Nerval's Lobster writes "Software developer Jeff Cogswell is back with an extensive under-the-hood breakdown of Facebook's Graph Search, trying to see if peoples' privacy concerns about the social network's search engine are entirely justified. His conclusion? 'Some of the news articles I've read talk about how Graph Search will start small and slowly grow as it accumulates more information. This is wrong—Graph Search has been accumulating information since the day Facebook opened and the first connections were made in the internal graph structure,' he writes. 'People were nervous about Google storing their history, but it pales in comparison to the information Facebook already has on you, me, and roughly a billion other people.' There's much more at the link, including a handy breakdown of graph theory."
Make a difference in your data center. Sign up for SlashDataCenter Update newsletter now.
An anonymous reader writes "Google has recently announced changes to its image search. The search provides larger views of the images with direct links to the full-sized source image. Although this new layout is being praised by users for its intuitiveness, it has raised concerns amongst image copyright holders and webmasters. Large images can now easily be seen and downloaded directly from the Google image search results without sending visitors to the hosting website. Webmasters have expressed concerns about a decrease in traffic and an increase in bandwidth usage since this change was rolled out. Some have set up a petition requesting Google remove the direct links to the images."
An anonymous reader writes "Most of us are familiar with advertisements in online web searching, and by now we've grown accustomed to scrolling past the 'sponsored' results to get to the real responses to our query. And we know the ads are context-sensitive; for example, searching for our favorite Federation Starship will bring up ads for a similarly-named car-rental agency. But now a Harvard University professor has found a more disturbing trend in those contextual ads: racism. 'Sweeney says she has evidence that black identifying names are up to 25 per cent more likely to be served with an arrest-related ad. "There is discrimination in delivery of these ads," she concludes. Sweeney gathered this evidence by collecting over 2000 names that were suggestive of race. For example, first names such as Trevon, Lakisha and Darnell suggest the owner is black while names like Laurie, Brendan and Katie suggest the owner is white. She then entered these plus surnames into Google.com and Reuters.com and examined the ads they returned. Most names generated ads for public records. However, black-identifying names turned out to be much more likely than white-identifying names to generate ads that including the word "arrest" (60 per cent versus 48 per cent).'"
Nerval's Lobster writes "Whatever your actual feelings about football and this weekend's Super Bowl, you have to admire Wolfram Alpha's willingness to crunch any dataset and see what it can find. The self-billed 'computational knowledge engine' has analyzed the historical data for both teams involved in this Sunday's Super Bowl XLVII. Its conclusion? The San Francisco 49ers and Baltimore Ravens are 'annoyingly similar' when it comes to numbers, although some players stand out as potential game changers — if the math plays out right."
New submitter Flozzin writes with news of some resolution to the long-standing dispute that some French publishers have had with Google for republishing snippets of news reports without sharing revenue earned from the ads run alongside. Now, reports the BBC, "Google has agreed to create a 60m euro ($82m; £52m) fund to help French media organisations improve their internet operations. It follows two months of negotiations after local news sites had demanded payment for the privilege of letting the search giant display their links. The French government had threatened to tax the revenue Google made from posting ads alongside the results."
mask.of.sanity writes "Github has killed its search function to safeguard users who were caught out storing keys and passwords in public repositories. 'Users found that quite a large number of users who had added private keys to their repositories and then pushed the files up to GitHub. Searching on id_rsa, a file which contains the private key for SSH logins, returned over 600 results. Projects had live configuration files from cloud services such as Amazon Web Services and Azure with the encryption keys still included. Configuration and private key files are intended to be kept secret, since if it falls into wrong hands, that person can impersonate the user (or at least, the user's machine) and easily connect to that remote machine.' Search links popped up throughout Twitter pointing to stored keys, including what was reportedly account credentials for the Google Chrome source code repository. The keys can still be found using search engines, so check your repos."
Nerval's Lobster writes "Wolfram Alpha has upgraded its Personal Analytics for Facebook module, giving users the ability to dissect their own social-networking data in new ways. Wolfram Alpha's creators first launched its Facebook data-mining module in August 2012. Users could leverage the platform's computational abilities to analyze and visualize their weekly distribution of Facebook posts, types of posts (photos, links, status updates), weekly app activity, frequency of particular words in posts, and more. This latest update isn't radical, but it does offer some interesting new features, including added color coding for 'interesting' friend properties, including relationship status, age, sex, and so on; users can also slice their network data by metrics such as location and age." Wolfram users could also use some of that new site-specific searching power to come up with some unsavory results.
An anonymous reader writes "Australian surgeon Guy Hingston is suing Google in the U.S. for 'auto-complete' defamation. Typing in his name brings up 'Guy Hingston bankrupt' in the auto-complete. The association seems to have come about because Hingston purchased an aviation group CoastJet which went bankrupt two-and-a-half years later. Hingston himself was also bankrupted. Hingston claims this association has cost him customers and is suing Google for $75k, plus court costs. Google has often found itself the target of litigation over auto-complete searches. Are auto-complete results even useful? Should Google be policing the auto-complete suggestions?"
Nerval's Lobster writes "Last week, Facebook announced Graph Search, a system for searching the social network's vast collection of users, photos, and 'Liked' interests. But how will Facebook power it? The Disaggregated Rack, which will separate compute, RAM, storage, and caching functions in order to remain flexible in the face of Graph Search's changing needs. By breaking up resources and scaling them independently of each other, Facebook can scale without needing to constantly open up new servers and upgrade new hardware."
moon_unit2 writes "An AI researcher at MIT suggests that Ray Kurzweil's ambitious plan to build a super-smart personal assistant at Google may be fundamentally flawed. Kurzweil's idea, as put forward in his book How to Build a Mind, is to combine a simple model of the brain with enormous computing power and vast amounts of data, to construct a much more sophisticated AI. Boris Katz, who works on machines designed to understand language, says this misses a key facet of human intelligence: that it is built on a lifetime of experiencing the world rather than simply processing raw information."
Dupple writes "The head of French telecoms operator Orange said on Wednesday it had been able to impose a deal on Google to compensate it for the vast amounts of traffic sent across its networks. Orange CEO Stephane Richard said on France's BFM Business TV that with 230 million clients and areas where Google could not get around its network, it had been able to reach a 'balance of forces' with the Internet search giant. Richard declined to cite the figure Google had paid Orange, but said the situation showed the importance of reaching a critical size in business. Network operators have been fuming for years that Google, with its search engine and YouTube video service, generates huge amounts of traffic but does not compensate them for using their networks. An editorial piece at GigaOm says Google is abandoning its principles and giving Orange 'the incentive to demand the same from other content providers.'"
redletterdave writes "Google just purchased a 2.4-acre plot in the King's Cross Central development in London, where the company plans to build a brand-new, 1 million square foot office. Google reportedly invested about £650 million ($1.04 billion) on the property, which, when finished, will be valued at more than £1 billion ($1.6 billion). While Google traditionally leases its overseas offices, the company's decision to buy rather than rent in this case was likely tax motivated, since Google can't repatriate its cash to the U.S. without paying a hefty tax."
An anonymous reader writes "Google on Friday announced yet another security improvement for Chrome 25. In addition to killing silent extension installation, the omnibox in Google's browser will send all searches over a Secure Sockets Layer (SSL) connection. Chrome already does this for users who are signed in to Google: when they search from the address bar, their queries are sent over HTTPS. As of Chrome 25, however, the same will happen for users who aren't signed in to Google."
Frequent contributor Bennett Haselton writes with some strong cautions on a Facebook "feature" that lets you search for random phone numbers and find the accounts of users who have registered that number on their Facebook profile. This has privacy implications that are more serious than searching by email address. Especially in light of the expanding emphasis that Facebook is putting both on search qua search and on serving as a VoIP intermediary (not to mention the stream of robocalls that the FCC is unable to stop), this might make you think twice about where your phone number ends up. Read on for Bennett's description of the problem and some possible solutions.
The Washington Post (among many others) reports on a development from Facebook that may excite many more users than does the much-hyped announcement about richer search capabilities: after launching a Canadian trial balloon not long ago, Facebook is expanding the reach of its free in-app VoiP communications with free voice comms via the company's smartphone app. "Excite" for some people will also mean "infuriate": to get the free candy, the recipient will need to have shared his or her number with Facebook, which many people will understandably be loath to do. From the WaPo article: "To use the feature, Facebook users must hit the 'i' info icon in the corner or a conversation or contact information page. That panel has a 'Free Call' button that you can use if your friend has shared a mobile number with Facebook and is available for a call. The company slowly has been building out the features available in chat — most notably with the 2011 Skype partnership that put video calling on the Web version of its site. When it released Facebook Messenger last fall, it became even clearer that messaging and mobile applications were priorities for the company."