I've written previously about how social search won't replace traditional search, how social relevancy rank can be used to deliver good results, and why the concept of social search is a return to a familiar state rather than something to fear. Today, I'll get more specific about the three flavors of social search that will improve user search experiences.

Collective Social Search

"Collective social search" is similar in concept to the wisdom of crowds, in that search is augmented by trends shared on a network (a la Twitter Trends) or results ranked against the real-time buzz of a group. Why might this be useful? Well, in some instances, we can't immediately find the information we're looking for; and pooled, aggregated data from the collective may point us to new avenues that expand our discovery process.

As of yet, no major search systems are doing this very well - and we don't know what type of interface would be optimal for sharing this information. The Cloudlet plugin inserts tag clouds (based on keywords) into search results; but tag clouds are known to be more of a distraction than a utility. BingTweets has been touted as such a resource, but it really only offers Twitter and Bing results on two separate pages. OneRiot shows only collective data from the real-time stream, although it may be integrated with Yahoo! results soon. And we are still waiting to see how Google and Bing integrate the Twitter firehose into their traditional search results - as opposed to merely including them as additional document-like resources.

Equally important will be understanding when collective social data should be shared with users: while performing the search or after? And for which types of searches?

My research on search strategies begins to address this question. Collective guidance may be useful when users are exploring a search space, possibly because the search domain is not familiar to them (i.e. they lack knowledge of how to drill down to an answer), or because they are passively exploring a problem. I find myself doing this all the time when I prepare recipes to cook. I want to browse recipes from many different sources before I decide what my own recipe will consist of. I don't have a specific recipe in mind (it's not an urgent, active request), and therefore I don't necessarily know when I've found what I'm looking for.

That said, it's hard to determine from keyword strings how active or passive a user's search is; i.e. it may be quite difficult to determine the type of search they're performing or how far along they are in their search process ("exploring" or "narrowing"?). Furthermore, the utility of collective social data for mainstream consumers will be limited, mainly because it doesn't come from trusted sources, unlike "friend-filtered social search" (see next section).

Friend-Filtered Social Search

Friend-filtered social search is approximately what Google is doing with its social search experiment: providing social data that your peers, friends of friends and wider "social circle" have shared. This data could appear alongside traditional search results (as with Google) or be exclusive results from within your peer network (as with TuneIn).

This is useful if your friends have shared relevant links, blog posts or tweets about a topic that you're searching for. If you were gathering ideas about, say, "the future of the desktop," you would see thought pieces, write-ups and links to projects from the main search algorithm, as well as stuff your friends are saying about applications they've encountered recently. If you trust your friends, they may serve as reliable filters, pointing you to relevant information.

The three major limitations of this approach are:

1. Your friends may have no archived social content that's relevant (or available) to your query. Searching within your Facebook network quickly demonstrates this problem. For this reason, augmenting traditional algorithms with friend-filtered social data may be better, rather than relying exclusively on data from one person's small exclusive network.
2. Current implementations are limited to keyword matching; whereas, searches that retrieve related posts based on topic, theme or timeframe might expose a wider set of results and combat the niche-social-network problem. This approach would be computationally harder than keywords alone, and exposing enough of the appropriate context remains a problem (see next item).
3. Understanding the context in which a post or link was shared is important. Without this, keyword- and even topic-matching might not convey to the user the relevance of a search result. Google provides limited context at the moment (showing only how you know a user, the source of the post and a short snippet). More testing is needed to learn how much and what kind of context is appropriate for social search content.

Similarly there is the issue of when friend-filtered social search would be relevant during a search. My instinct is that it will be useful throughout a search and for many types of searches (it is, after all, just another type of search result). This is critically different from collective social search and collaborative search.

Collaborative Search (a.k.a. Question-Answering)

"Collaborative search" is when two or more users work together to find the answer to a problem. This could look like IM-based question-answering (a la Aardvark ), Yahoo! Answers (which is relatively passive and asynchronous) or over-the-shoulder two-person search. In all of these cases, people speak to each other using natural language, which is incredibly useful for open-ended queries (e.g. "What is 'design thinking'?") or queries about unfamiliar domains (e.g. law, health, business, depending on your background). Such conversations, even not real-time ones, can assist people who don't know the right keywords to use (what's known as the "vocabulary problem").

My research has looked at the benefits of question-answering and at people's processes and preferences during search. Many users report that they want to attempt to search on their own first, or don't wish to interrupt their colleagues before they have given it a shot independently. This suggests that early social support should be passive (as with presenting collective or friend-filtered social data).

But later in the process, if the searcher gets stuck on a problem, they often turn to a colleague for help. If systems had a way of identifying difficult queries or search-process inefficiencies, they could offer more explicit social support to searchers. Perhaps the system could identify a domain-specific expert from the user's extended social circle. Information that this person has shared could be presented to the user, or this person could be suggested as a resource to chat with or email (depending on availability and preferences).

It should be clear by now that these three flavors of social search are complementary. Each has its pros and cons and is appropriate for different kinds of searches and during different stages of the search process. A powerful "social search engine" would be "smart" by making use of all three, while also exploiting the value of traditional algorithms.

Photos by: Who Wants to Be?, Claudia Lim and brewbooks.

Guest author: Brynn Evans is a PhD student in Cognitive Science at UC San Diego who uses digital anthropology to study and better understand social search.

But social search is not something to be afraid of. It's really just an extension of behaviors that we're used to in the real world, brought online, thanks to the advent of real-time social computing.

First, people have sought help from others for search problems and information-seeking tasks for a long time. Prior to the Internet, this was our primary way of getting information. We either asked a librarian to help us find something or, prior to that, we spoke to each other to spread information and seek help. We can reasonably treat this current trend in social search as a return to that familiar state.

Secondly, social search is no scarier than the Web itself. Social search will not make you easier to find. If you're the type of person whose social network data appears in search results, then you're already incredibly findable on the Web. Your tweets may reach more people than you want, but this also means that your ideas will be exposed to a wider audience, extending your reach and influence and presumably improving your brand. As before, if your tweets are currently open to the public, you're already reaching more people than you know.

The very real issue of relevance, though, comes up when dealing with social search. Noise, false information and spam could dilute a set of search results and distract users from their quest. But spam is context-sensitive. Well-targeted ads aren't regarded as spam; in fact, many people don't even know the difference between Google ads and the main set of results. Same with search. A Twitter post linking to an article on diet supplements may be perfectly relevant in a query on "dieting tips and tricks" on Bing. In any case, finding spammers on Twitter should be relatively easy judging by the follower-following ratio or similarity of tweets.

Social search is also not a fad. It's an improvement on limited algorithms that only index static pages. Chris Messina and Jyri Engeström remind us that we're in a transitional state from a Web of documents (in which algorithms were sufficient for surfacing relevant data) to a Web of people (in which PageRank no longer captures what's happening right now or happening among your group of friends). Facebook COO Sheryl Sandberg reiterated this at the Web 2.0 Summit, saying that there is a "shift going on from an information economy to a social economy."

Social search is one mechanism for leveraging the value in this shift.

So, as you get used to how companies like Google and Bing implement social search, think about how your community of Facebook friends, distant acquaintances in your Google contacts, nearby friends from location-based services and publicly bookmarked items might help you search better.

Guest author: Brynn Evans is a PhD student in Cognitive Science at UC San Diego who uses digital anthropology to study and better understand social search.

Sources: Gmail Contacts, Google Reader Subscriptions and Your Google Profile

Social Search can tab into three different sources, so you will need accounts on at least one of these services to make Social Search work.

Social search uses the Gmail contacts you have added to your friends and coworker lists and those you have chatted with on Google Talk. Social search also looks at your Google Reader subscriptions and the social networking profiles you have added to your Google Profile.

While you don't need to have a Google Profile, this service is a hub for your social networking profile on Social Search. Based on the information in your Google Profile, Google can auto-detect your social networking profiles and your friends on services like Flickr, FriendFeed, YouTube, Reddit, Digg, del.icio.us, BrightKite and many others.

How it Works and How to Trigger Social Search Explicitly

Once activated, Social Search results will appear at the bottom of the standard search results page and will be clearly labeled as "results from people in your social circle." As Google's search evangelist Matt Cutts pointed out to us in an interview earlier today, it is important to note that not every search will trigger Social Search results. When it does, however, the results should be highly relevant.

You can also explicitly trigger Social Search from the search options panel. There, Google will now also present a list of your friends that it thinks are the most closely related to the keywords you were searching for. By clicking on a name, you can restrict your search even further and just see results from this one person.

Social Search Makes Google Profiles More Useful

This new feature will also put a new emphasis on Google Profiles. Google has made some moves to make these profiles more prominent by highlighting some profiles when users search for people, but Google Profiles has generally not received a lot of attention from users. Now, however, as the hub for Social Search, users have an incentive to fill out their profiles - which, of course, will also give Google more information about you and your social network.

Privacy Concerns?

We talked to Google Fellow Amit Singhal, search evangelist Matt Cutts, and Maureen Heymans (the Technical Lead for Social Search) and Murali Viswanathan (the Product Manager for Social Search) earlier today and the team was obviously excited about this launch. The Social Search team was especially excited about the fact social search will now make your friends' knowledge far more accessible than ever before and that this will make it easier to find trusted product reviews and local search results.

We also asked the team about how they thought users would react to the fact that Google indexes and surfaces all of their social networking profiles and connections, which could spark some privacy concerns. In reaction to this, Matt Cutts pointed out that all of the info that Google indexes is already publically available on the Internet, including a user's friend connections. He also stressed that this was an opt-in experiment.

Social Search as "A Big Chess Move Against Facebook"

In theory, the idealism which infuses Lavva sounds like a good idea. "Search powered by the people," "results based on quality, not SEO," "transparency," etc. are the types of things Lavva likes to rave about when discussing their social algorithm. In practice, however, social search like this doesn't work. Obviously, it doesn't help Lavva's case that very few people know their startup even exists. Without users to rank the results, there's just no way Lavva can highlight the quality content.

Their idea just misses the mark, unfortunately. They would probably have had better luck if they overlaid their social algorithm on top of Google results, for example. That way the most relevant links would be retrieved first and then users could rank the results based on quality. (Rankings are done using little thumbs up and thumbs down icons beneath each result).

Do You Want to Chat about the News in Your Search Engine?

Another one of Lavva's hair-brained schemes is their "News Goes Social" page. Here, the engine aggregates top stories from a select few resources (CNN, Reuters, BBC, UN News, AP) and combines those with top search terms and the top links on Twitter.

While this in and of itself isn't entirely crazy (or entirely useful for that matter), how they want you to interact with the content sort of is. News stories have a "go social" link next to them which takes you to another page where you can chat, debate, and discuss the topic with other online searchers. After clicking through a number of these links, it was clear that no one was using this feature.

In a similar vein, users can sign into Lavva and click on the comment bubble icon under search results to leave their comments on the news story. Remember when Google tried this? Yeah, it was universally disliked then too. Frankly, this just isn't how people want to interact with search. And if Google couldn't make voting and commenting on search work, what hope does Lavva have?

Future Plans: Twitter and Facebook

While Lavva's service may get a little more interesting when they release their next update which plans to incorporate search results from sites like Twitter and Facebook, we doubt that alone will be enough for Lavva to make any impact. Even a startup as promising and innovative as the social search service that was Delver didn't make it, eventually selling out to Sears (yes, Sears!) in the end.

For now the best thing that can be said about Lavva is that it's powered by hydroelectricity, making it one of the greenest engines around. They plan to move to a solar-based system in the future, reports Seattle tech blog TechFlash. While we're happy that they're concerned about the environment, going green isn't going to be enough to make this social search attempt work.

Their latest release, Google@Omgili, features a sweet mashup with Google search, giving users a well-rounded look at the fair-and-balanced web alongside social buzz from and about sources such as blogs, newsgroups, video-sharing sites, forums, discussion boards, Q&A sites, and review sites.

The translation function isn't amazing, but for trending or hot-button topics, the results can be quite interesting.

The basic Google@Omgili search yields traditional web results with discussions that link to each result:

Once a user clicks on the "discussion bubble" next to the link, he can preview some of the social chatter linking back to that particular result:

Users can also choose to refine the search to show results from, say, forums and discussions only:

This product also incorporates the True Knowledge API and integrates Snap.com's service to give users a snapshot and video/RSS/Wikipedia previews. And of course, users can choose any number of social sites to share Google@Omgili with their networks (but not to share specific search results, which is disappointing) and can choose to add Google@Omgili to their search bar.

[UPDATE: The day after this post was published, Geva emailed us to report his team had added sharing of the results page on social networks. "We love feedback!" he wrote.]

Although the original Omgili search offering calls to mind many, many social/forum search engines - most of which have financially struggled or failed, see Twing and Delver - the Google mash gives this engine a much more interesting and useful product.

Overall, it's a nice way to see which links are generating (or supporting) subjective online conversations and might be a good tool for tracking down topic experts or sources. On the other hand, it might also be a way to get mired down in the chatter from forum crazies. Six of one, a half dozen of the other - after all, whether you're brilliantly insane or insanely brilliant, you've got to post your theories and rants somewhere, right?

Thurston started getting into this whole "Internet" thing in simpler times when the social web was called Usenet. He now carves out his niche at the overlap of the Venn diagram of comedy, politics, and tech. As an official Internet old-timer who makes it his business to stay relevant, Thurston has particularly useful insights on the business of curating applicable content with great efficiency and timeliness.

"I remember," Thurston said, "back in 1996 or 1997, when you could finish the Internet... You could stay up until two or three in the morning and go to sleep and know, 'I read the Internet today.'" Simpler times, indeed.

So, with the mind-boggling multiplicity of blogs, news sites, and social networks, how does a professional netizen maintain cultural and technological relevance? And what tools does the modern, socially cognizant webmaster use to track and optimize traffic in real time? Call us cruel, but we prefer you watch the video and hear it all firsthand.

Results are prioritized based on an algorithm of about 26 factors, filtered for spam, and unduplicated if links are shared through multiple URL-shortening services. There are two search modes: Users can browse real-time results or (in "pulse" mode) see links ranked by social relevance. We spoke with a caffeinated and exuberant Tobias Peggs, general manager at OneRiot, about 20 minutes before the new release went live at 9 a.m. "We're trying to get a sense of current social relevance; what are people talking about right now," he said. And more than any competing product currently available, OneRiot succeeds.

"One factor in the algorithm is velocity, the hotness factor: How many times a link has been shared on the social web within the past minute?" said Peggs. "We can get a sense of whether a story is increasing or decreasing in popularity from one minute to the next."

Users' online reputations also influence the score and help provide a means of ranking social relevance, kind of like Google's PageRank system. "The trouble is," said Peggs, "who's to say what's more important: A link shared by Scoble or a link shared by me? It doesn't necessarily mean the links he shares are more important."

A link's relative popularity (how many times has a link been retweeted or dugg) is also normalized so potentially relevant results don't get dwarfed by stories on CNN or major outlets.

New features of the current OneRiot release include searching within specific domains and blogs to determine which articles have the most current social buzz, searching for a specific URL to see how many shares it's received, identifying the first user to share the link, and expanding results to see every single share along with the user and any surrounding text.

So if OneRiot results are intended to be the freshest, most current links available, when do results disappear? "Because the results change in real time, users can come back and keep searching to stay on top of the latest information," said Peggs. In other words, top results will constantly fluctuate, and the maximum time a link will appear without refreshed bouts of sharing is a few days, meaning constant vigilance would be needed for social SEO in real time.

Also, because of the real-time nature of the product, Peggs noted, "Users tend to search more times per day per query than on a traditional search engine. Ultimately that's good for monetization because we have the opportunity to reach those users more times per day." Monetization plans at this point primarily consist of serving display ads.

When breaking down search behavior, Peggs stated that twenty percent of search queries are navigational (users seeking a specific URL) and 40 percent are queries for static information such as recipes or contact information.

"Google does an amazing job on that 60 percent," said Peggs, "but the remaining forty percent of users are looking for what's going on at this particular moment. What are people saying right now? Because of the way Google indexes the web and the amount of time it takes to index and rank pages, they are never, ever going to serve those socially relevant, very fresh search results. OneRiot is not a Google killer; it's not going to find your dentist's phone number. It's a completely different experience.

"What we're doing is showing people the real, true potential of real-time search," he continued. "It's way deeper than seeing a string of conversation around a keyword. We're uncovering the web content - blogs, videos, news stories. It's very different from what traditional search can do. Hopefully, people will start to see the potential of real-time search."

What's Glue?

Glue is a new browser plugin from Adaptive Blue. It uses semantic technology to connect you to your friends around things like books, music, movies, stars, artists, stocks, wine, restaurants, and more. The plugin places a bar - not a toolbar, just a bar - at the top of your browser window when you visit certain popular web sites like Amazon, Yahoo! Finance, Wine.com, IMDB, Wikipedia, Citysearch, Last.fm, and many others.

As you read about the album, movie, book, or whatever else it is that you're viewing at the time, you'll have a toolbar at the top of the page where you can see which of your friends had visited the same page, if they liked it, and if they left a comment.

Glue Is Not Co-browsing

Glue is not a co-browsing plugin like Me.dium nor does it try to socialize the entire web surfing experience like Socialbrowse (our coverage). Also, unlike Headup, another semantic browser plugin we covered recently, Glue doesn't bother you with pop-up messages as you surf. Glue simply provides a social element to web pages in context - there's no destination site to join and your social graph doesn't need to be re-created in order to use it.

How It Works

In order to tap into your network of friends, Glue uses APIs from popular social networks like Facebook, Twitter, and FriendFeed to import your friends. You can choose to import one or all of those friend lists into the plugin.

To participate in the Glue network, all you do is continue browsing the web normally. When you visit a supported site, the Glue friend bar appears. If you choose, you can view what your friends say about the item on the page, or you can ignore the bar and continue on your way. However, your visit is recorded and when one of your friends visits that same page, they can see that you've been there recently, though not the exact date or time your visit occurred. This information is only stored for the last 20 things you've visited on the web.

While surfing, if you want to share your thoughts about the item you're viewing, you can optionally use the Glue "like" button and/or the "2 cents" button which lets you add a quick thought about item. You can also click on the bar to see the profiles of your friends, other recent Glue users, and you can explore their interests even further by clicking into their profiles, which display in a pop-up box that appears when you click their avatar. You can also optionally click on "Actions" to explore the item you're viewing on other Glue-supported sites.

Making The Social Web Relevant

By providing this social experience in context, Glue can actually be more useful to you than simply joining isolated social networks surrounding your interests where your data and that of your friends is trapped inside the network's walls. It may also have some appeal over a lifestreaming service like FriendFeed, because you don't have to happen across the information - it's there when you're actively interested in something and have sought it out on the web.

In the official version coming soon, the company is also soon going to provide a method for any web publisher to "Glue-enable" their site by simply adding AB Meta to their sites, by inserting three lines of code in the header of a page.

Glue is the next generation of the Adaptive Blue plugin, a tool that currently has around 350,000 active users. Current Adaptive Blue users will find their plugin updated to Glue through the standard Firefox plugin update process. For everyone else, you can download the plugin here.

Although at the present time Glue is available as a Firefox plugin only, an IE version is in the works and an iPhone plugin will arrive in a few weeks.

Disclosure: AdaptiveBlue's CEO, Alex Iskold, is a feature writer for RWW.]]>Discuss]]> http://www.readwriteweb.com/archives/put_the_social_web_in_context_with_glue.php http://www.readwriteweb.com/archives/put_the_social_web_in_context_with_glue.php Products Tue, 28 Oct 2008 12:00:00 -0800 Sarah Perez ...And Continues To Be Awesome!

If you're a Firefox and Google Reader user and you haven't yet installed the Feedly plugin, you're going to want to install it today after you hear this: Feedly has now integrated its own results - that is, links to the relevant posts from your Google Reader - right into your Google search results. This integration essentially adds a layer of social search directly into Google, and all with no extra work on your part besides simply having installed the plugin.

To get you up to speed, Feedly is a service and a Firefox plugin that provides an alternative way to read your feeds. Its magazine-style interface presents the news to you in a more visually appealing way while still keeping it categorized by your own tagging system. As your read news items in Feedly, they're marked as read in Google Reader; as you "Recommend" items in Feedly, they're "Shared" in Google Reader; as you "Annotate," they're shared with a note, etc.

However, for voracious RSS readers like myself, Feedly is a much slower way to read the news, which is perhaps why it hasn't really taken off to a great extent among the early adopter set. That's a shame, though, because Feedly can do so much more than many people may realize.

For one, it's integrated with both Twitter and FriendFeed letting you tweet or share an item on either of those services (even FriendFeed rooms)with just one click. It also pulls in the blog comments, comments from Digg, and the FriendFeed conversations surrounding an article and lets you add your own thoughts to FriendFeed stream - without you having to view the page on FriendFeed itself.

Although it does show full feeds (just like Google Reader), it also allows blog owners to showcase their sponsors via a designated ad spot. On an area called "The Wall," you have the option to configure the page to show a mix of recommended items from Google Reader plus other sources. Those other sources can include your Firefox Bookmarks, your MyYahoo items, items from NetVibes, and items from Bloglines, as well as social connections like your Twitter stream, your FriendFeed stream, items from Yahoo Mail or Gmail. You can also select "best of" content to be mixed in from any number of categories from fashion to celebrities to tech or you can even upload your own OPML file.

Feedly Secrets

There are still a few secrets to discover about Feedly though. Despite all that it does, you may be surprised to know what more it's capable of. One of the biggest undiscovered gems is that as you "follow" friends on Feedly, you are immediately tapped into their Google Reader Shared Items - even if you're not "Google Friends." Their shared items will show up in Google Reader in a folder called "z.feedly.people." This is great for those who don't use Gmail or GTalk but still want access to people's Google Reader Shares. Friending people right now is somewhat of a difficult process because you have to find users on your own - there's no search feature for finding people. However, this friends feature is said to be getting an overhaul soon and we're looking forward to checking out.

Spring cleaning is a feature available from the "more" option that helps you eliminate the feeds you don't read. The interface color codes feeds to let you know how much you like them. The colors are as follows:

• Red: A candidate for deletion because you are not reading articles from the feed and/or the feed does not seem to produce articles.

• Orange: Produces a lot more than you are reading

• Green: You seem to like those feeds a lot.

• White: Everything else

Feedly + Google Search Integration

But today's big news is the Feedly+Google Search integration. As you search for a subject in Google, Feedly search results will appear at the top of the page. At present, the search results will only appear if you have related RSS items 7 days old or less (which explains why I had not yet seen this before).

This integration adds an immediate social filtering aspect to searching the web by promoting your favorite sources to the top of your search results. It even takes into account your reading patterns and the favorite metadata to sort the results.

With this feature, it doesn't even matter if you want to use Feedly to read feeds - this behind-the-scenes social filtering makes it a killer add-on for anyone who uses Firefox and Google Reader. You can download Feedly from here.

We told you about the new social search engine Delver earlier this year (see: "Delver Reinvents Search"), but at that time, the site was still under development. Today, Delver has finally launched their service.

When you do a search on Delver, the service taps into your social graph to return its search results. It organizes and ranks publicly displayed content that comes from various online services like social networks, web sites, blogs, bookmarking services, and photo and video sharing sites that your friends belong to. A 'breadcrumb' is shown next to each result, showing how that result is related to the user, so you can see how it's relevant to you. What Summize is to Twitter, Delver is for your entire social graph.

At the moment, Delver currently covers Myspace, Blogger, Flickr, LinkedIn, Youtube, Hi5, FriendFeed, Digg and Delicious. Other sources, like Facebook as well as other blogging platforms will be added to the service over the next few months. Facebook is harder to tap into because of their privacy controls, so to add Facebook content to Delver they're building a Facebook app. The app will permit you to explicitly share select items from Facebook - like photo albums, for example - with Delver's service.

Some Search Results on Delver

Search Buddies

A feature makes Delver even more useful, though, is the "search buddies" feature. This option lets you add certain friends as search buddies which will prioritize results from their networks higher than others. This way, you can give information sources you trust and value more weight than others. In addition, those you add don't have to be friends you're connected with on social networks - you can add anyone as a search buddy and they don't need to confirm the addition - it's not a "friending" feature, exactly, just a way to see more personalized results to your queries.

This feature is also useful for those people who aren't as active on social networks themselves, but have friends who are - they don't need to rely on their particular social graph per se, but can tap into sources (people) whose content is relevant to them.

Add Search Buddies To Make Search Results More Relevant and Personal

Save Your Stuff

Delver lets you save items you find interesting or informative to a sidebar of "Kept Items." The system automatically categorizes them as web items, images, music, videos, or people. In a month or so, Delver will allow for sharing these items more publicly through the addition of a Kept Items widget which could highlight findings on your own web site or blog.

Saved Items

Partner Program

Along with the launch of the search service, Delver is also launching a partner program that will allow sites that host user generated content to integrate Delver's technology into their web site.

Social Graphs - Social Media's Next Frontier

While the big networks like Facebook, Google, and MySpace all fight for control of your social graph, services like Delver provide an easy workaround to tapping into the power of your social network. In addition, you can also use Delver to discover interesting "friend-of-a friends" that you may want to follow, as Delver digs deeper than just who you are connected to yourself.

Your Social Graph on Delver

Delver is still at a very early age of development, but even so, the service demonstrates a lot of potential to become the next step in the social web's evolution.

But don't call Delver a "social search engine."

"That name belongs to services like Mahalo," says Liad Agmon, Delver CEO. "We prefer the term 'socially connected search engine'." That term makes sense because Delver is not a social network built around a search engine, but a search engine who indexes and queries your social network to deliver its results. Instead of just looking at a web site's popularity, Delver looks at information like whether your friends have tagged the site or if it's found on their social network profiles, bookmarking sites, photos and video sharing sites, or on their blogs. The results are more relevant because they account for who a person is and what they find valuable.

Agmon adds, "People want trusted information from their friends, but may not know who in their network is knowledgeable about a given topic. We make Web search more fun and meaningful by prioritizing results based on a user's network, while enabling the user to discover others in their extended network who share common interests."

Even without registering for an account, Delver will try to determine who you are by searching any public social network profiles you may have on sites like Flickr, Facebook, and YouTube. If you do decide to register on the site, though, you can then choose to associate your accounts with Delver in order to obtain even more accurate results. Delver currently indexes the entire web, and specifically indexes people's social connections on flickr, MySpace, LinkedIn, YouTube, hi5, facebook, Blogger, and, they are adding more all the time. When they go into public beta (circa May, 2008), an optional email import process will be provided as well.

Maximizing Your "Whole" Social Network

Many of us have friends, family members, or colleagues on sites like MySpace and facebook who aren't into using all the latest and greatest web apps and technologies. These friends may have a MySpace profile or a blog, but without visiting these sites directly, there was no way to gather information from these people before. Now with Delver, their profiles and contributions to your social graph are indexed.

No one has to sign up for Delver for you to have them included in your search results.

This is a real breakthrough since prior to Delver, the maximum value you would get out of social networks was directly related to how many of your friends would join. I don't know about you, but I still have plenty of friends who are on MySpace and nothing else, and are quite content with that. With each new social network I joined, the number of my non-tech friends that would follow me dwindled down to nearly nothing. Now it doesn't matter. They can stay on MySpace forever and yet the content they create there will be valuable to me.

Privacy Concerns?

It's important to understand that Delver doesn't display anything that isn't already publicly available. "If Google can get to it, so can Delver," says Agmon. But Delver just makes it so much easier to do so. You can access people's social information with such ease that anyone who hasn't been good about setting their profiles to "private" (or who doesn't know to do so), may be surprised to find themselves searchable on Delver.

Using Delver

After claiming your identity in Delver, your social graph is mapped and displayed for you beneath the Delver search box. Dotted lines connect you to your friends and your "friends of friends."

When you perform a query, results from all over your social web display.

You can narrow down your search to just display the people related to your search term or just media results by clicking the links at the top.

Each search result displays, via a  breadcrumb trail, your relationship to the person associated with that result. You can hover your mouse over their name to see their photo and their relationship to you. Even if you and them are not directly related as "friends" on a social network, you can still click the plus sign beneath their picture to add them as a connection. This will then add them into the mix of your search results in the future. This way, you can view the relevant bookmarks, links, blog posts, photos, and videos of people like you even if you don't know them personally...and they don't have to confirm the connection on their end.

Alternately, you can choose to exclude certain connections from your search results as well, which is perfect for eliminating those "who-is-that-guy?" friendships left over from your days of MySpace friend accumulation contests.

When Will It Arrive?

Privacy Issues

Integrating social search into search results is tricky, says Mayer, because people view search as a private activity. Most people just aren't that comfortable letting their whole network of friends know what they have been searching for. Google knows that they must respect that privacy, so they would want the user to explicitly approve any friend connections that would be used to add social elements to the search results. (In other words, you won't just log on one day and find your  Google search results re-ranked based on what your MySpace friends are doing.)

How Will It Work?

When asked how Google is planning on implementing social search, Mayer mentions a few different ideas they have which include labeling, identifying users like you, social network integration, and a social-influenced PageRank.  

Labeling: With labeling, Google users could annotate the search results and those notes could then be shared with friends on their social network or with others like them. She mentions that this has worked to some extent in Google Co-Op in certain areas, like health, but overall annotation is not a model that works well in its current state. However, the benefit of annotation is that it avoids the privacy issues because someone who is labeling search results presumably does not mind that others would see those notations.

Users Like You: Another option might involve Google taking a page from Amazon's book, and adding "others like you searched for ..." or "other people who did this search also did searches..." to Google's search results. Although useful, these related queries don't truly integrate results from your friends, nor do they influence the search result rankings, so they are not the best example of pure social search.  

Social Network Integration: To identify your friends and allow them to influence search results, Google may even try social network integration with search. Using aggregate statistics on your friends' searches would allow privacy to be maintained, but you would also be able to see trends that are important to you. Initially Google would leverage the Google user base and the connections that exist within it. However, 3rd party social network integration may come in time as well. Mayer uses the example of how you could see that several of your Facebook friends had searched for a particular topic one day - a stat that would be provided without user names. If a large number of your friends are searching for something, it's likely that you may be interested in that topic, too.

Social-Influenced PageRank: With today's version of PageRank, it's the link structure of the web that determines the most authoritative pages. However, Google believes that people would naturally give more authority to pages their friends visit. To bring in this influence, Google could take web history and then allow that data to influence rankings, so that pages that your friends visit would rank higher in search results. Today, Google web history is still an opt-in option and if it was going to be used to influence rankings, that would hopefully be an opt-in choice as well, but Mayer does not go into that level of detail.

The Future

So, what is the future of search? Mayer responds, "I think one way it will be better is in understanding more about you and understanding more about your social context: Who your friends are, what you like to do, where you are. It’s hard to imagine that the search engine ten years from now isn’t advised by those things."