Google’s Project Loon fills internet gaps with solar-powered balloons

Mobile data might seem near-ubiquitous, but the world still has major dead zones and huge expanses with poor coverage. Anyone that has fervently and consistently checked the availability of Verizon’s FiOS in their neighborhood knows this all too well, but that still places you on the well-connected side of the spectrum — there are parts of the world where Time Warner and Comcast (let alone FiOS or Google Fiber) would be a huge advance. Over the weekend, Google launched Project Loon, an initiative to help fill in those internet gaps through the use of networked balloons.

The goal is to provide broadband-like internet for the two-thirds of the world that doesn’t have access to a reliable internet connection. To do this, Google is using a network of actual high-altitude balloons — this isn’t some kind of metaphor — that fly in the sky above the Earth. The launch of the project consisted of launching 30 balloons, each of which are capable of providing internet access with speeds comparable to 3G — better than nothing, but probably not fast enough to download all those episodes of Falling Skies that you missed so far this season.

The balloons were launched over New Zealand as a test bed and fly around 12 miles high, each of which able to provide internet connectivity for an estimated hundreds of people within a 25 mile diameter. The height of the balloons was intentionally chosen to be out of the way of commercial flights, as well as birds.

The Loon balloons are networked to each other with a radio transceiver, and are also in contact with the ground thanks to another. Each balloon has a third transceiver for backup purposes, and is tracked with on-board GPS. Each Loon balloon is powered via solar panels, which takes around four hours for a full charge, and extra energy is stored in a rechargeable battery.

The balloons also contain equipment to monitor the weather, as objects flying in the sky will surely bump into some less-than-desirable weather at some point, but the balloons have been built to withstand the conditions they will be facing.

This test covers just 50 or so people in the Christchurch area of New Zealand, where the testers are able to connect to the network thanks to special Loon antennas. The antennas work specifically with the Loon network and aren’t compatible with WiFi networks and actually filter out standard WiFi signals. The signals being transmitted to and from the balloons are encrypted, though Google didn’t go into much detail about how.

As for positioning, the balloons are largely automated, though do allow for human control from technicians back at Google’s Loon home base just in case. Eventually, though, Google is looking to automate the entire process and keep the balloons traveling within specific wind patterns in order to maintain consistent coverage of desired locations.

Where once satellite internet was the only option, balloons could provide a cheaper, faster solution to large portions of the globe where it might never make sense to run fiber. Basically, Google is testing out an internet network built on the backs of balloons flying in the sky, ushering in an age of steampunk-like connectivity for people far off from city centers and otherwise lacking in high-speed internet access.

Source

A rare look inside an LTE cell site, operated by Sprint in San Francisco

There is one incredibly important field of modern technology that has always proven to be a bit of an impenetrable enigma for ExtremeTech: The hardware used by mobile carriers. We know in general terms how a cell site works, but for the most part the actual hardware and software and network topology is a proprietary black box. That’s why it’s very exciting when a mobile carrier such as Sprint invites a tech blog to take a close look at one of its new LTE cell sites in San Francisco.

In this case, Engadget visited Sprint cell site SF33XC664, located on a rooftop high above Van Ness Avenue in central San Francisco. In the photo above, the big structure at the end is the building’s elevator shaft, the tall gray boxes are Sprint’s LTE equipment, the smaller white boxes in the middle are AT&T’s equipment (it is common for cell sites to be shared), and the boxes in the foreground are Sprint’s older CDMA gear. Eventually, once Sprint’s Network Vision rollout is complete, the CDMA boxes will be be carted off and scrapped/recycled.

In the photo above, the box on the left (open door) houses Sprints’ LTE hardware, the cabinet in the middle is a battery backup system (good for 5-8 hours), and AT&T’s backhaul fiber connection is on the right. Historically most cell sites used microwave links for backhaul links to the carrier’s core network, but with the increased bandwidth requirements of 3G and 4G networks fiber links are now mostly used (especially in urban environments).

Inside the open cabinet is a ton of rack mounted network gear: Most notably, a Cisco switch at the bottom (connected to the AT&T fiber backhaul link), and three Samsung Smart Multi-modal Base Stations (SMM-2LD000). The top box deals with LTE data, while the bottom two boxes are CDMA data and voice. Because these Samsung base stations are multi-mode, Sprint can decommission its dedicated CDMA hardware and run a much simpler, cheaper, easier-to-manage setup.

These Samsung multi-modal base stations are the brains of the operation where all the magic occurs — and yet if you type the model number into Google, you get zero results (except for this story). This is generally because every base station has been tailor-made by the equipment maker (Samsung, Siemens, Ericsson) for the mobile carrier. In this case, Sprint started with a Samsung Smart MBS and probably had some custom firmware installed for its specific network topology. There is a Smart MBS website, but it’s almost completely devoid of useful information.

What we do know is that Samsung’s Smart MBS contains a software-defined radio (SDR), which basically means that Sprint’s network administrators can alter the technologies (LTE, CDMA, GSM) and frequencies (900MHz, 1900MHz, 2500MHz, etc.) used by the device through a simple software-based control panel, rather than actually installing new hardware. The SDR is the sole reason that Sprint is able to consolidate and simplify its network setup. In short, it is the MBS that manages the connections to every smartphone within its catchment area. It is essentially a huge wireless router.

The base station hardware would be nothing without amplifiers and antennae, however, which is what you see in the photo above. There are three directional antennae attached to the building, each covering a 120-degree arc. The two boxes next to the antennae are RF amplifiers, which amplify the signal from the base station before transmission. In the close-up photo below, you can see that the amplifiers are encased by a big heatsink — because amplifying signals all day generates a lot of heat. The output power of every cell site is different, depending on the local safety regulations for broadcast RF, but they generally max out at around 100 watts.

Ironically enough, Engadget actually gets surprisingly slow upload and download speeds when it’s up on the roof, next to the antenna. This is probably because the wavelength (i.e. the distance between the peaks) of cellular signals is between 15 and 50cm — and so if you’re too close to an antenna, before the wave can really stretch its legs, it causes all sorts of grief.

So, there you have it: When you make a call or access the internet, you now know exactly what the hardware that handles your data looks like. Your phone broadcasts its data, which is picked up by a cell site’s antenna, and then your signal is processed by a base station (such as the Samsung Smart MBS). The base station, which is just like a large wireless router, then forwards your data to the person you’re calling or the website you’re accessing, via the mobile carrier’s backhaul network.  Then, assuming you haven’t moved to another cell, data is transmitted back over the backhaul network to the same base station and broadcast by the antenna. If you have changed location, your phone will tell the new base station that you’ve moved — and data will be sent there instead.

Finally, your phone, which is constantly listening on a certain frequency for packets of data that are tagged with your unique ID, picks up these packets and decrypts them. Voila: A cellular network.

[Image source: Engadget]

Source

Xbox One: The wobbly third leg of Microsoft’s non-desktop trifecta

With the complete hardware, services, and pricing unveiled for theXbox One at E3, we now have the totality of Microsoft’s “next-generation” consumer-oriented lineup: Windows 8 on the desktop, laptop, and tablet, Windows Phone 8 on the smartphone, and Xbox One in the living room. On paper, this trifecta, seamlessly connected via Microsoft Account, SkyDrive, and Xbox Live, is almost perfect. In reality, though, this couldn’t be further from the truth. Where did it all go wrong for Microsoft?

From an objective standpoint, all of Microsoft’s new-for-2013 offerings — Windows 8, Windows Phone 8, and the Xbox One — are perfect. Windows 8 capitalizes on the slow death of the desktop and the rush towards mobile; Xbox One is a powerful and feature-rich games console that could dominate the living room; and Windows Phone 8 is a sharp and savvy smartphone OS that ties everything together, while on the move or as a second screen. As a tech writer and a self-confessed life-long Microsoft fan, I have never been more excited about Microsoft’s future than over the last two years of covering Windows 8, Windows Phone 8, and the “Xbox 720“.

From a subjective standpoint, though, each of Microsoft’s new offerings is intrinsically flawed and bogged down by crippling policy decisions no doubt handed down from Microsoft’s besuited higher echelons. Windows 8 and 8.1, despite “responding to customer feedback,” still forces users to use the Metro interface, even when a touchscreen isn’t present. Windows Phone 8 is one of Microsoft’s most polished products, but a smartphone OS is only as strong as its app ecosystem, and due to its minuscule market share WP8 still lacks the ecosystem to pull consumers away from iOS and Android — an unfortunate Catch-22 if I ever saw one. The Xbox One, depending on your point of view, is either an awesome all-in-one living room box that plays games, or an awful DRM-restricted games machine that acts as an HDMI passthrough for your cable box — the very same thing that the tried-and-failed Google TV attempted to do.

How did Microsoft manage to take three exciting, technologically advanced products and turn them into mediocre, humdrum devices that have had all of the fun and adventure sucked out of them?

Windows 3.0

A history lesson

For the past 20 years, Microsoft hasn’t done much more than double, triple, and quadruple down on the desktop ecosystem. Since the launch and success of Windows 3, almost all of Microsoft’s decisions have revolved around the maximizing of Windows-derived profits. The success of Office, one of Microsoft’s most profitable divisions, is entirely underpinned by Windows’ 95%+ desktop penetration — ditto the Server division. It’s even possible to draw a link from Windows’s dominance in the desktop market, to DirectX and PC gaming, to the Xbox.

To be honest, at the time, this all made perfect sense. Windows, Office, and Server were, and are, monumentally large profit drivers. But if technology has taught us anything it’s that nothing lasts forever — especially business models predicated on a large, bulky form factor that is virtually guaranteed to go the way of the dodo as technology advances. If anything, Microsoft has done incredibly well to maintain the desktop-dominated status quo for as long as it has.

Now, though, with tablets and smartphones exploding much faster than anyone could’ve anticipated, Microsoft is forced to adapt. Windows 8 and Windows Phone 8 are so different from their predecessors. The Xbox One isn’t that different, but it’s about as far from the Xbox 360 as Microsoft could get without completely redefining the games console paradigm. On paper, these massive changes all make sense, and if they were executed properly they really could give Microsoft the beachhead in the mobile market that it so desperately needs.

Gutless equivocation

Unfortunately, such changes simply can’t be easily made by a large multinational bureaucracy that generally works with three-year product cycles, rather than 12 months. From a strictly fiscal and pleasing-the-stock-holders point of view, too, Microsoft can’t just kill off the desktop. As it stands, Microsoft is massively profitable and will be for years to come. But at the same time, Microsoft knows that it must change now or face being squeezed out of the market by iOS, Android, and other upstarts. Faced with such a dilemma, Microsoft hedged its bets and created Windows 8, a Frankensteinian operating system that is the jack of all trades but the master of none.

Where does Microsoft go from here? It’s not too late for Windows 8, especially with Windows 8.1 coming up. If Windows 8 is a success, then there might be a knock-on effect that finally gets Windows Phone 8 off the ground. Finally, if the Xbox One works, it could be the perfect centerpiece of a new, non-desktop-oriented Windows ecosystem. That’s a lot of ifs, though, and given how wobbly the Xbox One looks in comparison to the PS4, and WP8′s consistently ailing market share, I think Microsoft has a rough few years ahead.

Source

iRobot Ava 500: The first ever self-driving telepresence robot

Today iRobot, which has been rather quiet this year, has a big telepresence announcement: The Ava 500, the “first self-driving business collaboration robot.” The Ava 500 was built in collaboration with the networking experts at Cisco, and was designed to bring enterprise businesses what could be the first viable telepresence robot on the market.

For the Ava 500 Cisco was enlisted to handle HD video and audio transfers with enterprise-level security and interoperability — basically, the telepresence part of the equation – while iRobot was able to play on both its strength and previous experience with the design of therobot itself. The Ava 500 was built around the company’s Ava bot, the goal of which was to produce an easy-to-implement, versatile, affordable robot platform for security and healthcare applications.

The Ava 500 is actually the second time the platform is being put to use. The first was the RP-Vita, what the company bills as a “telemedicine solution,” or basically a telepresence bot for hospitals that gives doctors remote access to patients and on-site care teams.

While a number of telepresence robots do already exist (with varying degrees of sophistication), the Ava 500 has a few very notable features. The partnership with Cisco is important, but the foremost of the selling points is that the robot uses autonomous navigation in order to map and direct itself through work areas. What Ava does is quite different from iRobot’s iAdapt (the Roomba’s combination of software and hardware sensors that it uses to navigate) — iAdapt is exploratory and does not rely on the robot constructing a map of its environment, while Ava does exactly that. Once the map is constructed the bot navigates on its own while actively avoiding obstacles.

Importantly, given the size of Ava and the office setting, the robot does not roll into objects in order to detect their presence the way a Roomba does. Ava will avoid collision with objects (like people) altogether. The robot is not driven from place to place; it’s simply told a final destination by a remote user. When that person has a meeting to attend, the Ava 500 will direct itself to the room where that meeting is taking place. The remote worker will simply need a compatible system (like Cisco’s EX90) in order to work with the communications software.

The Ava 500 is essentially a self-navigating Cisco TelePresence EX60 endpoint. In fact, the head of the robot uses Cisco EX60’s 21.5-inch station, which is the same system that can be used in a home or office. With this rolling on its own around a business, a worker on the other end can use an iPad to easily commission a bot and direct it before hopping on a Cisco workstation for videoconferencing.

“Hi Bob, how about an ad-hoc hallway conversation?”

Any trip can be done in private mode, where the display is left blank, or in public mode, where the user’s face is displayed on the monitor. When the latter option is chosen the worker can have what iRobot calls an “ad-hoc hallway conversation” — also known as a chat — with a coworker. When the user is done with the robot, he can sign off, and it will automatically return to its charging station.

iRobot is currently aiming the Ava 500 at enterprise businesses, who will be able to use it for collaboration, facility tours/inspections, remote management, training, presentation, and similar activities. The bot is expected to be available in early 2014, and priced at $2,000 – $2,500 a month, not including services.

Source

The NSA’s Prism leak could fundamentally change or break the entire internet

Most of the discussion of the NSA’s Prism program has focused on the domestic fallout, with some spillover into the question of what Google, Microsoft, Facebook and Apple did (or didn’t) do to safeguard user data when the NSA came knocking. There’s another facet to the issue that’s worth discussing, I think, even though it’s going to take months or even years to play out. The Prism leak could lead to fundamental changes in how the internet is controlled, administered, and routed.

Despite the focus on domestic implications, Prism is a system specifically designed to eavesdrop on foreign Internet traffic flowing through the United States — the NSA even says so.

 

This leak is a colossal embarrassment for the United States. Every time the UN-backed ITU has raised the issue of a more global approach to internet governance, the United States has fired back with both barrels and a tactical nuke. Last December, the House of Representatives passed S. Con Res 50 by a vote of 397-0. The opening paragraph of that resolution declared it vital that the internet “remain stable, secure, and free from government control” and stated that the structure of internet governance “has profound implications for competition and trade, democratization, free expression, and access to information.”

For a decade, the United States has fought to position itself as a neutral party that could be trusted to administer the internet in a manner that was beneficial to all parties. Now, the NSA has been caught gloating over the fact that internet traffic routing rules drive foreign data directly into its data centers by the truckload. Latin America, China, and to a much smaller extent, Europe, have precious little reason to trust the NSA and now, a great many reasons to guard their own digital borders. The question of whether the NSA actually did anything inappropriate is remarkably unimportant when there’s political hay to be made.

So what happens next?

Not every sea change kicks off with a trumpet fanfare and international brouhaha. If this had happened seven months ago, in the run up to the ITU’s latest vote on internet governance, it might have been a different story. As things stand, the short-term impact may be minor. Long-term, however, I think things will change.

This slide shows the amount of internet traffic that passes through the United States or Canada from other major regions. The percentage of worldwide Internet traffic routed through the US has been falling for years as nations have brought their own IXPs (Internet Exchange Points) online — African traffic, for example, routes almost entirely through Europe. These trends could accelerate sharply now that Prism is in the limelight, either out of a genuine fear of US spying or because politicians see a handy opportunity to launch their own regional efforts and projects. Either way, the total amount of traffic routed through US servers is likely to decrease at an accelerating rate.

The Prism disclosure could cause problems for the United States’ diplomatic efforts in other, seemingly unrelated areas. The ACTA (Anti-Counterfeiting Trade Agreement) treaty was narrowly defeated at the last minute, after the various member states of the EU raised concerns regarding the treaty’s negotiation process and copyright strong-arming. There’s a philosophical link between the idea that the NSA has the right to spy on foreign internet traffic and the treaty’s requirement that various EU member states agree to US-mandated copyright laws and the enforcement mechanisms those laws promote. Both the spy program and the treaty implicitly position the United States as the arbiter of legal rights in a manner that’s not subject to oversight from member states or its own citizenry.

I wouldn’t be surprised to see foreign nations simultaneously boosting their own regional networks and tightening their data protection laws. The long-term impact on companies like Google or Facebook is unclear. It’s one thing to require Google Germany to comply with German privacy laws, but how would Google go about isolating email sent from a German user to a US citizen?

Prism is designed to theoretically intercept such a missive — so what happens if Germany decides it doesn’t like that possibility? I honestly don’t know. It seems unlikely, however, that the status quo will continue. Nations that want to build their own spy systems and carry more traffic internally now have political cover to build the networks and blame Prism as the reason. Countries that were genuinely concerned about US overreach now have a concrete system to point to rather than a vague fear. Regardless of underlying motivation, there are precious few arguments for funneling traffic through the US.

Source

Making sense of the NSA Prism leak as the real details emerge

Few stories have broken as quickly or as confusingly as this week’s multiple revelations about US government spying. Initial, widely quoted, reports from the Guardian and from the Washington Post painted a dark picture of government “direct access” to the servers of nine major US internet companies: Microsoft, Yahoo, Google, Facebook, PalTalk, AOL, Skype, YouTube, and Apple. Separately, it was also reported and confirmed that Verizon has been routinely handing over all of its customers call logs to the government — with the Wall Street Journal reporting that the same is true of AT&T and Sprint. While the call logging story has been widely verified, claims about the Prism program are far murkier. As quickly as details of the program emerged they were contradicted by the NSA and the companies implicated, as well as by some other leakers. Digging into the details of what is actually known helps shed some more light on the story and how concerned we should be.

The direct access red herring

Nothing in the Washington Post article based on the leaked NSA PowerPoint presentation has generated more heat than the lightning rod phrase “direct access.” The Post story claimed that the leaked document revealed that the NSA is “tapping directly into the central servers of nine leading U.S. Internet companies.” The blogosphere keyed in on the words directlyand central to light up with horror scenarios of every chat and email being funneled directly to some massive NSA data warehouse. All this was at complete odds with the increasingly strong denials issued repeatedly by the companies involved.

Bloggers reacted with disbelief, many claiming that the internet CEOs must have been complicit, or perhaps were plausibly deniable. I for one think that’s highly unlikely. I don’t see the likes of Larry Page and Mark Zuckerberg either telling direct lies that could be found out later, or even being willing to look stupid by not knowing what is going on in their own server rooms. The similar claim that they might be denying knowledge of the program because they are bound to by law also doesn’t ring true. A simple “I can’t talk about it” would hardly have landed either of them in jail. So how do we reconcile these two differing realities?

Going back to a more direct quote from the NSA leak helps resolve this apparent paradox. In the same Post article we read “Collection directly from the servers of these U.S. Service Providers: Microsoft, Yahoo, Google, Facebook, PalTalk, AOL, Skype, YouTube, Apple.” This begins to make sense. Clearly all nine companies provide scads of information to the US government. Google’s own transparency report lists over 68,000 requests for 2012 alone — about half filed by the US and UK governments and almost always filled by Google. Given the volume of data, and its potential time sensitivity, it makes perfect sense that some type of automated, digital delivery system would be put in place to facilitate the transfer of data. Presumably legitimately approved requests for information could be sent to these staging servers and filled, perhaps automatically, by the company’s own data mining software.

What’s missing from the equation is the context of the five explosive slides from the 41-slide presentation that have been made public. Gellman, the Washington Post, and The Guardian have all declined to release any further information from the presentation, presumably worried that the other 36 slides all contain information that is better off not public. Unfortunately, the excerpt they have released doesn’t really tell us much about what Prism is — we need to rely on the article’s commentary for that. That leaves the door open for the possibility that all these CEOs are denying that their companies are involved in a top-secret Prism project because they aren’t — although data gathered from their servers clearly does ultimately make its way into the Prism system.

Prism: Hiding in plain sight

Rather than being the most secret of programs — so secure that none of the affected companies appear to know they are participating — it is entirely possible that Prism is actually just the reverse. Prism is also the name of a web data management tool that is so boring that no one had ever bothered to report on its existence before now. It appears that the public Prism tool is simply a way to view and manage collected data, as well as correlate it with the source. Clearly it is an important piece of the “toolchain” for intelligence analysis — which would explain why the leaked presentation says Prism was referred to more often than any other system in national intelligence briefings. But if the Prism tool is the same one referred to in the leak, it isn’t actually the source of any information (although it does have its own source designation), nor is it particularly nefarious. It is only a tool — which is more or less what Director of National Intelligence James Clapper said in response to the leak. What is much more important is to pay attention to what data is being collected, andhow.

Trying to dampen some of the rampant speculation that has resulted from its initial coverage, The Guardian has just released a fifth slide from the leaked presentation. It shows that, according to the presentation at least, Prism was not a direct intercept of communications from cables (although other programs are), and reiterates that it does get data from internet companies. Unfortunately, I suspect all this latest release will do is make everyone more curious about the rest of the context for these slides. Gellman has said that his source at the NSA expects to be made public eventually — and may have just been identified as a contractor and self-proclaimed whistleblower, Edward Snowden — so perhaps it is just a matter of time until we learn more about his involvement and motivations.

Subpoenas and metadata

Most of us probably haven’t read the fine print on what the government needs, legal-wise, to gather data about us. We may have some vague notion of “subpoenas,” “warrants,” and “court orders” but we probably don’t know much about the difference. In that case, reading Google’s transparency report provides us with the helpful information that, without any court involvement, a US government agency can use a subpoena to get all sorts of information about our communication. It can get our name, IP address, and call records, for example. Enough to build a connectivity graph and presumably allow for more targeted warrants or court orders (that do require judicial involvement) regarding the actual contents of email or phone calls. Equally important is the statement that those contents are not divulged by Google in response to “mere” subpoenas from the federal government.

Data logging declared less harmful than data mining

We’ve all heard the question, “If a tree falls in the forest and no one is there to hear it, does it make a noise?” While data logging doesn’t actually send trees crashing to the ground, the federal government has provided its own answer to this question by claiming that the act of logging data is not the same as actually looking at it. That is why the FBI has been able to get full access to the call logs (metadata) of Verizon and apparently AT&T and Sprint as well. Actually tapping into the calls requires a much more specific court order or warrant. Large-scale US government data mining of US citizens is also usually thought to be illegal, but with the revelation that nearly everyone’s call logs are being turned over to the government, I’m sure there will be a lot more questions asked about whether it is happening anyway.

Of course, one obvious purpose of all this logging is to later be able to to go back and do data mining on selected individuals or groups with the appropriate court-approved permission. If you’re the NSA’s big-data gurus, this makes perfect sense. Once you’ve got a lead on a potential terrorist or other enemy of the US, you can instantly find out who they have been in contact with, where they have been, and in turn who those people have been in contact with, etc. Without this preemptive data collection you’d have to start monitoring from that point forward, and it might well be too late to identify other key allies of the person suspected.

As to whether it is a good thing that our government is allowed to collect this data, it is ultimately up to Congress to settle the issue — and then for the executive branch to comply. Under both Republican and Democrat presidents, and Congresses, sweeping legislation permitting this type of surveillance has been passed since September 11, 2001 — including the Patriot Act, the Protect America Act and the FISA Amendment Act.

Enemies foreign and domestic

The NSA’s charter is to protect the US from foreign adversaries. In that role it has received greatly expanded powers from the Protect America Act of 2007, enacted along a mostly party line vote. The Act has been interpreted to allow snooping on conversations that have at least one endpoint overseas without any type of court order — although Clapper’s statement in response to the Prism leak says that all data gathered is pursuant to a FISA court order. It is this type of broad permission that is likely to have led to the deluge of information the NSA has been seeking, and apparently receiving, from internet companies. The Prism program clearly takes advantage of the fact that much of the world’s internet traffic travels through the US. As a result, concern over the leak is natural, as it’s easy to imagine terrorists suddenly working to use services and servers not located in the US.

As far as the impact of the program on US citizens, unfortunately, in this internet era it is no longer trivial to know in advance where both ends of a conversation reside. So the foreign intelligence programs “incidentally” scoop up plenty of data about US citizens and even domestic phone calls. In theory those records are destroyed once it is determined they shouldn’t have been captured, but with all the layers of secrecy and obscurity, it is no wonder there is massive skepticism as to whether that really happens.

Perhaps now is the time to hearken back to the pleas from Google, Facebook and other internet companies to make the entire process of government record requests more transparent. With the current mishmash of often-secret laws (my mind still boggles a little about how a democracy can have a secret law, that you don’t know what it says or whether you’re violating it).

If all of this starts to worry you about your own communications, PC Mag has an article with some quick tips on keeping your phone calls more private, and ExtremeTech has some tips on how to surf safely.

Source

How the US government has spied on almost every American for a decade

In the last 48 hours, it has come to light that the FBI and National Security Agency (NSA) have direct, government-mandated, warrantless access to servers at Google, Apple, Facebook, Microsoft, Skype, and other major internet companies. Furthermore, the NSA also requested — and received — data about every domestic call that’s routed through Verizon’s wired telephone network. While we only have proof of this one request made to Verizon, it is almost guaranteed that the NSA also demanded the same information from every other wired and wireless telecommunications company in the US. In short, the last 48 hours have blown the doors off what is probably by far the biggest case of domestic spying on innocent US citizens.

This data came out via two leaks. First, the Guardian newspaper in England got its hands on a top-secret court order, issued by the Foreign Intelligence Surveillance court, demanding Verizon turn over all of its “telephony metadata” — all of the data pertaining to a call, but not the actual content of the call — on a daily basis, for all domestic and international calls terminating in the US. Second, the Washington Post got its hands on slides detailing the US government’s Prism program — a top-secret program that gives the US intelligence community direct access to the servers of nine internet companies, including Google, Microsoft, Apple, Skype, and Facebook. As far as we can tell, both Prism and the collection of telephony metadata has been going on for years — probably since soon after the Patriot Act was enacted in 2001, following the September 11 attacks.

What isn’t known at this point is just how extensive the FBI and NSA’s data gathering antics have been, but it’s fairly safe to assume that the US government’s dragnet is a lot broader than just these two cases. The slides released by the Washington Post say that Prism is the most prolific contributor to the President’s Daily Brief (PDB), with one in seven NSA reports citing data gathered by Prism, for a total of 1,477 articles read by Obama last year.

Perhaps most worryingly, though, there’s no clear evidence that the court orders and Prism are only being used to gather domestic intelligence; as far as we can tell, Prism seems to care little about whether the target is domestic or international.

What does Obama know about you?

So, through these secret court orders and Prism, what data does the US government have access to? Telephony metadata consists of phone numbers, the unique serial numbers of any phones involved in the call, the start and end time, and sometimes the locations of the callers. It does not directly identify any caller by name, but it’s relatively easy to make the jump from a phone number, serial number, or location to an actual name. As long as you’re building up a detailed network of who talks to who, when, and where, names aren’t all that important either. While the Verizon court order excludes the actual contents of phone calls, it’s not unlikely that there have been other secret court orders that also give the FBI and NSA access to the conversations.

The Prism program is potentially a lot more nefarious. According to the slide above, the US intelligence community has access to just about everything that you do, say, or post on Facebook, Google (Gmail, Search, YouTube), Yahoo, AOL, Microsoft (Hotmail, Skype), and Apple. As far as we can tell, there’s no separation between domestic and international citizens, nor innocents or people suspected of wrongdoing: Prism, in a word, appears to give the US government completely unfettered, warrant-free access to almost all of your online activity and communications.

Data… lots of data

One question that remains unanswered is how the US government accesses the servers of these internet companies, or how it receives telephony metadata from Verizon (and probably other phone companies) on a daily basis. Due to the sheer scale of the data involved — probably on the scale of terabytes or petabytes per day — and the distributed nature of the internet, it’s simply not realistic for the US government to have a single mega-hub, where every telco and internet company sends their data.

Back in 2006, a whistleblower reported that the NSA had a secret room in an AT&T switching center in San Francisco, where it could listen in on all phone calls and internet traffic that passed through it. It’s entirely possible that the NSA/FBI/US government has similar rooms at switching hubs and data centers throughout the US. It’s easy to imagine the NSA having an office at Facebook’s data center in Oregon, for example, and then forwarding any interesting information to the Pentagon.

Facebook’s Prineville, Oregon data center, where the NSA probably has an office that secretly snoops on your data

These local offices probably have direct access to the local Facebook/Google/Microsoft servers, probably via a special interface that limits what kind of data they can obtain. The slides mention that each internet company provides different data, presumably as stipulated by each company’s technology and privacy chiefs, so the US government doesn’t have unlimited access to the actual memory and hard drives of these servers. Again, due to the sheer volume of data, we’re probably not talking about human spies leafing through your Facebook photos — your data is probably gathered and analyzed by a computer, with computer vision, voice recognition, and other specialized algorithms sifting out the occasional photo of a homemade pipe bomb out of the millions of food, baby, and lolcat photos.

These threats are probably then packaged up and sent along to FBI and NSA analysts in Washington and Maryland, where actual terrorist threats are picked out and presented to President Obama on a daily basis.

Is the US government out of control?

After learning about the Verizon wiretap and Prism slides, by far and away the most common response from the media is that Obama and the US government are out of control. Within hours of the Verizon wiretap being made public, the Obama administration gave us the same old line that we’ve grown to expect: It might seem like we’re stripping you of your First and Fourth Amendment rights, and Due Process, in the search for terrorists, but really, trust us, we know what we’re doing. This platitude was never that persuasive at the best of times, and at the worst of times — when the administration is caught doing something particularly heinous — it almost seems like they’re laughing in our plebeian faces.

Ultimately, any government will use the powers that it is given, whether for good or evil — and the Patriot Act gives the US government an astonishingly broad carte blanche when it comes to surveillance. Thethis i sad fact is, under the Patriot Act, the Verizon wiretap and Prism are probably perfectly legal. If anything, this story is a prime example of why laws should never be rushed through Congress, especially when those bills are as massively overreaching as the Patriot Act.

Moving forward, as a US citizen, an international citizen who uses internet services based in the US, or if you’re simply unlucky enough to be based in a country that routes most of its traffic through the US, all we can really do is pray that the Patriot Act is repealed. With terrorism in the US seemingly on the rise throughout the 2000s, and showing no sign of slowing this decade, it would appear that the Patriot Act, for all of its wanton decimation of our civil liberties, hasn’t done a whole lot to curb terrorism. It’s not that simple, though: The intelligence community would undoubtedly claim that there would be a more terrorism without these Big Brother-like measures — a claim that’s awfully hard to refute, when all of the data is top secret. With enough of an uproar, though, much like SOPA, the Patriot Act could be defeated.

For the time being, if you’re worried about Uncle Sam reading your messages and looking at your photos, your best bet is to stop using big, US-based internet services such as Google and Facebook, and to ensure that you always use encrypted connections (HTTPS) when surfing the web.

Source

Researchers create temporal cloak that can erase digital data from history

What if you could harness the fundamental nature of the universe to make your data completely secure? It might be possible sooner than you think. Researchers have managed to create a so-called temporal cloak that acts as a pocket in time through which data can pass and leave no trace. The temporal cloak is so complete, the event is entirely obliterated from history.

Recent years have been filled with announcements of rudimentary invisibility cloak technology that relies on bending light around an object. The temporal cloak idea is built on the same principles, but in this case light waves are pulled apart then compressed to generate time pockets that cloak events.

The first cloaking attempts in 2010 using laser pulses proved interesting experimentally, but the duration of the pockets was measured in a few picoseconds — not nearly enough to do anything useful. A team of researchers led by Joseph Lukens at Purdue University has been improving this process, carving out larger and larger pockets. The most recent breakthrough from this lab, as reported in Nature, makes use of a wave phenomenon called the Talbot effect. This time researchers have managed to completely obfuscate a huge amount of data.

The Talbot effect describes the way a diffraction grating produces an interference pattern of peaks and troughs when light is passed through it. The trick to creating the desired temporal effect is to convert the Talbot effect from something happening in space, to something happening in time.

The technique involves passing a laser through a waveguide with an oscillating voltage. The change in voltage alters the speed at which the light travels through the waveguide and produces an interference pattern over time. Because the frequencies are known, the peaks and troughs can be predicted. It is in the troughs, where light intensity is lowest, that we find the time pockets.

With some frequency tweaking, the team was able to expand the time holes to a whopping 36 picoseconds. A picosecond is a trillionth of a second, so this might not sound like a terribly long time. For lasers, it’s a relative eternity, though. The researchers were able to successfully hide a data rate of 12.7 gigabits per second in an optical cable — almost half the total bandwidth.

This process ends up a little too perfect when it comes to data security. It’s as if the cloaked data never existed, and that makes it difficult for a valid recipient to get the message — forget about a malicious third-party. However, Lukens believes future modifications will allow secret messages to be sent with the temporal cloak. In the meantime, the same process could be used to improve the fidelity of data streams. The troughs produced by the temporal Talbot effect hide data, but the peaks could be used to reinforce it. Timing transmissions to the peaks could reduce crosstalk on busy lines and lessen the chance of corruption.

Most of the scientific advances around cloaking technology rely on exotic materials and special conditions, but not this time. Temporal cloaking can be accomplished in standard fiber optic cable. The authors speculate this technology could be used to hide and improve communications in the near future.

Research paper: doi:10.1038/nature12224 – “A temporal cloak at telecommunication data rate”

Source

A standard router can turn your entire home into a gesture recognition device

Modern technology has been attempting to turn gesture control into a main form of input for quite a while now. Game consoles like the Wii, peripherals like the Kinect and Move, and accelerometers in phones are the main culprits of getting us to flail around in order to perform a task on a computer. A new project wants to cut out the hardware middleman though, and aims to turn the air around you into a platform that can recognize gesture-based input.

It’s already possible to set up a system in an office or home that can recognize the gestures you make, but it’ll be something that was cobbled together, and there will be hardware all over the place. If someone created some software to run the system, a bunch of Kinects placed throughout the home could get the job done. Dispensing with the need for much hardware, though, a team at the University of Washington has managed to coax WiFi signals into recognizing basic gestures. The system doesn’t need any kind of camera, nor does it need users to wear any kind of sensor. All the system needs is a modified WiFi router, and a few wireless devices to bounce the signal around.

Dubbed WiSee, the way the system works is likely more simple than it may initially seem. In this day and age, wireless signals fill up just about every room we’re in. If the human eye could see the signals, your home and office would look like some kind of trippy nightclub. Though WiFi signals can travel through most standard obstructions, such as walls, there is still a change in frequency in the signal when something passes through it — known as the venerable Doppler shift. So, the University of Washington team built a receiver that could not only notice when there was a frequency shift in wireless signals, but could measure the shifts in such a way where they could be translated into gestures. The device — which can be made out of a standard wireless router given the proper modifications — doesn’t just measure a specific wireless signal, but recognizes most of the signals being spewed out by standard devices found in the home, such as smartphones and laptops. This means that if you live in a modern home, the system doesn’t necessarily require anything more than the initial modified router.

The receiver requires a few antennas to properly function, but that’s certainly less of a hassle than setting Kinects up all over the house. The antennas hone in on a specific users’ movements, so multiple people can use WiSee in the same area at the same time. In order to prevent confusion, however, WiSee requires an initial gesture combination in order to unlock and gain access to the device. So, not only does this prevent an unwanted guest from using the system, but it prevents WiSee from accidentally registering unintentional gestures when in actuality you were just stretching out on the couch.

So far, WiSee can recognize nine different gestures in total, ranging from pushing and pulling, to more complex gestures like bowling. Five participants tested the system using 900 gestures in a two bedroom environment, as well as an office. WiSee managed to recognize 94% of the performed gestures.

Though the consumer market has only recently received technology as sophisticated as the Kinect, WiSee — or a system like it — does seem like motion and gesture control’s destiny. Removing the need for many peripherals significantly lowers the barrier to entry, and not having to be perfectly situated in front of a camera makes using the system a casual endeavor. At the moment, it doesn’t seem like WiSee has any sort of consumer release planned, but considering the only thing it requires is a modified router and some regular wireless devices you most likely already own, it doesn’t seem too difficult to turn it into a consumer product.

Source

Apple CEO Tim Cook discusses the iWatch, iOS 7, iTV, and the future of the iPhone

At the D: All Things Digital conference in California, Apple CEO Tim Cook has kicked off the proceedings by showing strong interest in wearable computers — but dissing the concept of Google Glass, while extolling the virtues of a (theoretical) wristwatch-type device.

Speaking to Kara Swisher and Walt Mossberg at D11, Cook said wearables are “incredibly interesting,” and could be a “profound area” if done right. Seemingly, he doesn’t think all that much of Google Glass, saying it won’t have broad appeal outside of specialized markets. The main problem when it comes to wearable computers, according to Cook, is getting people to actually wear it.  ”[There's] nothing [currently in the wearables space] that’s going to convince a kid that’s never worn glasses or a band or a watch or whatever to wear one… I wear glasses because I have to. I don’t know a lot of people who wear them because they don’t have to… To convince people they have to wear something, it has to be incredible.”

When pushed on whether Apple is developing an iWatch or not, Cook remained coy, merely saying “I see it as a very important branch of the tree. I think the iPhone pushed us forward fast and the tablet accelerated it. I think wearables could be another branch.” Cook said that he likes the Nike Fuelband, but “the [the wearable computers] that are doing more than one thing, there’s nothing great out there that I’ve seen.” In short, if there is an iWatchin the offing, Apple is going to make damn sure that it’s awesome. There’s pressure on Apple to release a new killer platform, to follow up from the iPad, but I wouldn’t be surprised if we’re still a year or two away from an iWatch.

In his D11 interview, Cook also tackled the rampant growth of Android smartphones and tablets, Jony Ive working on iOS 7, the elusive Apple iTV, and the future of the iPhone. The full interview is embedded below, too.

Mossberg pressed Cook on whether we can expect a range of different iPhone colors and sizes, like the iPod. “We haven’t so far. That doesn’t shut out the future.” Later, Mossberg mentioned that there seem to be lots of people who prefer smartphones with larger screens, or small tablets with stylus support, but again Cook made it sound like Apple is very happy with its current screen sizes.

When it comes to Android, Cook seems to be unphased. ”For us, winning has never been about making the most. Arguably we make the best PC, we don’t make the most. We make the best music player, we wound up making the most. We make the best tablet, we make the most. We make the best phone, we don’t make the most phones.” Cook also mentioned that it’s not just market share that counts — usage share is important, too. Citing an IBM study, Cook said that there are twice as many e-commerce transactions on iPads than all Android devices (tablets and smartphones) combined.

On the topic of iTV, “it continues to be an area of great interest for us,” said Cook. For now, though, the focus is Apple TV, which seems to be doing rather well: “We’ve now sold over 13 million [Apple TV boxes]. About half of those in the last year.”

Finally, Cook confirmed that Jony Ive — Apple’s lead hardware designer — is working on iOS 7′s redesign, and that we’ll see both iOS 7 and the next version of OS X at Apple’s Worldwide Developers Conference in June. “We recognized that Jony had contributed significantly to the look and feel of Apple over many many years, and he could do that for our software as well.” The current rumors are that iOS 7 will be much flatter and do away with the skeumorphic elements that underpin a lot of Apple’s interface design. We’ll find out for certain in a couple of weeks.

A bunch of interesting people are scheduled to talk at D11 over the next few days, including Elon Musk and Steven Sinofsky, so stay tuned for more coverage.

[Quotes sourced from various live blogs, including The Verge]

Source