In all of the razzle-dazzle coming out of Google I/O Wednesday morning, it's easy to forget about the company's core product: Search. But Google didn't forget about Search, showing off a number of interesting improvements, including a new hands-free feature that will bring voice-activated searches to desktop computers.

(See also Google Is Turning Search Into The Planet's Biggest Anticipatory System.)

Google Loves Star Trek

The new voice commands for Google Search are the culmination of Google's fixation on the Star Trek computer interface of the future. It's something that Senior VP Amit Singhal very much wants to see - and Google Chrome browser users on desktops and laptops will get to share in the experience, simply by saying the phrase "OK Google" and then speaking the desired search term. No need to even press a button.

Voice search is already familiar to iOS users, in the form of Siri, and Google Search for Android users. The roll out on Android no doubt served as a great testbed to get Google's voice-recognition algorithms better adapted to human speech patterns. 

According to a blog post by by Google senior VP Amit Singhal, the idea is to turn search into a natural language conversation: 

Soon, you’ll be able to just say, hands-free, “OK Google, will it be sunny in Santa Cruz this weekend?” and get a spoken answer. Then, you’ll be able to continue the conversation and just follow up with “how far is it from here?” if you care about the drive or “how about Monterey?” if you want to check weather somewhere else, and get Google to tell you the answer.

Reminder Cards & Knowledge Graphs

Voice was not the only addition to Google Search's feature set: the company also announced new reminder cards in its Google Now tool. Currently, Now's cards are based on a set type of search, such as sports team cards based on your frequent searches. Now users will be able to add reminders set to pop up as cards in Google Now whenever your location in time and space (your context) is close to the reminder. Like a note to buy milk when you go by the grocery store.

Google's search will also start to deliver information in a far more anticipatory way, using Google's Knowledge Graph to build and manage connections between searches and deliver information based not only on the user's immediate search, but potential searches that many people ask as a follow-up.

(See also Forget Searching For Content - Content Is About To Start Searching For You.)

All of these features move Google towards a more friction-free form of contextual search - and that promises to bring knowledge to users faster and more seamlessly.

Lead image by Nick Statt for ReadWrite.

Amid a blizzard of announcements at its I/O conference, Google unveiled a major change to its core search product in singularly low-key fashion.

Amit Singhal, a senior Google executive in charge of its search efforts, said that Google search, starting today, wouldn't just answer the question implied in the words we type into that ubiquitous search box. Google, Singhal said, would display results that "anticipate the next question."

Singhal gave the example of a search for "the population of India." The searcher wants a number, yes—but it's likely that they're on the hunt for more information about India after that. There's no reason why Google shouldn't get a jump start on answering those questions.

That word Singhal used—"anticipate"—is a powerful one in the field of artificial intelligence. Anticipatory systems, as we've written, are a hot field that's been moving from theory to practice. There are apps today that will recommend a restaurant, a purchase, even a date. And we've noted how anticipatory systems are the future of search.

But with this latest update, Google is launching the biggest real-world experiment in anticipatory systems ever, with hundreds of millions of its search users getting a glimpse of the anticipatory future today.

Google Maps, too, is getting an anticipatory update. Search for the "Walt Disney Family Museum," a popular San Francisco institution, and you might get another kid-friendly museum suggested, like the Exploratorium, based on the reasonable assumption that you're looking for a place to go with your family.

Singhal said the goal was to introduce "conversational search." To have a conversation, you need a conversational partner. The voice search features Google rolled out will likely get far more attention. But the anticipatory underpinnings of Google's new search are far more crucial. Human beings are remarkably good at processing context to anticipate what the people we're talking to will say next. Now we can really talk to Google. Whether it's through a keyboard, touchscreen, or microphone doesn't really matter.

Photo by Nick Statt for ReadWrite

Contextual search and the Internet of Things are two key factors in how search is evolving from users actively searching for information to users receiving information as they need it.

But there is another key component that must be added to the search equation: the rise of intelligent software agents that will not only anticipate the information you need, but also act on that information to help manage your life.

(See also Forget Searching For Content - Content Is About To Start Searching For You and How The Internet Of Things Will Revolutionize Search.)

The Dawn Of The Bot Age

Call it artificial intelligence, software agents or even bots, the technology for search-related automated prediction and action has been in development for a long time. (In fact, I used to cover this topic when I was the managing editor for BotSpot.com at the turn of the century.)

(See also Futurist's Cheat Sheet: Artificial Intelligence.)

In those days, bot development was focused on creating automated software to handle the routine tasks that were proliferating on the then-fledgling commercial Internet. Web crawlers, software that actively seeks out and indexes websites for search engines, were one very popular use of software bots. But there was always a goal beyond the mundane world of Web crawlers and software trying to reasonably fake a Turing test to appear human: developers wanted the software to take specific tasks completely off the hands of humans.

Over a decade later, we may finally be getting to the point where bots can actually do that.

It is not that the development of intelligent agents stalled during the first decade of the 21st Century. Instead, we may not have been quite ready to implement them. Automating routine tasks just didn't seem like the top priority during the beginnings of the mobile revolution.

Now things have changed. First, and most obviously, mobile devices are everywhere. Second, there are now legions of interesting Web services to automate. The final ingredient is the most important: With the rise of Big Data, there is now enough information available for a software agent to actually use to perform anticipatory actions. In that context, the challenges of applying software agents and artificial intelligence to business solutions is nothing compared to the potential payoff to users.

(See also How The Internet Will Tell You What To Eat, Where To Go And Even Who To Date.)

Dr. Moshe BenBassat, CEO of ClickSoftware, it is fair to say, lives and breathes this stuff every day. BenBassat envisions a world where personal agents, which he calls “butlers,” manage the day-to-day planning and implementing of workflows.

The "Butler" Did It

BenBassat offers an example to illustrate: Imagine a service technician who logs into his smartphone’s service app and pulls up today’s schedule. His first appointment: the Acme Bank downtown. A few more swipes pulls the address from the calendar app and then brings up a map in the navigation app, and off the technician goes. When he arrives at Acme, he finds and calls the customer contact, who has to come down to the lobby and admit him into the building. Once arriving at the customer site on the 17th floor, the technician discovers he has left the replacement part in his vehicle, so he goes back down to get it.

In a scenario like this, BenBassat estimates the technician would spend 7-12 minutes just swiping and typing on the phone to find and use the data he needs. Over the course of the day, that adds up to a lot of lost productivity.

In a world staffed by BenBassat’s butlers, the scenario might unfold like this: The service technician logs into his phone’s service and is immediately informed about the first appointment: the Acme Bank downtown. The butler asks if the technician if he would like a map to the appointment, and after agreeing, off the technician goes, using the map to reach his destination. Just before he arrives at Acme, the butler autodials the customer contact and informs her the technician is about to arrive, so she can come down to the lobby and let him into the building. When the technician leaves his vehicle, the butler senses that the replacement part is not in toolkit the technician is carrying, and prompts the technician to grab it from the truck, saving the trip back down to the vehicle.

"Butlers" like the ones BenBassat describes promise to play a huge role in changing search - and by extension the way we work. Proactive software agents will reduce the need to waste time looking for information. Instead, information will be delivered right when we need it. As software agents get better at figuring out what we want, that information will become more useful and actionable.

We are almost there now: Contextual search tools like Google Now, which takes into account where you are and what you are doing to provide useful information, are the first big step towards anticipatory and responsive software agents.

(See also Google Now Knows More About You Than Your Family Does - Are You OK With That?)

The Interaction Issue

There is still a ways to go. Social interaction is seen as the biggest obstacle to effective software agents. Agents are only as good as what they are programmed to do, while humans have internalized a lot of common-sense tricks for interpreting reality. We know what we mean when we say, "find me some pizza," but the software agent might give you a map of nearby pizza places - or just call up pictures of pizza.

In the consumer world right now, Apple's Siri is the most well-known example thus far of how a software agent will interact with humans, though it has its limitations, both in speech recognition and plain common sense. As that interaction is smoothed out, though, it is not hard to imaging giving agents like Siri or Google Now's voice search more permissions to act on the information at hand, instead of just reporting it. Once that hurdle is overcome, all of that predictive and contextual information that the Internet is starting to finding for us will have a smooth, human-like interface and better able to help us manage our days.

(See also Who Has The Advantage: Siri Or Google Now?)

Why Search Anymore?

Searching for anything - be it on the Internet, your inbox or on your personal devices and services - will be far less necessary, both in business and personal contexts. Search is not just firing up Google, after all - it also includes combing through your own data for relevant information. When your spouse has a last-minute meeting and can’t pick up the kids from after-school sports, for example, you won't have to go though a complicated dance of multiple phone calls, texts and emails as you re-arrange both your schedules and stress out over making sure someone gets there on time. Instead, your search agents could analyze and coordinate both your schedules and create a single suggestion to line everything up. All you'd have to do is agree to the changes.  

The combination of automated agents, contextual search and a sea of data from our devices, services and the Internet of Things, search is poised to become vastly more useful and efficient than it already is. The pieces are getting there with agents like Siri and contextual search like Google Now. If it all works as promised, information we need will be delivered to us just when we need it, without our having to invest time and effort looking for it.

Image courtesy of Shutterstock.

Today, Google's Googliest project makes the leap from Android to iOS. Google Now, announced last June at the company's I/O 2012 conference, is part smart search and part personal assistant — but don't call it Siri. The service will make its debut on iOS through an update to Google's core Search app, available in the App Store.

According to Google's blog post on the release, "Today, with the launch of Google Now on iPhone and iPad, your smartphone will become even smarter. Google Now is about giving you just the right information at just the right time. Together, Google Now and voice search will make your day run a little smoother."

Google Now for iOS will be nearly identical to the Android release, though it won't enjoy the same deep integration as it does on Google's own mobile platform. That means no homescreen widget, of course, and no "swipe up" gesture for instant, fluid access. The iOS version will also be missing a few of the cards you'd find on Android: For now, cards for boarding passes, nearby events, Fandago and Zillow will remain an Android exclusive.  

A 20% Project That Took Off

We spoke with Google's Baris Gultekin, co-creator of Google Now, about the product's migration to that other platform. According to Gultekin, Google Now is the latest product home run with humble beginnings as a year-long 20% project (Google encourages employees to dedicate 20% of their time to a pet project that interests them).

"In the early days it was all about keywords," Gultekin explains. "With Google Now, you don't even have to search. We're really interested in having computers do all the hard work." 

For Google Now, the heavy lifting comes easy. A smart search app on steroids, it provides instant access to a spread of useful information, delivered via "cards". The cards are wholly dependent on context. As Gultekin puts it, "The product is different given the situation you're in." You might see a card for commute traffic around rush hour, or a card for your flight reservation the morning before you head to the airport.

Google Now Is Google, Now

Google Now is an umbrella project of sorts, tying Google's vast web of products together. Naturally, the product is also right at home on Google Glass, the company's futuristic eyewear that also aims to make this whole business of carrying the Internet less interruptive.

Google is betting big on Google Now, so it will be interesting to see if the service takes off in Apple's ecosystem. Google iOS ports like Google Maps are wildly popular, but will iPhone users take notice of Google Now? 

From its perfect morsels of context-dependent info to its uncanny knack for knowing what you needed to know before you knew you needed to know it, Google Now is a powerful tool — and a fun one.

Try it out today in the App Store and have fun pitting it against Siri in voice-powered search time trials.

You know you want to.

As mobile devices dictate the terms of search and how results are being conveyed to end users, there's another phenomenon that will greatly influence the future of search - very soon, we're going to be swimming in more data than we will know what to do with.

The rise of the Internet of Things means billions of physical objects will soon generate massive amounts of data 24 hours a day. Not only will this make traditional search methods nearly impossible to use, it will also create an environment where instead of looking for things in the world, those things will be seeking us out to give us all sorts of information that will help us fix, use or buy them.

(See also Forget Searching For Content - Content Is About To Start Searching For You.)

Search Collides With Internet Of Things

When talking about the Internet of Things, it is important to get past the hype and explain exactly what it is: vast numbers of automated physical devices and objects connected to the Internet. These devices are usually routers, switches, phones… but increasingly devices like security cameras and remote climate sensors are being added - and over time we can expect everything from cars to refrigerators to join the party.

(See also Futurist's Cheat Sheet: Internet Of Things.)

At this moment in history, the relationship between search and devices on the Internet resembles the model of searching for Web content: you search for data from devices, and you get it.

The Shodan search engine follows this model very well. Using Shodan is pretty much like using Google or any other search engine: enter your search string and up pops quite a few results matching devices that Shodan has found on the Web. Routers, cameras, municipal control systems… they're all there to see.

CNN, true to its alarmist nature, recently described Shodan as the "The scariest search engine on the Internet" - and indeed, there are some disturbing aspects to using Shodan. Entering the search term "cisco-ios last-modified" - a term suggested by Shodan itself as popular - reveals a number of Cisco-based devices on the Internet that had not been updated in a very long time. Such devices could be very vulnerable to attack through well-known exploits.

Shodan exposes what security experts have known for a long time: a lot of devices connected to the Internet have been largely forgotten even as they continue chugging along performing tasks that range from mundane to mission-critical.

But Shodan, while interesting, will probably not represent the future of searching Internet of Things. Instead, as the world around us "wakes up" and becomes increasingly connected to the Internet, the world will talk to us without our having to look for it.

When The World Touches You

Here is one vision of how this potential future might work:

Imagine, says  J Schwan, CEO of Solstice Mobile, walking into an office building in which you are a maintenance technician about to start your day. Over the course of the previous day and evening, various tickets were filed by the building's tenants (or maybe even by the devices themselves): a squeaky door here, a hiccuping elevator there. An app on your phone lists the problems to be fixed - ordered by priority and optimized for maximum efficiency. This is a system that Solstice is working on now for clients.

Other examples abound. A department store could be populated with products connected to the Internet via RFID tags. Customers could walk in, request a certain item, and see a map to that item's exact location displayed on their phones. Or even see it as an augmented-reality display on their Google Glass or other wearable computer.

The same store's managers could use augmented-reality displays to view sales figures of the items the see on the shelves, perhaps in the form of a heat map that instantly indicates which items sell better in what position in the store. Just by speaking a couple words, they can have the stock room shift the position, or even update electronic price tags on the shelves to attract more buyers. Those price updates, by the way, then show up as alerts on shoppers' devices.

Not Far Away

This world is not far off. Smartphones and other mobile devices can already tap into public search engines to discover more about the world around them. You can use augmented reality to see results displayed graphically on device screens.

As more and more objects join the Internet, they'll create information that will be added to the potential data you can receive, raising the level of information available by orders of magnitude. This will be both a boon (more data to help make decisions) and a curse (so much data you could drown).

Searches, as active tasks, will become less needed - and less practical - as not just the Web, but the whole world around us will automatically send us information based on our preferences. And the questions we will have to ask will be far simpler and less arduous to construct.

Instead of entering "2013 best televisions" in your browser to get back lots of data in the form of articles and content, you'll be able to stand in the store and ask "which one of these TVs is the best for me for me right now?" and see the answers (based on your budget, space and other preferences) right in front of you.

Knowing the tenacity of advertisers, you probably won't even have to ask.

Image courtesy of Shutterstock.

The world of search is about to be flipped completely on its head. As part of that sea change, today's reactive Web-based searches are about to give way to proactive, geo-fenced answers that will pop up before you even frame the question.

In many cases, you won't be searching for content - content will be searching for you.

Putting The New Search In Context

Search, to date, has mostly worked something like this: You type a word or phrase into a search bar in a browser or mobile app and a search engine with a funny name returns a list of Web pages it deems related to your query.

In recent years, search has gotten a lot better in a number of ways. One key improvement takes location into account. If I type "Notre Dame" while I'm in my hometown, then it's very likely I will get results about the University. If I were located near Cleveland, though, I might get results about Notre Dame College. And if I were in France, surely my results would focus on this beautiful edifice. 

Location is part of what experts call "contextual search," which becomes even more important with the rise of mobile computing. Where we are and who we are makes a big difference in the search results we want, and contextually aware search engines are working to use that information to decide what results to return to us.

According to J Schwan, CEO of Solstice Mobile, there are four aspects of contextual searching that all have to work together:

• Where
• Relevance
• Push
• Security and privacy

First, there's the where - what Schwan refers to as geo-fencing. Where you are, as noted above, makes a difference in what search results are most appropriate.

Then there's relevance, which dictates results through explicit preferences that you have set, the results delivered to other users in a similar context and what is going on around you at that particular time (traffic, weather, business hours, etc.).

The third aspect Schwan highlighted is relatively new, but fast-becoming more important to contextual search: push. Rather than waiting for users to search and then reacting to that query, data providers and search engines are working on how to push data to users based on their context. Google Now does this now on Android and its Chrome browser extension: cards based on your search results, location and even email messages will appear that give you the traffic report to get home or inform you of the latest sports score.

The final aspect is the wrapper of security and privacy that has to work with all of this to ensure a user's data doesn't go where it's not supposed to.

Squinting For SEO

Contextual searching is perfect for mobile, because, well, mobile users are by definition moving around. But the mobile form factor also makes contextual search more important.

Many people may have honking big 27-inch monitors on their home PCs, but relatively tiny smartphone screens inherently limit the amount of information we can access. In that context, it's even more important for mobile users to get the right results near the top of the results screen.

This is even more true when adding natural interfaces to search, such as voice-activated searching using systems like Apple's Siri. Forget search strings, Siri has to process natural-language queries and either speak or display usable results on a small screen. 

For search-engine optimization (SEO), this is a huge challenge: With contextual search, it's no longer enough to get your business or product listed on the first Web page of results. On a mobile device, as well as in push situations, SEO is really effective only if you can push your results into the top position, or at least into the first few lines.

Wearable devices like Google Glass and the rumored iWatch could put even more pressure on search results. We don't yet know what their interfaces will look like, but it seems safe to assume that there may be even less real estate available to display search results. 

This is one reason why the search engines are working so hard to deliver knowledge rather than just Web page links in their results. Google and Bing both now feature "knowledge boxes" that try to encapsulate the pertinent information about a topic in one glance. This "knowledgization" of search results is conducive to mobile search because it parses data into easily displayed and digestible chunks - essential for the smaller screen.

We may already be seeing the early effects of this trend. Last Fall, Google reported its first-ever drop in search volume. Some of this decline is no doubt attributable to competition - such as Bing, Yahoo or even local searches through services like Yelp. But how much of it is due to pushed content and knowledge replacing what might have otherwise been searched for? If the information being received is of better quality, then perhaps we won't have to search as much in the future.

By incorporating context and working towards knowledge - useful information instead of just plain data - the next evolution of search will take advantage of new opportunities and cope with new demands and challenges.

Will that help us make better decisions? We can hope.

Lead image and Notre Dame images courtesy of Shutterstock.

Roku debuted its next-generation streaming set-top box, the $99.99 Roku 3, on Tuesday night, complete with a faster processor, an updated interface, plus an innovative remote control that hides a headphone jack.

But Roku's secret sauce lies deep inside its box, under a heading marked "Search."

For many, Roku is synonymous with video streaming, combining virtually every online streaming service available - Amazon Instant Video, Crackle, Hulu Plus, Netflix, Vudu, and many more - under one umbrella. Although Roku has dabbled with games - it added a motion controller with the Roku 2, allowing users to play popular games like Angry Birds - the company's success has been based on simply providing the most movies and TV shows for the lowest price. In January, Roku announced that Time Warner Cable would bring up to 300 additional "live" channels to its customers who own a Roku, allowing them to tune into live TV without the need for a dedicated cable box.

(See also Roku Meets Time Warner: Baby Steps Towards A New Kind Of TV.)

In fact, the most distinguishing feature of the Roku 3 is its new remote control, which hides a headphone jack that turns off the volume of the TV and routes it through the jack, allowing a viewer to watch television without disturbing her partner. (In-ear earphones are included.) Aside from that, the Roku 3 features an updated processor for faster scrolling, Ethernet, 1080p support, a microSD slot... and that's about it. The one thing that users will immediately notice, the new UI, is also being rolled out to most of Roku's other devices in April, including the Roku LT, Roku HD (model 2500R), Roku 2 HD, Roku 2 XD, Roku 2 XS and the Roku Streaming Stick.

Search Becomes A Focus

All of that variety - Roku says it hosts up to 750 streaming channels - inevitably led to some form of consumer confusion. Last October, Roku unveiled its cross-service search application, allowing users to search for movies, TV shows, actors and directors across Netflix, Amazon Instant Video, Hulu Plus, Crackle, VUDU and HBO GO. Within the Roku 3, search has been made part of the core functionality, meaning that users don't have to load an app to discover which video service hosts "Cheers."

"It's what we call one-stop search," said Lloyd Klarke, director of product management at Roku. "Instead of having to go into Netflix's catalog and search there, come back out, go to Amazon, come back out - here, the objective is focus on what you want, we'll help you find it, and then you can find out where to go, and watch it."

After searching for an actor, like Daniel Craig, a list of his movies appear. Click a movie, and you can see which services offer the movie. Search results are divided into two groups: those services you subscribe to, and those you don't. Within each group, the results are ordered by price, automatically guiding you to the cheapest option. (Some results include the caveat "From..." indicating that a more expensive option might be available.) HD availability is also called out.

A Growing Problem

Several other companies have made a stab at cross-service search, but none seem to have done it quite as well as Roku.

Last year, Verizon launched its Viewdini app for LTE phones and tablets that are subscribed to its services, but customers have complained of crashing and limited search offerings. Clicker.tv (now TV.com) lets users search for a show and then find a particular episode; hovering over it brings up as list of services that offer the episode for purchase, but with no indication of the price. Clicker.tv/TV.com also doesn't seem to offer any way to narrows the search according to which services you subscribe to, let alone launch them. Flixster may be the best of the bunch: Searching for a movie brings up a list of services that offer it for rent, purchase and subscription streaming, as well as the struggling UltraViolet digital locker format. But the service indexes only movies.

Update: A couple of commenters have noted below that both Google TV and the Microsoft Xbox 360 both search across services, too. That's true, although Roku's implementation appears to be as efficient. Another, important consideration is price: while some Google TV "buddy boxes" are priced at the $99.99 price point of the Roku 3, several implementations are directly integrated into more expensive TVs. However, Google TV's PrimeTime feature taps into Google search, giving it a leg up in search complexity. The point, however, is that to my knowledge, no independent Web service offers the same search functionality as the Roku 3, which is one of the most cost-effective hardware streamers available.

Roku's search service isn't perfect - with so many channels to choose from, and thousands of titles to index, it may never be. It's also doubtful that Roku will ever be able to adequately index live content from TWC and any other providers with which it strikes deals. 

Apps like Peel, Thuuz and Netflix itself point the way toward recommending what you might want to watch, based on your historical preferences. But for me, recommendations are often guided by threads at forums I frequent, by people I know recommending movies they've discovered and I might want to watch. Being able to quickly find those movies and TV shows, as cheaply as possible, makes Roku's cheap streaming set-tops even more valuable. 

Microsoft's latest attempt to drag down Google has again fallen flat. But that hasn't deterred Redmond, which insists it will press on with its high-profile "Scroogled" attack-ad campaign on Google's Gmail and search. 

"Scroogled" launched last Thanksgiving across television, newspapers, YouTube and social media. The ads claimed that Google users were being Scroogled — a phrase intended to evoke the miserly Ebenezer Scrooge from A Christmas Carol — because the search giant reads their email (in order to sell advertising in Gmail) and delivers slanted results for product searches (because Google mixes paid ads with normal search results).

The campaign, however, hasn't exactly had the desired effect. While more than 114,000 people signed a Microsoft petition asking Google to "stop going through your email to sell ads," there's little sign it had any impact on Gmail or Google search. A report from KQED late last week suggested that Microsoft was throwing in the towel on the Scroogled ads, although Redmond now insists that's not true, as per this statement it provided to ReadWrite:

Scroogled will go on as long as Google keeps Scroogling people. We know Google doesn’t like it when the facts come out. Chapter two of the consumer education campaign has shown people care about their privacy. More than 3.5 million people visited scroogled.com, and nearly 115,000 people signed a petition asking Google to stop going through their Gmail. Stay tuned for the next chapter.

In other words: Pay no attention to that Microsoft official behind the curtain. We have always planned to run the Scroogled campaign until the end of time — or until we triumph over Google. Whichever comes first.

ReadWrite editor-in-chief Dan Lyons noted last month that the Microsoft campaign was little more than “another pathetic smear campaign" against Google:

The gist of the scare campaign is that Google is a scary, scary company that reads your private emails in order to send you targeted ads. "Even if you don't use Gmail, if you send email to someone who does, Google goes through those emails to generate advertising revenue too," Microsoft warns in material sent to reporters. Oh, and Microsoft points out that six class-action lawsuits have been filed against Google over this issue, and asks people to sign a petition "to tell Google to stop going through your personal email messages."

The latest ComScore search data for the U.S. reveals that core search market share for Google sites rose to 67% in January from 66.7% in December, a 0.3% increase. Microsoft sites also saw an increase, albeit a smaller one, in search market share — to 16.5%  from 16.3% , a 0.2% increase. Yahoo, AOL and Ask all experienced minor drops.

In other words, if Microsoft really wants to boost its search presence, it may want to consider acquiring the also-rans. Its frontal assault on Google appears to be doomed to failure.

Email usage is harder to track for a number of reasons, but KQED quotes Microsoft official Stefan Weitz acknowledging that the campaign hasn't exactly unseated Gmail. "It's a habit," Weitz said, apparently referring to Gmail. "It's like smoking. It's hard to get folks to stop doing it."

Here's another look at the late, unlamented first phase of Microsoft's Scroogled campaign:

Frustrated and bitter that laws like SOPA and PIPA have yet to get pushed through Congress without those pesky constituents objecting to turning the U.S. government into muscle for entertainment industry, the Recording Industry Association of America (RIAA) is taking out its anger on Google. The music industry lobbying group is accusing the search engine giant of failing to effectively demote search results that lead people to those nasty little download sites.

In a blog post on the RIAA site yesterday, Steven M. Marks, EVP & General Counsel, RIAA made it clear that the music copyright association thinks that Google, despite making some headway, remains a day late and a dollar short.

"We recognize and appreciate that Google has undertaken some positive steps to address links to illegal music on its network," said Steven M. Marks, the RIAA's executive vice president and general counsel. "Unfortunately, our initial analysis concludes that so far Google's pledge six months ago to demote pirate sites remains unfulfilled. Searches for popular music continue to yield results that emphasize illegal sites at the expense of legitimate services, which are often relegated to later pages. And Google's auto-complete function continues to lead users to many of those same illicit sites."

(This isn't the first time: see also RIAA Slams Google's Anti-Piracy Efforts, Demands Even More Unreasonable Measures.)

Testing The Claims

I wanted to see if the RIAA might be overstating its concerns, something that they've been known to do before. So I performed a little one-man experiment, using the song "Some Nights" by Fun. as the guinea pig. Your mileage may vary, of course, but my quick-and-dirty test revealed that the RIAA may have some valid claims.

A search for "Fun. album" returned a first, second, and third page of results that were absent of any results that would seem to contain illegal downloads, with the bottom of the third page containing three DMCA takedown notices that point to the Electronic Frontier Foundation's ChillingEffects.org for more information.

But down in the "Searches related to" section of all of the results pages, "fun. some nights download" was among the listings, and a click-through pulled in the plenty of links to aggregate MP3 download sites, mixed with a few legitimate sites, like iTunes (#3), Amazon (#7) and the official video on YouTube (#8).

As for the RIAA's claim that Google's AutoComplete will suggest search terms that could lead users to sites containing unlicensed copies of songs, I found this was indeed true. By the time I typed "fun. some", Google had filled in four results:

fun. some nights
fun. some nights lyrics
fun. some nights meaning
fun. some nights mp3

On a whim, I turned on SafeSearch to see if that would make a difference. Results did differ on some search results, such as "fun. some nights download", where legitimate sites (like the Wikipedia entry for the album) were moved up slightly on the first page of results, but the sketchy download sites were still in full-glory display.

I should also note that the RIAA did not take Microsoft's Bing service to task, even thought the same experiment on Bing yielded very similar results, even in the auto-complete results. Type in "fun. some" on the Bing home page and you get these helpful suggestions:

fun. some nights lyrics
fun. some nights
fun. some nights meaning
fun. some nights torrent
fun. some nights video
fun. some nights album download
fun. some nights mp3
fun. some nights review

Search Engines As Police?

Based on these (admittedly quick) search tests, it seems like the RIAA has a point, and Google is failing to block apparent pirate sites on its search results, and its demotion policy announced in August 2012 isn't really working all that well, either.

But let's be clear: Google has said all along it wasn't going to block site results from any site unless it receives a specific copyright removal request from the rights owner.

"Only copyright holders know if something is authorized, and only courts can decide if a copyright has been infringed; Google cannot determine whether a particular webpage does or does not violate copyright law. So while this new signal will influence the ranking of some search results, we won't be removing any pages from search results unless we receive a valid copyright removal notice from the rights owner," senior vice president of engineering Amit Singhal wrote back in August.

At the time, what Google said it would do was add a new signal to how it ranks search results.

"Starting next week, we will begin taking into account a new signal in our rankings: the number of valid copyright removal notices we receive for any given site. Sites with high numbers of removal notices may appear lower in our results," Singhal stated.

The RIAA is contending in its statement this week that Google has failed to live up to that promise.

One has to wonder, though, if trying to keep up with the sheer number of sites that provide access to unlicensed media content is comparable to spitting on a forest fire. If the signal for page ranking depends in some way on number of takedown attempts, perhaps the RIAA and other rights holders are not sending enough signals. Or maybe these sites know who to game other ranking signals to boost their status on Google and Bing search results.

It is very easy to point fingers at Google and Bing and accuse them of not doing enough to keep people away from pirated media. If you forget, of course, that this not their job.

Complaining about the auto-completing results would seem to be a more valid concern, until you remember that there could be legitimate results for "download X."

The RIAA wants to protect the rights of its artists and producers, a valid concern. But it is not clear at all that Google, Microsoft and the other search engines should be relied upon as key allies in the recording industry's ongoing quest to stomp piracy. Search engines' missions are to provide data, not analyze that data for legality.

Despite what they're asking for here, I suspect even the most vehement anti-piracy activists would not care for some of the implications of a world where search engines were to undertake that goal.

How much would you pay to be the default search engine in the iOS mobile operating system?

According to a report at TechCrunch on Tuesday, the magic figure that Google will be paying Apple for the privilege could be a whopping $1 billion dollars.

That's because Google and Apple, while enemies in the mobile arena, also need each other to survive. Apple's valuable iOS users - millions and millions of eyeballs that happen to be wealthier than most - are coveted by Google, and Google's massive search data is needed by Apple for those same users. (I could also add that Apple desperately needs Google's mapping data, but that's not what this story's about, and that's just rubbing salt on the Apple Maps wound.)

Apparently, even as Google is trying to kick Apple's butt with Android, Apple gets 75 cents for every dollar Google makes on iOS from advertising and data collection. Which puts the iOS price tag somewhere around that billion-dollar mark, according to the Morgan Stanley report cited in the article.

A billion dollars in revenue is nothing to sneeze at, and on the surface it seems like Apple is being silly with this continued effort to separate itself from the ever-pervasive realm of Google's services.

What it may be thinking is simple math: if Apple can get $1 billion just as a cut of someone else's business, imagine what it could make if it had full control of that business? Thus, we have efforts like Apple Maps.

Clumsy as Apple's efforts might be, the fact that there's a lot of money to be made in the realm of mobile data and advertising means such efforts may ultimately be worth the birthing pains. With its rich and loyal user base, Apple may be the eventual winner - if it can ultimately remove Google and it's billion-dollar ad payouts as the middle man.

Image courtesy of Shutterstock.

Get ready for the social explosion. Facebook Graph Search has begun to roll out to earlybirds who opted into the beta of the new social search bar, which was announced last week at the company's Menlo Park HQ. If you haven't seen the feature yet, get ready for an onslaught of experimental searches elbowing their way into your News Feed as your Facebook friends and frenemies discover it and tinker around - likely to your horror or amusement.

The feature will grow into Facebook's so-called "third pillar", alongside Timeline and News Feed. As Zuckerberg drove home during its debut press event, Graph Search is a bit rough around the edges for now. Facebook clearly has big plans to build out the social search engine, which remains in beta for the time being. 

If you don't count yourself among the afraid-of-change sort that refused to embrace Timeline until roughly a year after its launch, you can opt in now and give it a go. Sign up now on the Graph Search beta page to get in line or check out the new viral Tumblr, "Actual Facebook Graph Searches", themed around the awful, bizarre or otherwise zany results you can dredge up with the new intra-network search tool. 

Got any good ones? Don't keep 'em to yourself.

Microsoft's Bing search engine will ride the coattails of Facebook's Graph Search for years to come, gaining knowledge and possibly ad revenue as Facebook expands Graph Search's capabilities.

That can only be a good thing for Microsoft, right? Maybe not, this could be a deal with the devil, with Bing ending up digging its own grave.

Facebook rolled out Graph Search at an event at its new Menlo Park, Calif. headquarters on Tuesday. The natural-language search engine  can, for example, search for "music that people who like Mitt Romney like," or "photos of my friends in 1989." Although it's limited to four topics at the moment - people, places, interests and photos - the queries that users can run against the millions of photos and connections stored among Facebook's billion users are powerful, powerful tools for users and brands alike.

But Graph Search isn't perfect. Mobile capabilities aren't yet in place, and Instagram, multiple languages and addressing the entirety of the Facebook Social Graph are features that will keep Facebook's development teams busy for years to come, chief executive Mark Zuckerberg said.

Bing Rides To The Rescue?

That's where Microsoft's Bing comes in. For every search that Facebook can't handle, Bing will step in. When you launch Graph Search (and just a few hundred to a thousand people have it at the moment) Facebook suggests results to the query as you type, Google style. Although Facebook subtly teaches you what types of searches you can make against it, it will defer to Bing in certain situations: "restaurants in San Francisco," for instance. Facebook assumes that you'll want recommendations from friends, but can't be certain.

In the latter case, a Bing results page will appear, surrounded by the Facebook framework. If you need to click further, it will open up a separate tab.

Who Gets Paid?

What this means is that for a certain percentage of searches by a certain percentage of users, Facebook will be referring users to Bing. Whatever the actual percentages, for a site with a billion users and counting, that number should be significant. What isn't clear, however, is which site will receive any associated advertising revenue; Bill Hankes, director of Bing PR at Microsoft, declined to comment. (Facebook, of course, is profitable; Microsoft's online division has traditionally lost money.)

In search, context is key. And where Microsoft benefits, according to Hankes, is through a better understanding of the intent behind the search. "Where Bing social searches will be aided is actually in the Web searches conducted from the Facebook context, because we can infer intent, and that intent is different," Hankes said. "So let's talk about mobile. If you conduct a search on mobile, your intent may be very different than if you conducted it on your PC. Because your mobile device... where you are, and what time of day it is, and if you conduct a search for "happy hour" or "Mexican," you're going to get a very different result than if you search on your PC.

"Facebook posseses an entirely different corpus of information than what is on the Web, and when you're in Facebook, you're doing different types of things," Hankes added. "You're doing different types of searches.  When somebody asks for "restaurants in San Mateo," the intent may be different than if you were doing a general-purpose search on Google or on Bing. It sounds nuanced, but it's one way we can improve the algorithm."

What Happens Later?

It's not hard to believe that as Facebook expands both its user base as well as Graph Search, the numbers of people exposed to Bing will grow. Even a tiny fraction of Facebook's user base should boost Microsoft's market share, but Hankes was noncommittal. "It's early days, and we're going to see how it goes," he said. 

But eventually, we could see a fragmenting of the search market. According to Zuckerberg, Graph Search helps give users the tools to "map out the Social Graph." Users begin to have access to data that was previously reserved for large corporations. (At this point, Facebook doesn't have an API - application programming interface - in the works to let developers hook into Graph Search, which means that those users and brands will be on an equal footing.) 

Bing - like Google - has begun to try and provide answers to questions, rather than lists of links. But that's the same territory Zuckerberg and Co. have staked out, too. How one approaches the question, however, differs: Microsoft and Google seem to be fighting over the Web of facts, while Facebook has marched in to claim the Web of recommendations. (Google, of course, is fighting the war on both fronts, with its basic search engine and Google+ social network.)

As the years pass, however, it's fair to believe that Google will maintain its dominance in overall search, while Facebook will continue to control its own data. And as Graph Search improves, there's likely to be less of a need to supplement it with Web search.

What Does Bing Own?

Where does that leave Microsoft and Bing? Most likely, Bing will be left outside of Facebook's walls, looking in, while it fights Google for the rest of the Web.

Not surprisingly  Hankes doesn't see it that way. "To assume that, you have to assume people's behavior will be the same as it was yesterday," he said. Instead, users are increasingly turning to mobile, to tablets, and to multi-purpose gaming consoles in the living room: new locations and modes of search that Microsoft is itself trying to define.

Within the first two categories, Microsoft would seem to be well behind Apple and Google, which control the phone and tablet worlds. As for the third, at least within the United States, the Xbox 360 has outsold the Nintendo Wii and PlayStation 3 for more than a year.

But of you accept that Facebook's socially driven Graph Search is valuable, than Microsoft's control of living-room, big-screen content means something too. So Bing could end up a nomad of sorts: migrating between Facebook, the Web and the living room.

The question is whether that nomadic existence is viable for Bing, or if it leaves Microsoft's search efforts with the crumbs that search's true owners leave behind. 

If you've ever run the virtual gauntlet of online dating, you're well aware that most people are creepy. But the stigma of online dating is fast dissolving - and starting with Graph Search, Facebook is right there at the doorbell, tie straightened and bouquet of flowers in-hand.

On Tuesday at a press event at its Menlo Park, California, headquarters, Facebook took the wraps off of its newest feature, a search engine for people, places, photos and interests within a user's web of Facebook friends. During the presentation, Tom Stocky, Facebook Product Director, brought up one big use case that caught many off guard: online dating. "For single people, friends of friends tend to be a good start," Stocky said.

With Graph Search, you can query the social network for phrases like "friends of my friends who are female, like "Homeland" and are single" and get a custom-built singles search culled right from your own social graph. And since Facebook commands a pool of data as deep as it is wide, it has a lot of dirt on its one billion users - from the stuff they Like to the people they hang out with to their core demographic information like location, age, gender and sexual orientation.

If Mark Zuckerberg were to take you out for a night on the town, he'd even know to order a bottle of that local pinot you love. (Graph Search, for all its other implications, is also the perfect pick-up line generator.)

A Really, Really Big Phonebook

Seemingly content with the fact that just about everyone who might consider signing up for an account is already a member, Facebook now wants to fuel more social discovery - connecting people online first, so they can connect offline. The dating use case is obvious, but Graph Search could also prove handy as a recruitment and job hunting tool, Facebook noted.

"Mark says we're going back to our roots," a Facebook spokesperson told me following Tuesday's announcement. "In the beginning... before we had News Feed, [Facebook] was like a directory service - you had to use it to find people. With News Feed it became way more about picking up with the people you already knew. And now we're hoping that one of the things about [Graph Search], being the third pillar, is that people will start to find ways to make new connections, whether that is friends, people who share the same interests, dating, recruiting - whatever."

How Online Dating Sites Work Now

Dating sites like OKCupid, eHarmony and Match.com obsessively tinker with their romance recipes, but Facebook has a huge advantage: it already knows your friends. By eschewing the layered anonymity that's the cornerstone of the online dating community, the social giant might have just tapped into key the secret ingredient.

Sophisticated algorithms might not stand a chance against the insight and connections of the people we choose to surround ourselves with in real life - but they can try. Match.com pairs you up with so-called "behavior twins" behind the scenes and then nudges you toward the people that they've interacted with. Circl.es, another Web matchmaker, hooks into Facebook to use users' real names and interests, but it blocks out your existing Facebook friends so you get only fresh faces - and not embarrassing encounters with exes.

Hooking Up In The Early Days Of Facebook

I signed up for Facebook on May 21, 2004. The social network, then still known as The Facebook, had just expanded to New York University, the ninth school it added (Harvard being the first). Before I arrived at college, I did something kind of creepy on the nascent social network. Using Facebook's search tool, I hunted for NYU students who I might want to get to know - namely other female students who identified as "interested in women."

From my outpost in suburban Texas, I Friended some of them -others I just sort of observed at a distance. Facebook profiles were pretty open back then, but they didn't have much info. Still, that sexual orientation tidbit in the Basic Information section was a telling one, and I wasn't the only one using it to connect to people I'd later befriend (and later, er, "date"... in at least one instance).

Because the Web makes winnowing the human population down to people who might consider sleeping with you in the first place a lot easier for those of us seeking same-sex partners, the LGBT population tends to be over-represented in online dating stats. But while sexual orientation may not have been salient to my straight classmates in the same way - plenty of them used Facebook to date too, back then.

Before it was a running joke, we're talking roughly eight years before Poke made its second coming as an app, meeting someone cute at a party and poking them the next day was just what you did (no, really, it was!). These days, Facebook isn't really a social discovery engine in the same way. People still use it to hunt for potential partners - in fact, i've been asked on more than one date via Facebook in recent months by a friend of a Facebook friend - but usually that's only after some sort of real-life encounter.

Online Dating Is Broken

Online dating is a booming business - to the tune of $2.1 billion in early 2012 - but it's also a broken one. According to the co-author of a meta-study of online dating outcomes, all of those mysterious romance recipes might just be leading us on. Eli Finkel, Associate Professor of Social Psychology at Northwestern University, summarizes the findings in "The Dubious Science Of Online Dating," a takedown published in The New York Times:

"[Online dating] sites tend to emphasize similarity on psychological variables like personality (e.g., matching extroverts with extroverts and introverts with introverts) and attitudes (e.g., matching people who prefer Judd Apatow’s movies to Woody Allen’s with people who feel the same way). The problem with this approach is that such forms of similarity between two partners generally don’t predict the success of their relationship."

Graph Search Could Evolve

Will Graph Search evolve into something better? Facebook admits that it doesn't really know how users will end up using any new product as it rolls out - and that includes Graph Search.

In an ancient version of its FAQ, the company explained that this ambiguity was the exact idea behind everyone's favorite retired-then-resurrected Facebook feature. "When we created the poke, we thought it would be cool to have a feature without any specific purpose," said Zuckerberg. "People interpret the poke in many different ways, and we encourage you to come up with your own meanings." 

Of course, Graph Search has many implications beyond the world of online dating, but it's a pointed example of how Facebook's third "pillar" casts a wide net. And true to recent form, the company intends to trawl up just about all the competition while it's at it. Will users burned out on established online dating sites sculpt Graph Search into the next Match.com? 

Could be, and the one social network to rule them all apparently doesn't mind casting a line in that direction.

The big Facebook news tech blogs were all freaking out about turned out to be just another shot in the platform wars.

There's nothing exciting about Facebook's Graph Search. It's just another way to lock in free users to a mediocre, incomplete service, just like Google wants to do with Google+. Until there's a personalized, natural-language search box that can search whatever and wherever we want, I don't think anything else matters.

The way I see it, there are three companies offering rudimentary artificial intelligence to consumers at a grand scale: Google, Apple, and now Facebook. (Sorry, Microsoft, you're a distant fourth.) All of these companies want to provide you convenient answers when you ask them questions in your own language. Hundreds of millions of people use more than one of these. Many use all three. But they can only be used one at a time. While you're using any one of them, you're not using the others. Therein lies the problem.

Asking questions and getting answers is one of the most important things we want to do with computers. It's the thing that impresses us most. Our sense of living in the future is bolstered by this very feature. When it works, we're amazed. When it fails, we feel like our magical gadgets have let us down.

And right now, each contender has its own box we have to use to ask different questions of different data sets. We have to remember the strengths and weaknesses of each box. If our question fails in one box, we have to go to another box and start over again. Facebook Graph Search is just another box.

The next interesting kind of search is a single, personal, natural-language interface to all of the data sources of our choosing. Wake me when we get that.

Facebook Is Too Creepy

Facebook's Graph Search will tell you things about your friends and "friends" and "likes" and your "friends' likes" on Facebook. It's Facebook's first bit of AI stalking enhancement. It lets you ask natural-language queries about your social sphere, like "single people near me who work at Facebook," and it displays all the answers it can find. This makes privacy more important than ever. If you don't take care of your privacy settings on your posts and profile, Facebook's AI search is going to dig that stuff up.

So everyone who's smart will limit the amount of information that can be found about them on Facebook, since now it's so easy to find. People will realize that they should share only things they want people to find, and that's a good thing. But it also means searching on Facebook will show you only a narrow band of things.

If the answer is not on Facebook, Graph Search searches the Web with Bing. Oh, great. What if your search and browsing history, your email, your saved places, and (for nostalgia's sake) your RSS feeds are all on Google? Well, you'll have to go to another site or app and ask a different AI for that.

Google Is Too Geeky

Google's mission has always been to "organize the world's information." Lately, the end goal of that mission has become clear. Google gathers information about absolutely everything, and relates that information to us personally, because it wants to build the Star Trek computer that can answer all our questions. It even has Ray Kurzweil, one of the most starry-eyed proponents of powerful AI, working on the problem.

The issue is that Google doesn't know us very well. It knows a great deal about our searches, our email, oftentimes our location and all of that, but it doesn't have a clear enough picture of our relationships to draw many conclusions about them. It's trying to get us to point them out with Google+, but Facebook is still the place where the lion's share of online friendship happens. It's where we indicate who and what we really care about.

Because Facebook and Google are in competition for our attention, they don't work together. Both companies offer cheeky bromides about being willing to work together, but that just means they each want the other to compromise its business and hand over the valuable data. That's not going to happen under today's economics. So for different kinds of questions, we're stuck with different search boxes.

Apple Is Too Paranoid

The biggest question mark in consumer AI is Apple's Siri. Apple seems to realize, at least partially, that the AI has to have many, many data sources in order to have good answers. The problem is that it won't partner with Google, which is the best data source in most businesses, because they're competitors at the OS level.

There are finally great iOS apps for the core Google features, but they don't talk to Siri. If you ask Siri for directions, she uses Apple Maps, which are not as good. If you ask her for a list of restaurants, she uses Yelp, which might be fine, but not if you're a Google Maps or Foursquare user (unless Apple buys Foursquare).

The problem with Siri — and iOS in general, really — is that you can't switch the data sources. You can't choose to use Google Maps with Apple's AI, and if you choose to use Google's AI on Apple's device, it's a hack. It's not integrated with the system.

Now, Facebook is integrated into Apple's operating systems, so it would be interesting to see Siri use Facebook's Graph Search. Now we're getting somewhere. But Google is still the best AI for a great many kinds of questions, so we'll still need two boxes.

Search That Works For Us

We won't really be in the future until it works like this: We've got our own AI assistants who know us intimately but protect our information. We can ask them questions, and they will use every data source to find the right answer for us, not merely the best one available on a proprietary service.

Until then, all these companies will keep building their own boxes, and we'll have to run around between them pretending it's convenient.

Lead image courtesy of Shutterstock. Facebook photo by Taylor Hatmaker.

Now we know what Facebook's mysterious announcement is. Today at its Menlo Park, California, headquarters, Facebook took the wraps off Graph Search, an instant smart search that sifts through the company's massive social graph to return relevant results.

Facebook's Third Pillar

According to CEO Mark Zuckerberg, News Feed and Timeline are the two extant pillars of Facebook, and Graph Search is about to be the third leg of the stool. But Graph Search is no Google - if you ask Zuck, anyway.

"Graph search is not Web search. We're not indexing the Web here, we're indexing our map of the graph," Zuckerberg said. "Graph search is very different. For example, you need to be able to easily ask a query like "Who are my friends who live in San Francisco?"

Zuckerberg noted that rather than directing you to a link or page where users can find the answers to specific queries, Graph Search will deliver the answer, functioning like a social-powered Google Instant Search - but all within Facebook of course. 

To interact with Graph Search, users can type a query in natural language into Facebook's meta search bar. According to Tom Stocky, Director of Facebook Product Development, "This isn't keyword search... Graph search is structure. It understands search made up of these simple phrases."

How Graph Search Works

Type "Photos of my friends taken in Paris, France" and, assuming your photos are geotagged, you'll get exactly that. Type "Friends of my friends who are single, male San Francisco residents who are from India" and you'll get a hyper-specific subset of potential romantic partners, according to Stocky. Better yet, type "Photos of my friends before 1990" and you'll get a wealth of social blackmail.  And when the social graph doesn't have the answer? A partnership between Facebook and Bing will deliver Web relevant web results to fill in the gaps.

Yes, that is Facebook COO Sheryl Sandberg as a baby.

According to Zuckerberg, Graph Search is a beta product in its very early stages. "It's going to take years to index the whole map of the graph. We're going to start rolling it out very slowly." Facebook Graph Search will be available in a limited beta beginning today at Facebook.com/graphsearch. 

"It's a really different product from anything else out there. This is who we are. We love building things like this."

Microsoft had a pretty lousy year in 2012, putting out a string of big products – Windows 8, Windows Phone 8 and the Surface tablet – that all turned out to be be disappointing.

But those pale in comparison to what may be the biggest disappointment in Microsoft’s history — its failure to convince antitrust regulators to take action against Google.

After a 19-month investigation and despite much prodding from Microsoft, the Federal Trade Commission has reached a settlement with Google that basically amounts to a slap on the wrist.

This is a crushing blow to Microsoft, which has spent millions of dollars on lobbyists and phony grassroots groups over the past several years hoping to land Google in hot water.

Indeed, Microsoft’s obsession with Google doesn’t just border on crazy. It is crazy, and not just a little tiny bit crazy but full-blown, bunny-boiling, Ahab-versus-the-whale nutso.

The Shadow War

For years Microsoft has devoted massive resources and energy to waging a sneaky shadow war against Google, fielding an army of lobbyists and front groups that exist almost completely to spread anti-Google propaganda, including ICOMP.org, the Association for Competitive Technology, FairSearch and SafeGov. 

They call themselves “industry groups,” and they have lots of members, but they’re basically Microsoft fronts devoted to hating on Google. Take a glance through ICOMP’s “newsroom” and “Voices” section and white papers and “ICOMP in the News” section. It’s pretty much all Google, all the time — and all negative. It’s a whole website devoted to bashing Google, and frankly it’s kind of incredible, in a twisted way.

FairSearch does much the same. Its site is loaded with Google bashing. FairSearch's spokesman is Ben Hammer, who works for the Glover Park Group, a lobbying and PR firm in Washington, D.C., that used to lobby for Microsoft. Before taking on his role at FairSearch, Hammer worked on the Microsoft account for Glover Park.

FairSearch also hired Patrick Lynch, the former attorney general in Rhode Island, as a consultant —  and then Lynch began lobbying states to investigate Google and publishing op-ed pieces in newspapers bashing Google for "abusing its power," among other things.

Microsoft even hired Randall Long, a former FTC attorney who had worked on investigations into Google during his time at the commission. A watchdog group protested and called the hire "suspicious."

In 2009 Microsoft spent millions of dollars to kill a proposed deal between Google and Yahoo. Read this Wired story about what Microsoft did, and you’ll be flabbergasted.

That battle was fought by an internal SWAT team that is devoted to hobbling Google. The team is led by a lawyer, John Kelly. (See his LinkedIn bio here.) Also on the team is Kate O’Sullivan, a former exec at Burson-Marsteller, a top Microsoft PR agency that has a hand in running ICOMP.

The SafeGov 'Grassroots' Campaign

More recently, as Google started trying to sell Google Apps to government agencies - competing against Microsoft Office by offering much lower prices - Microsoft helped create a new group called SafeGov.org, another so-called “grassroots” organization whose mission supposedly is to discuss issues around government computing policies, but really ends up being just another front for bashing Google.

SafeGov’s "experts" include Doug Miller of Milltech Consulting, who used to work at Microsoft doing “competitive strategy,” and earlier this year disclosed a consulting relationship with Microsoft. There’s also Jeff Gould of Peerstone Research, another SafeGov expert who has disclosed his consulting connection to Microsoft in the Washington Post and on a site called The Open Enterprise, but not in his SafeGov bio. Finally there’s Bryan Cunningham of Cunningham Partners who co-founded a company called Polaris Consulting whose lobbying clients included Microsoft —  though again this connection is not mentioned in his SafeGov bio.

SafeGov’s other experts include four people from the Chertoff Group — a lobbying organization run by former Secretary of Homeland Security Michael Chertoff. They've done their part to bash Google. Chertoff criticized Google in the Wall Street Journal in July 2012. Richard Falkenrath, a principal at Chertoff and a SafeGov expert, criticized Google in the Financial Times in February 2012. Yet another SafeGov expert, Andrew Weis of Civitas, warned about Google and privacy in the Chronicle of Higher Education in February 2012.

All For Nothing

And now all Microsoft's plotting and scheming has come to nothing. Or almost nothing.

The FTC has settled with Google on the allegations of search engine manipulation. As FTC Chairman Jon Liebowitz has stated, the investigation is closed and the FTC is satisfied that Google was not acting in anti-competitive fashion. "On balance, we didn't believe that the evidence supported an FTC challenge to this aspect of Google's business under American law," Leibowitz said in a press conference on Thursday. There won't be any protracted (and distracting) legal battle. There won't be any embarrassing depositions. Google will just make a few little changes that it has already agreed to and the Feds will back off.

Of course FairSearch is howling about the settlement and claiming the FTC "failed in its mission to protect American consumers." Microsoft is still holding out hope that European regulators will do something to hobble Google. And that may happen.

But here in the States, the whole campaign is starting to look not just evil but also ridiculous and even pathetic, a failed crusade that smacks of revenge and failure and sad psychodrama.

Why Not Just Make Better Products?

Microsoft has spent the past 10 years missing out on every big new trend — search, social, mobile. Instead of looking inward and trying to fix its own problems, Microsoft has become ever more obsessed with Google.

The crusade extends to PR as well. Last year Microsoft hired Mark Penn, the pollster and PR guy who led Hillary Clinton’s 2008 train wreck of a campaign.

Penn’s mission is to create smear campaigns about Google. One of his big brainstorms was the “Scroogled” campaign which tried to persuade people that Google’s search results are tainted by advertising. Microsoft ran full-page ads in newspapers and spots on Monday Night Football. It even created a site, scroogled.com, and urged people to complain about Google on Facebook.

Then came the “Droid Rage” ploy, where Microsoft used its Windows Phone account on Twitter to ask Android users to send in Android malware stories to win a prize. The campaign backfired when Android fans used the hashtag to mock Microsoft.

Going negative might work in politics, but when you're selling products it's probably wiser to tout the virtues of your own product. The risk Microsoft is taking is that by howling about Google, Microsoft starts to look like a company that can no longer compete, a desperate dinosaur that has toppled into the tar pit.

Worse, Microsoft starts to seem a little unhinged. Nevertheless, don’t expect this freak show to end anytime soon. Supposedly Penn has been going around Washington trying to recruit consultants, telling them that Microsoft has armed him with a $50 million budget to go after Google.

Imagine the kind of cool product that a team of great engineers could dream up with $50 million. Instead, we’ll get more negative ads and ineffective lobbying, while the Surface continues to post disappointing sales figures, and Windows Phone languishes in the cellar of the smartphone market, and while Android continues to gain market share in smartphones and tablets and the FTC can't find a way, despite all of Microsoft's finagling and pleading, to make a case against Google. 

Keep up the great work, Microsoft.

See also: Google Escapes Unscathed From From FTC Settlement and Microsoft's Fight Against Google Continues With YouTube App Complaint.

Image by ReadWrite.

Internet porn surfers everywhere were crying foul last week over Google's decision to change its "SafeSearch" feature to exclude adult content from searches.

Anyone that uses Google to find adult images will likely notice something different when they type in their dirty little specifics: less skin, more filtering. The change isn't that huge, you just have to be a little more specific in what you're looking for. The update to the feature changed the SafeSearch options from "Strict", "Moderate" and "No Filtering" to whether or not to select "Filter explicit content," which seems to correlate to some combination of the old Moderate and Strict settings and filters out most anything that anyone could find objectionable.

Except medical content apparently. A search (research purposes only) for "penis" still yielded results of naked male organs. However, most of them where injured, sickly or replicas made from baked goods - not necessarily sexy. Parents of young children, or those who are really nit-picky about porn in their household can also lock SafeSearch so that no explicit content can ever be seen, even accidentally.

Some Folks Missing Stuff

This seemingly innocent change in the feature has made quite a few reddit users and bloggers pretty upset. The big deal seems to be that there was no mention that this change was coming before it actually happened. Google made the change only to the U.S. site (U.K. site still has the old version of SafeSearch) and went on about its day.

While there still hasn't been an official announcement, ZDNet pointed out that "This default setting, however, does appear to be more reminiscent of the 'Moderate' SafeSearch setting rather than a total overhaul to the system."

You Can Still Find Smut

If this change bugs you beyond belief and don't want to get specific with your Google search engine ICM has some advice. In a press statement, ICM Registry advises searchers to include "XXX" to their inquiry, where the results are expected to be the exact amount of explicit your were expecting to get. The release also includes a plug for its newly launched site, adding: "Of course age appropriate Internet users can best and most quickly find adult content through ICM’s www.search.xxx dedicated porn search engine too, without using the “XXX” trick."

As long as Google doesn't limit our ability to search for the latest "Tard the Grumpy Cat" meme or videos of dogs playing with Christmas ornaments, a few changes to the more-explicit content searches will probably be something most folks can get over.

Besides, other search engines, including Microsoft's Bing, still make it plenty easy to search for porn.

The ReadWrite DeathWatch is known for serving up plenty of doom, gloom and grumpiness. But for the Holiday Season, we're going to take a slightly different tack, and highlight companies and technologies that Cheated Death - that might have died, but didn't.

First up for the Cheating DeathWatch is Microsoft, which somehow managed to stay relevant even as the market and the media have pivoted away from the desktop computer arena that made the company rich and famous.

Where Microsoft Was

After buying DOS to enter the operating system market in 1981, Microsoft knocked out IBM and held off Apple to become the undisputed champion of the computer operating system. Along the way, Microsoft imitated and intimidated as much as it innovated, gaining a reputation as a bully - bulldozing its way to success with marketing, money, and industry leverage.

Whatever its tactics, Microsoft was successful. Windows killed IBM's OS/2, Office killed WordPerfect and Lotus 1-2-3, and Internet Explorer demolished Netscape. Microsoft was king of the tech world, closing out the millennium at its highest valuation ever.

And then everything fell apart.

As first RIM, then Apple and Google brought computer functionality to mobile devices (from smartphones to tablets), they shifted the center of the tech world to mobile devices and away from Microsoft's desktop stronghold - and Microsoft couldn't answer. In the lucrative search market, Microsoft's bid to buy Yahoo failed, and its Bing search engine shows no signs of dethroning Google. In the enterprise, Linux was becoming a viable alternative to Microsoft's products. Only Microsoft's XBox was able to become a leader in a new market. Pundits were increasingly ready to write off Bill Gates' creation as a dinosaur, not able to keep up with swifter, smarter competitors. The DeathWatch was on.

Where Microsoft Is Now

But Microsoft didn't give up. Love it or hate it, Windows 8 is everywhere, from PCs to phones to tablets, and people are actually taking it seriously. It's early, and Microsoft has a lot of catching up to do, but even tech insiders can now bring home a Windows-powered Surface tablet or Lumia 920 phone and hold their heads high.

Even the gaming market is looking up. Halo 4 sales hit $220 million on opening day, with killer reviews. Microsoft's E3 preview of SmartGlass displayed a well-planned move toward cross-platform gaming, and the rumored Xbox Surface could back up its mobile gaming ambitions.

Microsoft is refreshing its whole lineup. Office is finally moving to the cloud, IE 10 is completely rebuilt, and bing has a new consumer focus, thanks to deeper relationships with Facebook, Amazon and Yahoo. There's even the first new logo since 1987.

Can Microsoft pull off this reinvention without a hitch? Not a chance. There's evidence that the tide may already be slowing. Still Microsoft has gone from an aging granddad to a revitalized contender in just about a year. Depending on who you ask, it might even be kind of cool.

How Microsoft Got Here

From one angle, Microsoft came back by doing what it always does - rolling in late and large. It finally understood that mobile, touchscreens and data portability were big, the same way it realized the Web would be huge back in 1997 - years after everyone else. It the media blitz for Windows 95 or Internet Explorer 4 all over again - only with at least slightly better ads:

But Microsoft actually did something kind of bold. In its own words, it "shunned the incremental" for once. Windows 8 is a major departure from other operating systems - including Windows 7. The Surface may not be quite sure whether it's a tablet or a laptop, but it's a novel flagship for a hybrid OS that looks great in ads and challenges third-party manufacturers to do better.

Microsoft has also finally learned to adapt. It has dropped Windows prices to compete with Apple, addressed the Google Docs threat by moving Office to the cloud, and given non-PC devices a nod by expanding Windows to ARM processors.

Can Microsoft Stay On Top?

One of the biggest drivers of Microsoft's mind-shift was Steven Sinofsky, the President of Windows and Windows Live who left the company in November. While Microsoft's stock took a dip after the announcement, there is a of potential upside. Sinofsky made enemies at Microsoft, and some analysts think his absence might smooth collaboration in the future.

Sinofsky's replacement, Julie Larson-Green has been with the company since 1993, and arguably had as much influence on Office and Windows 7 and 8 as Sinofsky did. She's smart, knows the lay of the land, and has a lot of friends. The question, really, is whether she can use that political capital to push the company forward. Sinofsky could be brutish, but he got things done.

Larson-Green's first task is to pull more support and innovation from its hardware partners. Microsoft has always prospered by getting its partners to do much of the heavy lifting, but many of its biggest and most important relationships have well-earned spots on the ReadWrite DeathWatch. We're looking at you HP and Nokia.

Those challenges are very real, and Microsoft is not out of the woods yet. But make no mistake, Microsoft is still a player.

To see more ReadWrite DeathWatches, check out the ReadWrite DeathWatch Series, which collects them all, the most recent first.

On Wednesday, Microsoft began running a national ad campaign to warn users to avoid being “Scroogled” by Google search results that have been influenced by paid placement.

Specifically, Microsoft’s campaign attempts to portray Google as basing its Google Shopper search results on the amount of money each vendor has paid, rather than the accuracy of the results themselves. More practically, Microsoft claims that popular online retailer Amazon opted out of Google’s search, but that Bing users will be able to compare prices against the Web’s largest e-tailer. 

Who Is More Unbiased?

With the new campaign, Microsoft is taking the side of the angels, using semi-patriotic language reminiscent of the recent presidential election. 

“Today, Bing renews its commitment to the old rules – to honoring our side of the bargain with shoppers by delivering better, more objectively ranked search results,” Mike Nichols, a corporate vice president and chief marketing officer for Bing, said in a blog post. “We won’t let who pays us for ads or other services affect what you see in your search results. Search results are one thing; ads are another. We won’t switch to pay-to-rank to allow some shopping search results to appear higher than others. We don’t believe shoppers should risk paying more, simply because they started their search at Google.”

Hoist By Its Own Petard?

Now that the program is in place, Microsoft is positioning its own results as complete and unbiased, and a better alternative for shoppers this Holiday season. And the company is going full bore: not only has it launched a “Scroogled” Web site that invites shoppers to submit their examples of being “Scroogled,” but it has peppered the site with links to the Google founders’ letter, where Larry Page and Sergey Brin claimed, in the famous “Don’t Be Evil” paragraph, that their company’s search results “are unbiased and objective, and we do not accept payment for them or for inclusion or more frequent updating.”

Damningly, Microsoft is also using Google’s more recent words against it, citing the language used by Google in announcing an update to Google Shopping announced this spring:

“Ranking in Google Shopping, when the full transition is complete this fall, will be based on a combination of relevance and bid price - just like Product Listing Ads today,” Sameer Samat, vice president of product management for Google Shopping, wrote in a blog post. “This will give merchants greater control over where their products appear on Google Shopping.”

While Microsoft’s campaign must be seen as a classic example of FUD (Fear, Uncertainty, and Doubt) tactics it has honed over the decades to fight everything from Netscape to Linus, the company suggests that shoppers won’t be best served by the pay-to-play rankings. Like Amazon, merchants may choose not to pay to rank results preferentially for low-margin products, which could affect their supply, and specialty merchants might be excluded as well. Microsoft also called out Google’s search results, which claim to be defined by “relevance”.

If one clicks on a particular product, Google opens a separate page listing stores and prices, with a box that encompasses all of the results and defines them as “sponsored.” “Google is compensated by these merchants,” an informational blurb reads, but only if the user clicks on the “i.” But on the main results page, Google hides the pay-to-play results under the link, 'Why these results?" - a difference in language Google couldn't explain.

“While good deals and wide selection take a back seat on Google Shopping thanks to its policies, the way to look at the deception taking place is how Google hides its disclosure policy to customers behind a little link on the side of the page and offers a vague explanation of the factors that may influence ranking, including payment,” Stefan Weitz, the senior director at Bing, said in an emailed statement to ReadWrite.

For its part, Google wouldn't specifically address the pay-to-play charge, but characterized Google Shopper as a useful tool.

“Google Shopping makes it easier for shoppers to quickly find what they’re looking for, compare different products and connect with merchants to make a purchase," a Google respresentative said in an emailed statement. "With new 360-degree, interactive product images, social shopping lists and a fast growing inventory of more than a billion products worldwide, Google is a great resource for shoppers to find what they need, at great prices for their loved ones this holiday season.”

Does Pay-to-Play Affect Search Results?

The problem in comparing the two search engine results is that the results are rarely, if ever, clear-cut. In one test ReadWrite conducted (without input from Google) Microsoft’s results seemed less relevant. But on others, an observer could conclude that a vendor paid for placement - but might also believe that the search engine simply returned a more relevant result.

Take a search for “telescope,” for example. On Google, the first five results are for Celestron telescopes, then the Tasco Specialty 49TN. On Bing, the first two results are for Celestron products, then the Tasco 49TN, the BARSKA StarWatcher, and then two more Celestron scopes. The searches are different, but is one better than the other?

The answer might lie in comparing the searches that you (or an acknowledged expert) might make himself. When shopping for the “best smartphone” on Google, Google ranked the Sony Xperia S as the most relevant result - an odd choice, given that few U.S. reviewers had actually reviewed it, let alone given it top marks. (CNET gave it three stars, though PhoneArena’s review said that Sony “hit the nail on the head”). Google’s next choices were the Samsung Captivate Glide (ranked 3 out of 5 or 4 out of 5 in most reviews, and then the Samsung Galaxy S II, the older version of the well-reviewed Galaxy S III now on store shelves. Google did not place an Apple product in the first page (18 products) of results.

Bing, for its part, chose the Samsung Infuse 4G, released in May 2011, the superb Motorola Droid RAZR MAXX, and the Sony Ericsson Xperia arc as its top three choices. At fifth, it placed the Apple iPhone 4s. At sixth, however, Bing chose the T-Mobile MyTouch 3G, a phone that was released in 2009 and available from just one store. The T-Mobile G2, one of the very first Android smartphones, is seventh, one place above the Microsoft-powered Nokia Lumia 920. Wow.

(Interestingly, if one searches for "smartphone" - not "best smartphone" - on both sites, Bing's results bury the iPhone 4S farther down, bump up the Lumia, and include the original, much-despised BlackBerry Storm. Google places the Apple iPhone 4S and the iPhone 4 in the first three results.)

A more significant problem is the exclusion of Amazon in search results. This kind of thing happen all the time; search flight listings on Kayak.com and you’ll quickly find that the results exclude Southwest Airlines. But Kayak at least lists the flight times, even as it excludes the prices; on Google, you’ll need to remember to check Amazon to see if the company even sells the item. That’s a problem.

“So for this holiday season, we just want to make sure shoppers know that when searching for that perfect gift for Cousin Harry on Google Shopping, the results they are seeing are partially optimized to benefit Google’s revenue, not a shopper’s pocketbook,” Microsoft's Nichols wrote. “If this practice has you concerned, then please try Bing for an honest search.”

The conclusion? Yes, more Microsoft FUD. But this time it just might be justified.

Somewhere in Hollywood, the publicists for Kim Kardashian and her sisters are raising a toast. According to Microsoft's Bing search engine, the Kardashian sisters were easily among the most searched-for terms of 2012.

Monday night, Microsoft released the most popular search terms of the year, grouped into various categories, including several devoted to celebrities. Kim Kardashian was the most-searched-for person of the year, with her baby, her role as a reality-television star... and what else? putting her name on Bing users' fingertips. 

A Year-End Tradition

Lists of the most-searched-for terms have become a year-end staple, with Microsoft, Google, Amazon and others releasing their users' most-popular terms, celebrities, albums, books, and other categories. Although Microsoft holds a minority position in the U.S. search market (16%, in October, versus 66.9% for Google, according to comScore), it's likely that most if not all of Microsoft's most-common terms will appear in other lists.

Here's what we know about people using Bing this year.

Microsoft's top searched-for celebs of 2012:

1. Kim Kardashian
2. Justin Bieber
3. Miley Cyrus
4. Rihanna
5. Lindsay Lohan
6. Katy Perry
7. Selena Gomez
8. Jennifer Aniston
9. Nicki Minaj
10. Taylor Swift

The holidays will undoubtedly be a bit less cheery for Lady Gaga, Jennifer Lopez and Britney Speats, who plunged from the top 10. (The 2011 list included Bieber, Kim Kardashian, Jennifer Aniston, Lindsay Lohan, Lopez, Spears, Katy Perry, Megan Fox, Lady Gaga, and Miley Cyrus.)

Interestingly, even during a nasty, balloon-budgeted election year, celebs in 2012 were sought more often than the Presidential candidates. Mitt Romney won the Bing search "election," placing 43rd to Barack Obama's 46th place. Royals bowed to entertainers, too. Kate Middleton was No. 12, while sister Pippa dropped to the bottom of the list at number 75.

The Kardashian sisters also swept the reality-show category, with Kim, Khloe, and Kourtney topping the first three spots. And don't look now, but in the fifth spot, here comes little Honey Boo Boo, the redneck reality star who made most of America cringe.

Khloe and Kim, with partners Lamar Odom and Kanye West, ranked third and fourth, respectively, among the top celebrity couples. 

Tech & Sports

Top tech terms included the Apple iPhone 5, the iPad, the Samsung Galaxy S III/S3, the Amazon Kindle and iPad 3. Bing also ranked social media, where the only surprise was that MySpace finished third after Facebook and Twitter.

Payton Manning topped the list of sports stars, as his comeback with the Denver Broncos captivated fans. Tiger Woods and devout quarterback Tim Tebow finished second and third, respectively, followed by sexy French Open winner Maria Sharapova and Kobe Bryant. Even international soccer star Ronaldo made the list, in ninth place. It seems the Kardashian connection propelled Lamar Odom (No. 7) above LeBron James (No. 8).

The top Olympians, meanwhile, included Serena Williams, Lolo Jones, LeBron James, Michael Phelps, and gymnast McKayla Maroney, whose "not impressed" meme surely helped her ranking. 

Biebermania Subsides - A Bit

Justin Bieber fans were unable to propel their idol to the top of the Bing pops, but Bieber and the late Whitney Houston ranked first and second, respectively, among musicians, with Katy Perry, Selena Gomez, and Rihanna in the top five. (Bieber fans also launched an unsuccessful Twitter campaign to keep Bieber's song "Baby" the most-watched video ever on YouTube. Korean pop star Psy stole that crown, however, with his viral video hit "Gangnam Style".) 

Bing also revealed the top 10 songs of 2012, based on searches. (Look for these tunes to be 2012's top sellers, too.):

1. Psy's "Gangnam Style"
2. Carly Rae Jepsen's "Call Me Maybe"
3. Justin Bieber's "Boyfriend"
4. Gotye's "Someone I Used to Know"
5. Fun.'s "We Are Young"
6. One Direction's "What Makes You Beautiful"
7. Drake's "The Motto"
8. Maroon 5 featuring Wiz Khalifa: "Payphone"
9. The Wanted's "Glad You Came"
10. Katy Perry's "Part of Me"

Surprisingly, Bing did not list the top 10 most-searched for videos, although Microsoft did say that music videos dominated search - with Fun.'s viral hit, "We Are Young," coming in first. (That's probably because most, if not all, of the top videos resided on YouTube, owned by rival Google.) 

Bing also broke out other categories, such as Most Searched Fashion Designer or Label, Most Searched Destination and Most Searched Holiday Food. To see all the lists, check out Microsoft's blog post.

For some reason, though, Microsoft did not report the most popular search terms over all, but the company responsed to my request with these five:

1. Facebook
2. Insurance
3. Netflix
4. Twitter
5. Coupons

Kind of dull, actually...