Hugo Barra is returning to Silicon Valley, just over three years after he left Google to help turn Chinese smartphone maker Xiaomi into a global company.
During Barra’s time in Beijing, Xiaomi has grown far beyond its home market with its strategy of selling stylish Android phones on thin profit margins. In January, it made a splash at the CES trade show in Las Vegas, capping a series of international launches that had taken the company into over 20 countries, including India, Indonesia, Singapore, Malaysia, Russia, Mexico and Poland.
Barra, once Google’s vice president for Android product management, announced in a Facebook post Monday that he planned to return to the U.S. for personal reasons.
He had concluded it was time to leave Xiaomi, now that the company’s global business “is no longer just an in-house startup,” he wrote.
But Barra’s real motivation for leaving Beijing and returning to Silicon Valley now is personal, he said. “The last few years of living in such a singular environment have taken a huge toll on my life and started affecting my health.”
He also said he wanted to be closer to friends, family and what he considered to be his home.
Xiaomi CEO Lei Jun has asked him to remain an advisor to Xiaomi indefinitely, Barra wrote. He will remain with the company until after the Lunar New Year, which will be celebrated at the beginning of February this year.
Barra announced his move on Twitter and Facebook, a contrast with his departure from Google, which he announced on Google Plus. His public profile there appears not to have been updated in 18 months.
Microsoft is expanding the free services it offers to help enterprise customers adopt its products.
The company announced Monday that its FastTrack services now encompass Dynamics 365, Windows 10 and Microsoft Teams. Those come in addition to Office 365 and the Enterprise Mobility Suite, which were already covered under FastTrack.
Each service is designed to help companies get the most out of products they’ve already subscribed to and adopt products that the company has recently released. FastTrack includes adoption guides, tools to find a Microsoft partner company to help with migration, and consulting help from Microsoft employees on particular issues.
FastTrack began life a year and a half ago as a one-time service to help companies migrate to Office 365. Since then, the service has evolved to support continued use of Microsoft products and adoption of new features.
Organizations evaluating a Windows 10 deployment will be able to use FastTrack in planning a rollout and upgrading users. For Dynamics 365, FastTrack will offer expert guidance on how to get the most out of Microsoft’s suite of business apps, which include CRM and operations management tools.
Getting customers to use the subscription services they pay for is important to the long-term success of products like Office 365 and Dynamics 365. If companies aren’t using the subscriptions, they won’t renew them.
It’s a common issue for most enterprise software companies, including Oracle, IBM, Salesforce, SAP, and smaller players like identity and access management vendor Okta.
“In our world, if the product doesn’t get deployed, [customers] don’t renew. They’re not successful, we’re not successful,” Okta CEO Todd McKinnon said at a press event last week.
Including Windows 10 in the FastTrack service should help Microsoft get customers to upgrade to the new OS. It will provide a centralized set of resources for IT administrators dealing with difficult aspects of migrating their users, like ensuring apps are compatible with the new OS.
On the Dynamics 365 side of things , Microsoft is offering technical talks, workshops and ongoing calls from its engineers to help with adoption of its products, which launched last year. It’s a suite of different products that compete with the likes of Salesforce, Zendesk and ServiceNow.
The Microsoft Teams support for FastTrack includes guides that help walk users through scenarios that might be helped by the live collaboration service.
Cyril Belikoff, a senior director of Office marketing at Microsoft, said in a blog post that the company is making those resources available ahead of the launch of Teams so it’s easier for administrators to get people to use it.
Samsung Electronics said its profit rose almost 20 percent in 2016 despite nearly flat sales and the costly recall of the Note 7 smartphone.
The company reported an annual net profit of 22.7 trillion won (US$19.5 billion), up 19.2 percent, on sales of 201.9 trillion won (US$173.5 billion), up just 0.6 percent.
Samsung saw some of its biggest sales gains in memory chips on the back of strong demand from smartphone makers. The company is one of the world’s biggest memory chip manufacturers.
The mobile division saw sales fall 3 percent in 2016 to 97.8 trillion won. It’s Samsung’s biggest division by sales and was hit by the recall of the Note 7 smartphone and slowing demand for high-end phones.
A day earlier, Samsung divulged that battery problems were to blame for the fires and explosions that led to the massive recall. It said batteries from two manufacturers suffered from internal problems that caused a short circuit and fire.
Looking ahead, Samsung said it expects the smartphone market to slow down while artificial intelligence services present a new business opportunity. To date, Samsung handsets have featured Google’s AI personal assistant, but the company reportedly is developing its own assistant.
Samsung also signaled that it plans to boost the competitiveness of its mid-to-low-end smartphones by adding features that were previously only available in more expensive models.
In the consumer electronics business, sales barely rose but profit more than doubled to 2.6 trillion won. Samsung said it sees high-end QLED televisions and large-screen 4K televisions as key to growth in 2017. It will also attempt to expand the business-to-business side of its home appliance division.
President Donald Trump has named Commissioner Ajit Pai, an outspoken opponent of the FCC’s net neutrality rules, as the next head of the agency.
The choice was widely expected after Trump’s election last November. Pai is the senior Republican on the commission, having served since 2012. He doesn’t need to be confirmed by the Senate because he is already on the Commission.
Pai attacked the reclassification of broadband as a utility in 2015, saying it would place excessive burdens on service providers, other internet players and consumers. The expansion of broadband service through a competitive marketplace has been one of Pai’s themes as a commissioner.
“I look forward to working with the new Administration, my colleagues at the Commission, members of Congress, and the American public to bring the benefits of the digital age to all Americans,” Pai said in a statement Monday.
Former FCC Chairman Tom Wheeler proposed the net neutrality plan and former President Barack Obama strongly supported it. Trump and Republicans in Congress have criticized the rules and are expected to target them as part of an overall push for deregulation.
Although Pai can take over as chairman without a confirmation vote, he will need to be renominated and reconfirmed at the end of this year when his current term on the Commission expires.
The FCC customarily includes three members from the president’s party and two from the opposition. If Trump follows suit, he will name one more Republican and one more Democrat. In addition to Pai, Democrat Mignon Clyburn and Republican Michael O’Rielly are on the Commission now.
Pai grew up in Parsons, Kansas, the son of immigrants from India, according to his biography on the FCC’s site. In addition to several positions as an attorney in the federal government, he was an associate general counsel at Verizon from 2001 to 2003.
Mike Pompeo was sworn in late Monday by U.S. Vice President Michael Pence as the new director of the Central Intelligence Agency, amid protests from surveillance critics who worry about his conflicting views on a number of key issues.
The oath of office was administered to him after the Senate voted in favor of his confirmation in a 66-32 vote.
Critics of Pompeo, a Republican representative from Kansas, are concerned that he may weigh in with the government on a rollback of many privacy reforms, including restrictions on the collection of bulk telephone metadata from Americans by the National Security Agency under the USA Freedom Act. There are also concerns that the new director may try to introduce curbs on the use of encryption and bring in measures to monitor the social media accounts of people.
The new CIA chief wrote in January last year in an op-ed that Washington is “blunting its surveillance powers” with measures like the USA Freedom Act. Pompeo had previously voted in favor of the legislation.
“Congress should pass a law re-establishing collection of all metadata, and combining it with publicly available financial and lifestyle information into a comprehensive, searchable database,” Pompeo wrote in the op-ed he coauthored with constitutional lawyer David B. Rivkin.
During a recent confirmation hearing, Pompeo did not provide details of limits he would accept, if any, to the new surveillance powers he has suggested.
His proposals have not unexpectedly irked critics of the government’s surveillance, who are concerned that Pompeo has not clarified when asked for details about the kind of data he would like the government to collect.
“After two rounds of submitted questions and a hearing, we still don’t know what Congressman Pompeo meant when he referred to ‘all metadata’ or how he defines ‘publicly available financial and lifestyle information,'” said Ron Wyden, Democrat senator from Oregon, in a speech in the Senate.
Wyden said that on issue after issue, whether on surveillance, torture or Russia, Pompeo had “taken two, three or four positions, depending on when he says it and who he’s talking to.”
The CIA last week updated rules governing the collection, retention and dissemination of information on U.S. persons with an eye to addressing concerns about the collection and handling of information on U.S. persons in the course of overseas surveillance. One of the measures was to put a five-year limit on the holding of sensitive data. The rules introduced under the administration of former President Barack Obama come into effect on March 18.
For years, Apple Macs dominated the classroom. Then Chromebooks took over. Now, Microsoft is once again pushing for kids to learn via new, cheap Windows 10 PCs, managed by a new Intune application designed specifically for education.
Chromebooks aren’t much more than web browsers built on top of a bare-bones OS. That can be a drawback for businesses, which demand a rich platform for computing. In classrooms, though, the relative simplicity of the platform can be a positive, as they’re easier to manage.
To counter that advantage, Microsoft is debuting Microsoft Intune for Education, a cloud-based version of the Intune application- and device-management service to make managing 30 or 40 laptops per classroom less of a chore. Schools can customize over 150 granular settings, assign them to a student, and apply them to hardware, apps, browsers, the start menu, Windows Defender, and more, Microsoft said in a blog post. For example, a particular student can be assigned to a specific group, and be automatically assigned the same apps as other members of the group.
Managing a few devices can be handled by just logging into the service with an Office 365 login. A separate “set up school PC” app can handle larger deployments, Microsoft said. Intune for Education will be available in preview in the coming weeks and more broadly available this spring for $30 per device, and via volume licensing for 69 cents per month per teacher.
“Ultimately, both Intune for Education and [Google’s] Chrome Management console provide cloud-based device management for classroom devices, but Intune for Education integrates better with the rest of the existing school infrastructure and allows access to full Intune for management of Android, iOS and Windows devices,” Microsoft said in a statement.
According to a September 2016 report by FutureSource, about half of the computing devices in U.S. classrooms are Chromebooks, the inexpensive Chrome OS-powered devices manufactured by Samsung and others. Microsoft hopes to regain dominance via its partners, who are marketing inexpensive PCs that can compete on price while offering capabilities like touch and electronic inking that Chromebooks lack. Still, at least in the classrooms we’ve seen, Chromebooks seem to have a solid foothold.
Although Windows Ink is in its infancy, Microsoft pointed to apps like FluidMath and Drawboard PDF as two educational apps that would benefit from a touchscreen and stylus—and that Chromebooks wouldn’t be able to match.
“We believe Windows Ink is a truly valuable tool especially in the classroom, in fact, handwriting and neuroscience studies report 36 percent better test scores for kids that have diagrammed their notes,” Microsoft said in an emailed statement.
At the BETT conference this week in London, Microsoft said it will show off existing PCs designed for education, such as the HP Stream 11 Pro G3, starting at $189. Three others were announced as well:
With four usage modes, the 11.6-inch convertible notebook can be folded back via its 360-degree hinge to approximate a tablet. The $299 notebook includes a stylus and touchscreen supporting Windows Ink, a battery good for 13 hours of work, a shock-absorbent bumper and spillproof keyboard.
The TURN T201 is a PC designed for students aged 12 to 18, with a Microlens microscope and a thermal probe.
Lenovo has also updated its education PCs with the addition of the ThinkPad 11e, either in a traditional clamshell or Yoga form factor. The 11e’s battery supports 11 hours of battery life and comes with a stylus. The N24 does as well, plus a 11.6-inch IPS display. Lenovo offered no other details, save for its expected release in mid-2017.
Microsoft also added some new features to its Minecraft: Education Edition. “Global Pause” allows teachers to force students to take a break, while improvements like text-to-speech for in-game chat has been added for greater accessibility. Other new features include an updated user interface for managing in-game settings, new “mobs” includong polar bears, 256-block world heights for greater building capabilities, plus gameplay from the first night to The End dimension.
This story has been updated at 9:37 AM with additional details and to correct the volume pricing of Intune for Education.
Cord-cutters who’ve been pining for a la carte TV subscriptions might want to start paying attention to Amazon Channels.
The service, which launched in late 2015, allows Amazon Prime members to add on-demand video subscriptions from premium networks like Showtime and Starz, international services like Acorn TV and DramaFever, and streaming-only offerings like AMC’s Shudder and NBC’s Seeso. Amazon Channels isn’t exactly a la carte nirvana—there are no live channel feeds, and no basic cable offerings like FX and Comedy Central—but it does help make streaming video less chaotic. Each channel integrates with Amazon’s existing video apps and Fire TV devices, and ties into a single billing mechanism that, unlike cable, lets users add and drop channels at will.
The initiative seems to be gaining traction. Amazon now offers more than 100 channels, up from 75 in November, including HBO and Cinemax, which signed on last month. And while Amazon doesn’t share specific subscriber numbers, the company says it has “millions” of Channel subscriptions.
What does all these mean for cord cutters and TV networks? I jumped on the phone with Michael Paull, Amazon’s vice president of digital video, to find out.
Although Amazon Channels removes some of streaming-video’s billing headaches, and might spare you from jumping between so many separate apps, subscribing to multiple channels won’t save you any money. HBO still costs $15 per month, equal to a standalone HBO Now subscription. The same goes with Starz ($9 per month), Seeso ($4 per month), and Shudder ($5 per month).
Paull says that over time, Amazon might consider packaging channels together at a discount. Though he wouldn’t get into specifics on how this might work, it’s not hard to imagine Amazon as a TV bundler for the streaming age.
“I can’t really comment on the rights that we have and the relationships we have with our partners. What I can say is based on conversations we’ve had with many of the channels that we have on our platform, they would be open and excited about putting packages together,” Paull says.
The other potential snag with Amazon Channels is that you need the Amazon Video app (or an Amazon Fire TV device) to access the content. So if you subscribe to HBO through Amazon Channels, you can’t easily watch Game of Thrones on your Apple TV or Chromecast. It’d be nice if subscribers could use the standalone HBO Go app as a workaround.
Paull says Amazon is open-minded about TV Everywhere authentication, noting that Showtime subscribers can use their Amazon credentials to log into the Showtime Anytime app on other platforms. But getting other providers to offer this feature takes some work on both sides.
“It is a two-step process,” he says. “One is, we need to give them the rights to do it, and then expose the capability, and two, they have to implement it … but we generally are very customer-focused, so we want people when they pay for programming to be able to watch it anytime on any device.”
Amazon Channels is only as valuable its content, but right now it doesn’t offer the vast majority of channels people get with a cable subscription. Might that eventually change?
Paull says he can’t comment on discussions with media companies, but notes that Amazon is already working with networks like NBC, AMC, and Lifetime—albeit on services that aren’t part of a standard cable bundle.
To lure media companies into participating, Amazon is trying to sweeten the deal with usage data. Compared to cable, Amazon can provide more granular—but not personally identifiable—information on what people are watching. Networks can then use that data to help them decide what to put in their Channels.
“One of the big things that we’ve helped some of our partners to discover is the importance of regularly refreshing programming,” Paull says. “So an example is, based on the experience we’ve had on our platform, we’ve learned that having a new tentpole show on a monthly basis is really important.”
Perhaps unsurprisingly, Amazon has also found that exclusive and original series help keep people from jumping ship on a particular subscription.
“I think having some programming that’s available on a non-exclusive basis is good, especially if it’s shows and movies that people will be excited about. But what drives these services, both in terms of acquisition and retention, is exclusive programming as well as original programming,” Paull says.
Amazon also isn’t above creating its own Channels to meet a particular need. That’s what happened with Anime Strike, a $5-per-month channel curated by Amazon. Paull says Amazon talked to a variety of media companies about doing an anime channel before deciding to put together its own. So far, he says, Anime Strike is “exceeding our expectations.”
Still, Paull isn’t worried that Amazon-curated channels might scare off media companies who have similar content. He sees Amazon Prime, for instance, as a “premium” service akin to Showtime and Starz, who were part of the original launch. Despite all the competition, HBO and Cinemax eventually jumped in.
“In our mind, we don’t want to have one option per category,” Paull says. “We want to have multiple options per category, and we think that will contribute to the flywheel that we have in terms of being able to acquire new customers and upsell them.”
Sign up for Jared’s Cord Cutter Weekly newsletter to get this column and other cord-cutting news, insights, and deals delivered to your inbox.
This story, “Interview: How Amazon Channels is building a la carte TV” was originally published by TechHive.
Google has cool technology to recognize images and speech, and IBM’s hardware can diagnose diseases and beat humans in Jeopardy.
Combine the two, and you get a powerful computer with serious brains.
IBM is merging Google’s artificial intelligence tools with its own cognitive computing technologies, allowing deep-learning systems to more accurately find answers to complex questions or recognize images or voices.
Google’s open-source TensorFlow machine-learning tools are being packed into IBM’s PowerAI, which is a toolkit for computer learning. The two can be combined to improve machine learning on IBM’s Power servers.
A computer learns as more data fed into its system, much like how a human learns. PowerAI and TensorFlow tools can help track patterns and classify data, and spit out approximate answers to queries. The answers will be more accurate as a computer learns more.
IBM already has Watson, which uses advanced software and hardware to find answers to complex questions. But PowerAI is more customizable compared to Watson, which is more of a pre-packaged set of hardware, software, and cloud technologies targeted at use models like medicine and finance.
Integrating TensorFlow into PowerAI will solve a big problem of installing Google’s machine-learning technologies on Power systems, said Sumit Gupta, vice president of high-performance computing and analytics at IBM.
Right now, customers have to download TensorFlow from Google’s site and install it separately, which can be time-consuming. The optimization of TensorFlow for PowerAI will make provide a hassle-free way of installing TensorFlow, Gupta said.
Google also upgrades TensorFlow every month or so, and keeping up is a headache for customers, Gupta said. IBM will update PowerAI with the latest TensorFlow tools every quarter.
IBM isn’t creating a fork of TensorFlow for PowerAI, though that could be on the roadmap, Gupta said. The company has created a version of the open-source Caffe deep-learning framework for its PowerAI toolkit.
Forking TensorFlow to work with specific hardware or applications is commonplace. Specific versions of TensorFlow have been created for use with GPUs from Nvidia, embedded devices, robots, and drones.
Integrating and optimizing TensorFlow for PowerAI also speeds up machine-learning. The new PowerAI is designed for IBM Power Systems S822LC server, which has the speedy Nvidia NVlink interconnect to connect to GPUs, where most of the computer learning takes place. The NVlink is faster than the PCI-Express 3.0 interface on most servers today. The S822LC server is designed to work with Nvidia’s Tesla P100 GPU.
Those who don’t own IBM hardware but want to test the latest version of PowerAI with TensorFlow can do so on a cloud service provided by Nimbix.
Beyond computer learning, PowerAI can help in fraud detection or recommendation engines, especially for shopping. As a computer learns from input, patterns, and classifications, it’ll be more adept at identifying suspicious transactions, which is valued by financial institutions, Gupta said. The recommendations engine can help online stores sell more products.
PowerAI is not as popular as Watson, which became famous as a high-powered computer that beat a human in jeopardy. But PowerAI, TensorFlow, and Watson overlap, and also complement each other to build powerful learning models, Gupta said. A company using Watson may plug in TensorFlow or PowerAI for specific computer-learning tasks.
Many machine-learning frameworks like Caffe, Torch, and Theano are now integrated into PowerAI.
Tired of conventional passwords? So is Nuance Communications, a tech firm that is promoting the human voice as a way to secure user accounts.
The company’s voice biometric product is among the technologies that promise to replace traditional—and often vulnerable—password authentication systems, which can be easy to hack. That isn’t the case with Nuance’s solution, the company claims.
“To determine if it’s you or not, we are looking at over 100 different characteristics of your voice,” said Brett Beranek, Nuance’s director of product strategy.
The need to move beyond passwords hasn’t been more urgent, given that hackers are routinely finding ways to steal them. Last year, Yahoo, LinkedIn and Dropbox all reported major data breaches involving account details such as email addresses and hashed passwords.
With such information, a hacker can plunder through an email account like suspected Russian cyberspies did to U.S. political figures in last year’s election.
However, security provider Nuance is trying to change the status quo. Already, banks and financial institutions have been deploying the company’s voice biometric technology to verify user identities.
“This is more secure than a password,” Beranek said. “We’ve had our customers report a significant reduction in fraud over PIN and password security solutions.”
The technology was first deployed in a customer call center back in 2001. Since then, it’s also been used in finance-related mobile apps and to secure PCs at a handful of organizations, Beranek said.
Every human voice is unique, he added. Factors like a person’s larynx, the shape of the nasal cavity, and whether the subject is missing a tooth, will all determine the way someone sounds. People can also speak in a more monotone or lively manner, or space out their words in varying rhythms.
Nuance’s technology has been built to analyze these differences to accurately determine who is who, Beranek said. It’s been refined to the point, it can weed out voice impersonators, digital recordings and synthetic voices that try to dupe its system.
“In most cases, we can differentiate consistently between identical twins,” Beranek also said.
Replacing passwords is one thing, but what if your security system could also detect and kick out intruders who managed to break in?
SecureAuth is another company that’s been working on this very technology. It’s offering a system for companies to go “passwordless,” which can also spot any unusual activity on a user account.
The approach leverages an existing device many people have: smartphones built with fingerprint readers, said Keith Graham, CTO of SecureAuth.
Essentially the hardware is replacing the password. When logging on to a system, the user will receive a notification sent to their phone that can only be unlocked through a fingerprint scan. Clicking on the notification will then grant access to the system.
However, SecureAuth’s authentication process is also on the lookout for unusual behavior from the user even after logon. For instance, it’ll examine inconsistencies with the person’s keystrokes, mouse movements, where the user logged in from, at what time, along with the configuration settings on the device.
In that way, SecureAuth can assess whether someone accessing the system is possibly a hacker or not.
“It doesn’t only matter how big a lock you have on the door,” Graham said. “It’s about how quickly you can respond to remove the attacker from the environment.”
In the future, authentication systems may very well act more like “fraud detectors,” said David Mahdi, an analyst with research firm Gartner. Tech companies are aggregating so much data about their users, they’ll be able to notice suspicious activity from normal behavior, and boot out suspected bad guys.
“The more data points you can look at, the better,” he said.
Companies including Google and Microsoft are also developing better authentication systems that rely on other biometrics such as facial recognition or even how a person walks. So it’s maybe only a matter of time before these technologies become more available and displace the password.
However, one major obstacle is getting the industry behind the same standards on authentication, Mahdi said. Many enterprises are also using legacy systems that they fear will break if upgraded.
“A lot of laptops have fingerprint readers,” he said. “But all the organizations that I talk too aren’t using them, because the laptops have a specific driver, and I’ve heard its a nightmare to push out updates to them.”
It’s also true that any biometric system, such as fingerprint and voice, might have drawbacks. For example, what if your phone runs out of battery, or if you’re in an extremely loud area?
“They all have their pros and cons,” said Mahdi. “There’s not one method to rule them all.”
Nuance Communications said its own system isn’t perfect. For instance, its voice biometric technology will have trouble working with people who have laryngitis or other throat-related illnesses.
However, the company says its solutions is an improvement over passwords, which users often forget. Nuance’s product can also be combined with other technologies for two-factor authentication.
“You could have two separate biometrics,” Beranek said. “It could be fingerprint.”
The war on insecure webpages has begun, and Mozilla fired the first shot.
Recently, Mozilla rolled out Firefox 51 to its mainstream user base. With the new release comes an insecure warning on any page that offers a login form over an HTTP connection instead of HTTPS. Chrome plans to follow suit with version 56, expected to be released to mainstream users on Tuesday, January 31, as Ars Technica first pointed out.
HTTP uses an open, unencrypted connection between you and the website you’re visiting that could be intercepted by anyone monitoring traffic between you and the site. For that reason, it’s never a good idea to share login or credit card information over an HTTP connection. Most major sites offer the encrypted version—HTTPS—but every now and then you’ll come across a site that doesn’t.
In Firefox, users who’ve recently updated their browser will see a lock icon with a red strike through it next to an information icon in their URL address bar. These icons appear together when a user lands on a login page with an insecure HTTP connection. If you click on the icons you’ll see a plain-language explanation that the site is not secure, and a warning that any logins on the page could be compromised.
Chrome, meanwhile, will take a slightly different approach. Instead of a red strike through, Google’s browser will display an information icon along with the message “Not secure.”
Chrome will only display the warning while the login fields are visible. If you land on a site that requires you to click on a drop-down menu to show the login fields, for example, the “Not secure” message won’t show up until you reveal the login text entry boxes.
Mozilla announced in January 2016 it was working on an HTTP security warning, which first appeared in the developer edition of Firefox 46.
Google announced its plans to display a warning for insecure logins and credit card fields last September. The company also said this would only be the first step in a “long-term plan to mark all HTTP sites as non-secure.”
:Most major sites and services already use HTTPS for login connections, but every now and then you’ll come across a site that uses the insecure HTTP. When you do see an insecure login site, try typing https:// before the website name and hit Enter to see if that changes anything. Some sites do offer HTTPS connections but not by default—using the HTTPS Everywhere browser extension would automatically check this for you. If there isn’t an HTTP option on the site you’re visiting, you’ll have to weigh the risks of logging in over an insecure connection versus not using the site at all.