Social Search Overview
Happy Day 3, everyone.
[I don't know if my head is just getting big now that I've met Danny, but someone just walked into the session room and asked to look at my SES schedule and I'm pretty sure it was Seth Godin. Like, The Seth Godin. I know! Could it be? -- (10 minutes later) It was Seth Godin. I just spoke to Seth Godin. I am so cool!] [So. Jealous!--Susan]
I have my Grande White Mocha in hand ($4.99 for a GWH? I was totally ripped off due to my tourist status…but it sure is delicious) and it’s time to enjoy this morning’s Social Search Overview session. Chris Sherman (Search Engine Watch) is moderating the session which features Grant Ryan (Eurekster), Tomi Poutanen (Yahoo), and new late additions Apostolos Gerasoulis (Ask.com) and Seth Godin (Squidoo) speaking.
Seth Godin and the father of Ask.com in one session? I am in geek heaven right now. I’m so star struck I’m sitting frozen in my seat but what I really want to do is jump up and down a whole bunch of times and maybe run up and down the aisle. However, that would probably be seen as "not cool" so I’ll try and stay planted.
Okay, back to the session.
According to the fancy wording in my SES program agenda, "humans are hot again". And not just any humans (humans like Sadie), us geeky humans. The search engines are categorizing the Web by tapping into our obsessive tags, watching our clicks and using our search history to try and see things the way us dorks see them. Things have sure changed since high school.
Chris Sherman starts off by explaining social search. He says it is simply Internet wayfinding tools informed by human judgment. I’m sorry, but there is nothing simple about that definition, Chris. In fact, I’m not even sure what it means. Chris says informed can mean many things – including egregiously uninformed. No good industry standard definition yet.
The interesting thing, says Chris, is that we’ve always had social search. The very first guide to the Web was created in 1990. You probably know that Yahoo was originally created by a team of human editors. It was basically the first social directory. Meta tags were then created to help content owners influence search engines – and were a massive failure.
Tagging is pretty much Meta tags in a different form.
Algorithm search itself is social because fundamentally search engines reflect human bias in the form of programmer choices. They determine what’s quality and what’s spam. Search engines also observe human behavior like click paths, popular URLs, etc and use this information to modify their algorithms. As of late, we’re even seeing new personalization efforts to refine search for everyone.
So if social search has been around for years, why is it getting so much buzz right now?
Chris says it’s because algorithmic search has plateau with innovation being much harder than it used to be. We’re also realizing that humans are still better at some things than computers (like whistling). A major factor is that many, if not most, of the players in social search are leveraging the work of volunteers.
There are many different types of social search, including:
- Shared Bookmarks & Web pages: Del.icio.us, Shadows, MyWeb, Furl, Diigo
- Tag Engines (blogs & RSS): Technorati, Bloglines
- Collaborative directories: ODP, Prefound, Simbo, Wikipedia
- Personalized Verticals: Google Custom Search, Eurekster, Rollyo, Trexy
- Collaborative Harvesters: Digg, Netscape, Reddit
- Social Q&A Sites: Google Answers, Yahoo Answers, Answerbag
The problem with social search is the scale and scope. When people go out and tag we have a problem with language. What’s orange? A fruit, the color of a shirt, a sunset? The trouble with tagging is that language is ambiguous, there’s a lack of controlled language, humans are lazy, and some of them are just plain idiots.
The other problem with social search is the spammers and people trying to game the system. Cheaters ruin everything.
What will ultimately work?
The combination of algorithm and people-mediated search will become the new format. Yahoo is doing a great job of this as of late. We’re going to see people trusting other people’s judgment – unless those people are idiots. Trust networks are going to evolve, and we’re going to see increased personalization and user control over result filtering. We’ll be able to say I don’t want results from Site X, I want them from Site Y. Social search will work best for non-text content like photos, music and video.
Seth Godin is up and says people are here because search is broken. People need more traffic. Seth built Squidoo because he saw a need for human beings to create a post search engine filter. Users can use Squidoo has a launching pad to find what they are ultimately looking for. The purpose of a Squidoo page is to make people leave it.
If you’ve got a Web site that’s not getting enough traffic the wrong thing to do is trick and engine into sending traffic you don’t deserve. The right thing to do is to open up the deep pages of your site and expose the stuff that’s not getting attention. Don’t do it yourself; have your fans, readers, and the people who buy from you to build out the content. If they built out content because they like you, they’re putting relevant and meaningful content in front of potential customers.
Apostolos is up next. I’m trying not to make eye contact in case it causes me to melt into a puddle. He says it’s his first time speaking at SES, which makes me that much cooler, I think.
Apostolos says he’s here to say that a new generation of search is going to start appearing. You need to figure out how you are going to rank in the future with this new technology.
He defines social search as a knowledge that you use to guide you into a region of what you want. I think that’s a really awesome definition. If you want to find exact, encyclopedia-like information, then a traditional engine will help you, but it’s not going to take you to the region of knowledge you’ll have in front of you through a social engine. The Web is static. If you have knowledge (education, experience, etc) you can guide and give what the user wants.
It’s about understanding who you are. If you search for Yale, social engines will give you info about other Ivy League schools like Harvard and Princeton. It’s one search for an entire knowledge region or community.
The future of search is the integration of social knowledge to guide you to the right community. Apostolos says Ask.com has been using social search for awhile but they just haven’t been talking about it and uses their Image Search as an example.
Social search is an important future area that we need to support.
Next up is Tomi Poutanen from Yahoo. All of his slides are pictures. I like Tomi.
Tomi highlights the three ways Yahoo is using social search:
- Flickr: Tomi shows the difference between Google Images and Flickr by searching for golden retrievers. Who doesn’t like puppies in the morning? I know I do.
- The different between Flickr and regular search is that you can take human interaction into account, which therefore makes the pictures are more engaging.
- Del.icio.us: Tomi asks how many people use del.icio.us and 75 percent of the audience raised their hand – that’s pretty impressive given the geeky nature of the product. Way to go, guys! He does a search for "nyc hotel" and shows how the results that appear on del.icio.us are considerably more relevant than what you’d find from a traditional engine.
- Yahoo Answers: Answers enables you to ask as question and the community will answer. It’s about people helping people (Isn’t that song?). The service is also being show on the bottom of Yahoo’s SERP to complement the search experience.
The main problem with social search is that there’s no way to track who said what. There’s an anonymity there that isn’t found on TV or in print, and it’s dangerous. People can say anything they want.
During the question and answer period, an audience member asks about potential revenue model for social search.
Tomi responds by bringing up the late Google Answers which bombed because people weren’t willing to pay for information. The motivation that drives people to share is noncommercial interest in helping each other and sharing information. Introducing a financial incentive changes that and it makes it participating solely about the financial incentive. Once this happens people will start worrying about not getting paid enough instead of producing quality content. Going down the path of paying people is a really risky path you can’t reverse and it changes the dynamic of the communities.
Seth agrees that money changes everything. There’s a long history where you work til free until you get the top of the pyramid and then you get money. The other thing money changes is that it eliminates anonymity. If there’s money on the table, you know who’s producing the content. Seth says anonymity is the worst part of the Web.
Apostolous says allowing money in social search is the same as the monetization of the Web. He also says that money almost destroyed the quality of the search engines.
I have to say, I really agree with the panelists’ thoughts on introducing money to the equation of social search. If things take off the way everyone is expecting, it will likely be a necessary evil, but providers are going to have to be really careful with how they go about it. One wrong move and the entire model will be ruined.
This was a great, great session. [So. Jealous. --Susan]