Bing and Google in an Arms Race for Features
$MSN is making some gains in its attempts to get a foothold in the search business. Google is by no means running scared by is adapting a Bing-like look and in some cases tweaking its functionality.
$MSN is making some gains in its attempts to get a foothold in the search business. Google is by no means running scared by is adapting a Bing-like look and in some cases tweaking its functionality.
If you’ve spent any appreciable amount of time on Twitter, the term “TweetUp” should be pretty familiar to you. In Twitterspeak, it means a gathering of all connected tweeters in a particular city – not on Twitter, but in actual three dimensional space (hard to believe people still like to interact with each other in the flesh, isn’t it?). But as of this week, “TweetUp” has new meaning now that a new service has emerged which has chosen the term as its namesake.
It’s being widely billed as a Google for tweets, so it should come as no surprise when I tell you that it’s essentially a search engine for Twitter with a twist. That twist being that 1) it uses an algorithm to find the “best” and “most relevant” tweets for a given search, which will be displayed at the top of the results, and 2) it will allow users to pay to be listed at the top of the results for a given TweetUp search.
So first of all, it’s designed to help you search Twitter without having to sift the through piles of garbage tweets you and I see every day. This sounds great in theory, but I have to admit, saying that one of its searches displays the “best” results sounds like an awfully subjective judgment for something that uses an algorithm to determine what’s best. The algorithm seems to analyze the “authority” of a Twitter user who sends a particular tweet, along with metrics from URL shortening outfits like bit.ly to determine what’s best and most relevant.
But the other, and more important aspect is the idea of sponsored results. As you might have guessed, this is a Google Adwords type model, and indeed those who pay to have their tweets (or their Twitter profile itself) appear atop certain search results will bid on the keywords in conjunction with which they would like their tweets to appear. Now, being an avid tweeter and just generally aware of social media marketing, I can see many companies being interested in this type of thing. As for individual tweeters, I’m not so sure there will be much interest. But I guess anyone looking to “build their brand” through social media, whether it’s a company of 500+ or just a freelance professional, might be intrigued by the prospect.
In addition, TweetUp has deals in place with Twitter application developers like Seesmic, which will use the TweetUp’s search in their applications.
The natural question to me here is not who will pay for this service (at least in the short term), it’s – will it have any effect on those searching Twitter. Because TweetUp offers this pay model, in addition to its natural search offering, users may end up encountering not the best and most relevant tweets related to their search, but rather tweets that might not be related at all. Further, even if they are relevant, they may not be anything you’d really want to read.
And in all honesty, I don’t know how often users really even search Twitter. Very rarely do I myself search, and if I do, it’s usually just clicking on a hashtag that’s popped up in my Twitter feed. Plus, obviously I can still use Twitter’s normal search function if I really want to find certain content-specific tweets.
I guess like anything else it comes down to how well the service functions. If it truly plugs a whole that Twitter users have been looking to fill, then I’d expect it to be at least a relative success. But if it’s just a whole bunch of paid tweets that don’t offer the quality users are expecting, then I’m not so sure. I can’t say it enough – content is king, my friends. So I guess I’ll just have to give it a shot and decide.
This one completely passed me by last fall, but apparently there is a unique Social Search Engine called Aardvark that unlike traditional search engines, proclaims to find “people, not web pages that have specific information.”
Pretty interesting concept, as you might imagine – instead of using keywords to search Aardvark, like one would with Google or Bing, users ask Aardvark a question, and instead of returning “the most authoritative” web pages related to their search like Google or Bing, Aardvark connects them with other users that its engine believes can help answer their question(s). When you sign up, you tell Aardvark what topics you think you can answer questions about. Then, when a user has a question about any of those topics, Aardvark queries you and if and when you answer the question, Aardvark emails (or IMs) the user your response. So it really is connecting people to people, as opposed to connecting people to static content.
Anyways, Aardvark caught my eye this morning because as TechCrunch reports, yesterday they released findings of their study in a paper that they’ve titled, “Anatomy of a Large-Scale Social Search Engine.” The title is a play on Google’s seminal paper, “Anatomy of a Large-Scale Hypertextual Search Engine,” and just like Google’s paper, Aardvarks has been accepted to the WWW2010 conference, the same conference to which Google’s paper was accepted back in 1998.
Now, whether or not Aardvark is the Google of Social Search engines remains to be seen. Its search engine is less than a year old, and is still gathering force (though it supposedly already has over 90,000 users). Still, it is a relatively novel concept and Aardvark may really have something if they continue to invest tirelessly in making (and/or keeping) their model appealing to users.
Of course, the idea of asking a community a question really isn’t that new. Users could, and undoubtedly do, ask the same questions they might ask Aardvark, on Twitter (where they could even tag their tweet with a topically relevant hashtag), Facebook or on any number of other social sites, and expect to find an answer. But the way Aardvark does it is slightly different, and it even looks at your blog posts, existing online profiles, and tweets to determine what topics you might know about, to augment its offering.
The thing Aardvark adds above other platforms is that users have already declared themselves experts on certain topics, and whenever a user (in their area) has a question about that topic, Aardvark inserts the question into that users “Answer” queue (screenshot below is just an example of a question, though I cannot confirm or disconfirm that I’ve declared myself an expert on hair gel…guess that’s what I get for listing my Hoboken, NJ location in my profile, right?).
From there, that user can answer the original users question if they know, or just leave it for someone else.
As for how well it works, I cannot say. I myself just signed up for an account today, and haven’t really had the time to give it any kind of rigorous test. Nevertheless, if you’d like to know more about what Aardvark’s engine does or the data it discovered and is publishing in its paper, I encourage you to check out TechCrunch’s report (mentioned earlier), and/or Aardvark’s own blog post (also linked to earlier) for further specifics.
Oh, and as always, feel free to add any other insights you may have in the comments.
ScanBiz Mobile solutions is one of a slew of iPhone/mobile startups that, with almost no overhead, is putting out applications for handheld devices. The true challenge here is how to get your application to rise above the rest. It goes without saying that it has to be a quality application. But at least half the battle is standing out, getting noticed. Marketing savvy and PR are essential if you’re going to make money.
ScanBizCards is a very cool app that scans business cards. But that’s really only the beginning. Founder Patrick Questembert and Danny Panzer have devised numerous applications. For example, you can connect to Linkedin and send out introductory emails and enter info in your address book after just a scan. More applications are on the way.
It’s a worthwhile app to add to your arsenal….
Don’t look now, but thanks to a new collaboration between the makers of TweetDeck and TwitJobSearch, Twitter may actually help you get a job, which oughta prove pretty handy when lose your current one by Tweeting when you should be working.
TechCrunch describes the cleverly-titled, “JobDeck” as “basically just a custom branded TweetDeck client that comes with two additional default columns: ‘Job Search Experts’ and ‘TwitJobSearch’.” And that’s pretty much accurate. The two additional columns are just a feed of Tweets from a number of Tweeters that TwitJobSearch has deemed “job search experts,” and a feed of TwitJobSearch’s own Tweets (a mixture of company news, jobs users have tweeted, and jobs that recruiters have requested TwitJobSearch post). But, you can add a column for your Facebook and/or LinkedIn feeds, which can also help, and as with the traditional TweetDeck, you can always create a custom Twitter search for the type of job you’re seeking.
Just as an aside, the JobDeck app is new, but TwitJobSearch is not. If you’re currently in the market, you may want to check out their web offering as well. There you’ll find the most salient feature, which is the ability to search the vast depths of the Twittersphere for whatever job you desire, as well as a pretty cool “Job Map” that shows the locations of all jobs Tweeted within the previous 72 hours (which are clickable – providing the ability to search by location as well as by keyword).
It also gives you the ability to refine any searches you perform by location (city or country), salary, and job type (full time, part time, contract, etc.), and aggregates Tweets from a number of “sponsors” (including Adidas, KFC, and Ernst and Young) who are all offering jobs.
They even have an iPhone app you can download, which I can’t really vouch for since I’ve yet to give it a whirl, myself.
Anyways, if you’re already using TweetDeck anyway, what have you got to lose, right? JobDeck is out now and should be available for both Mac and PC users. So go forth, try it for yourself and let us know what you think!
I gotta say, I’ve yet to really get into “real time social media search engines,” as they’re commonly called. It’s not I don’t think they’re a valuable tool, it’s just that heretofore, they’ve seemed kind of primitive. Plus, obviously I’ve been content to labor inefficiently, searching Facebook, Twitter, YouTube and more one-by-one.
Still, Sobees‘ new real time social media search offering is pretty slick.
Unlike a couple of other similar tools, the web-based application’s interface separates Tweets (and other real time social media sources), videos, pictures, web search, and news search (as seen in the screenshot above), into five modules, all on one screen.
Compare that to OneRiot (which Sobees actually searches itself), where you have separate tabs for “web” and “video,” and you don’t have the option to sort by platform…
Or Collecta, which throws everything together in one long stream…
Consider also, the settings control that Sobees offers…
…and it seems like Sobees may have the makings of a slightly better alternative.
Add to that the fact that Sobees’ track record in real time social media – it already offers its own desktop client that aggregates Facebook, Twitter, MySpace, FriendFeed and LinkedIn activity (though it is Windows only), and has plans to branch out into the mobile space – and you’ve got a pretty promising future, in addition to an already solid application.
It may not be nearly as in depth or sophisticated as some other enterprise oriented social media monitoring tools like truCAST, but if you don’t have tens of thousands to blow and you’re looking for an efficient (and free) way to keep an eye on what’s being said about your company throughout the web, Sobees Search just may be a good fit for you.