to Gartner—are what Steve Ballmer calls a “non-niche device,” which to him, are things like TVs, PCs and phones. So the Windows model still applies, right? That’s the approach Microsoft took for years. So, just about anybody who could pay for the license could shove Windows Mobile onto their phone. Some people did great things with it, like HTC’s HD2. Other people did less awesome—okay, shitty—things.

What’s amusing is that, despite the Windows Mobile model clearly not working that well, Google came in with Android and applied basically the same strategy, except Android’s actually free to vendors—and if they agree to certain conditions, they can include Google’s applications and be branded as “Google” phones. Not surprisingly, the same strategy’s leading to the same outcome—some people do awesome things, like the Hero. Some people commit atrocities. Some software works on some Android phones and not on others. Fragmentation amok.

The philosophy at play is the same: Open platform, device choice.

Windows Phone 7 Series ends all of that for Microsoft. (Not so coincidentally, it comes out of E&D, the same division that created Xbox and Zune.) Other people still make the hardware, but Microsoft’s got an iron grip on the phone, and how software and hardware come together, more so than ever before.

When An Open Door Closes, Someone Pries Open a Window

Ballmer phrases it as “taking responsibility for the experience.” What does that entail? A Windows 7 Phone Series…phone must have a high-res capacitive 4-point multitouch display, 5-megapixel camera, FM radio, accelerometer, Wi-Fi, GPS, set CPU and GPU benchmarks, and even a particular button set that includes a dedicated search button. Very little is left to the hardware guys. The shape of the phone, and whether or not it has a keyboard, basically. And Microsoft’s only partnering with a select group of OEMs—Joe’s Mom can’t build Windows Phone 7 Series phones. (Yes, I’m going to keep writing the OS’s entire name out because it’s a dumb name)

This level of involvement is a radical break for Microsoft. It’s them admitting that the old way wasn’t good enough. That it was simply broken. That their partners effectively can’t be trusted. They have to be told exactly what to do by Microsoft, like goddamn children. It’s Microsoft finally saying, “While we can’t make our own hardware”—since phones are a mega-category, that could limit growth and once again piss off partners—”we’re serious about the software.” Coming from Microsoft? That’s huge.

It’s a necessary step, because Microsoft’s position in mobile is way different from its position in desktops, way different from the position it expected to be in. They’re not the dominant OS. They don’t lord over a vast ecosystem, commanding 90 percent of the smartphones on the planet. They’re just another competitor. Meaning they have to be different, and compelling, in a much different way than if their expectations had played out. If Microsoft was in the same position in mobile as they are on the desktop, do you think they’d be shitcanning the entirety of their mobile platform? Nope. They’d be expanding the ecosystem, working to make it more ubiquitous, more entrenched. Not a breath of fresh, rainbow-colored air.

Still, Microsoft isn’t exactly alone. Google may be shedding Android licenses like cat hair, but they’re covering their asses by following this same tack, too. I’m talking, of course, about the Nexus One. Heralded as the Google Phone. It’s the Anointed One, the truest of all Android phones. And you know why? Because Google told HTC how to build it. Google designed the phone themselves to be the exemplar of Android. It’s basically saying no other phone was good enough. Not even the Droid, released just two months before it. Google had to make it them goddamn selves. That was the only way to achieve Android perfection.

An interesting side effect is that it puts the company who made that phone, HTC, in a fairly awkward position. HTC and Asus, as I mentioned earlier, are unique: For years, they slaved in near-anonymity, making phones and PCs for the brands you’re familiar with. HTC, at one point, made 80 percent of the Windows Mobile phones out there, which were sold under monikers like the T-Mobile Dash. Now, they’re busting out with huge campaigns to be on the same brand level as the Dells and Palms of the world. They even design their own software, which is increasingly how these companies distinguish themselves, since everybody’s using basically the same guts in everything, from laptops to phones. While they obviously still make money, these OEM superstars are effectively re-marginalized, hidden by the bigger Windows brand.

Worse off, still, it would seem would be the brands who don’t make the hardware, the Dells of the world. They’re a middleman in the worst sense—their brand is squeezed, and they’re passing on guts made by another company entirely. It’s almost like, “Why do you even exist?”

Assuming Microsoft does get a toehold with Windows Phone 7, the ecosystem might loosen up. It might have to, in order to expand outward. Meanwhile the march of random third-party Android phones will keep on stomping through, but make no mistake: Microsoft and Google, former champions of the open platform, have basically admitted that the only right way to build a phone is to do what their chief rivals Apple and RIM already do: Design the software and hardware yourself. Now, they’re serious.

Vudu, despite being a swanky video service with delicious 1080p streaming, never made any money. So why did Walmart just drop $100 million on a failing service? Peter Kafka reports it’s because Vudu convinced Walmart its video compression was nigh-wizardry.

The standard assumption was that Walmart wanted Vudu’s baked-in deals with Hollywood studios, so it’s interesting they were after the tech. There is some truth to Vudu’s claims, for sure: They were the first to roll out 1080p streaming at very nice bitrates compared to most services, and the video looked amazing for it. (But the compression wasn’t that damned magical, since action scenes with its HDX 1080p streams would still spike up to 20Mbps, and averaged around 9Mbps. Zune’s 1080p service requires an internet connection of at least 10-12Mbps, so it’s probably in the same ballpark.) Either way, their tech was good enough to interest Cisco as well, Kafka hears, which probably helped drive up the price.

Amazon, here comes Walmart. [MediaMemo]