The Five Best Real-Time Search Engines: Which One Should You Use?

406635986_fa8da57692_bReal time search has exploded the past few months. The general mission statements of most of the new startups highlight Google’s lack of a real-time search function—its inability to track what’s happening on the internet right now, beyond daily results on its Trends page.

The consensus is that it’s time for a new kind of search, one that finally brings the internet in-line with the sort of ‘breaking news’ we might see on TV, or that ubiquitous, useless ticker at the most of most 24-hour news channels.

Writing in Seattle’s The Stranger, Paul Constant recently explained why real-time is important:

The thing most people will probably, eventually use Twitter for is its clean and efficient search engine.

The search function on Twitter is an amazing thing: It’s a focused laser beam into what people are thinking about right now.

And it’s permanently set to “now”: It doesn’t have anything to do with the past or with archiving. That is an innovation in and of itself.

An inordinate amount of the internet is devoted to archiving and filing all the world that existed before the internet.

Twitter is reflexive, instinct driven, present. It doesn’t care about the past. It’s hard to find a post older than a week old. It’s work.

Twitter’s own search has been covered to death online, and the consensus is that it needs competition and improvement. As a result, an almost comical number of competitors have stepped up to sort and catalogue what’s happening in real-time on the internet right now.

Below, we look at 5 of the most interesting ones and let you know how they work, what they do for you, and whether you’ll want to use them every day.

Friendfeed

friendfeed-search-xinjiang_1247063052946-w595

Friendfeed was the first search I tried. To me, it ended up looking more like a blog search—most of the stories were people posting major newspaper articles pulled by Google News, or from the New York Times directly, or from other major international sources like Spain’s El Mundo. I saw a few results pulled from delicious.com and also some posts by the New York Times’ own friendfeed stream as well. The real-time updating was nicely done, but for the subject I chose, there simply isn’t enough at-the-moment information coming out to make a search like this worthwhile.

After about 30 seconds of waiting for someone else to make a comment, tweet, or social-bookmark another article, I moved on. As it stands, I could get a nice idea of how a story is developing through conversation online, but a quick Google News search would get me more relevant information a whole lot faster, considering almost everything in the feed is just links to old-fashioned journalism anyway. Perhaps with a still-rapidly-developing story things might be different, but again, a huge amount of conversation (even about ultra-fresh breaking news) comes through as links to newspaper websites.

The Verdict:

A nice but very simple interface that pulls from a wide range of sources (tumblr, twitter, delicious, digg, facebook, youtube) to get its information. Whether that information is any good is another story. If you use it just to search your own friends’ streams (and have a lot of friends online) it might be great—otherwise it’s search away and hope for the best.

Scoopler

xinjiang-scoopler-search_1247063045089-w595

I liked Scoopler’s approach, as it shows real-time results updating in the center column, while displaying various ‘popular’ results (sorted into videos, links, and images) on the right. I tried following a few other ‘hot topics’ as suggested on the left, and most of the twitter results that showed up were near-unreadale, full of 6 or 7 #hashtags each and several links to spam sites.

The spam isn’t surprising—any new communication format online gets abused by spam, but usually we’re able to block it out, either by not following spammers on twitter, or relying on Google’s algorithms to block it for us. Scoopler, although I’m sure they’re trying valiantly to do so, hasn’t done it yet, although they are a brand new search engine.

The Verdict:

An OK mix of the real-time flow and a filtered, more popular resulsts list, but ultimately not very satisfying in helping me find anything. If you like studying the online zeitgeist, this is better than Google Trends, in most respects.

Topsy

xinjiang-e28093-topsy-search-results_1247063036160-w595

Topsy is actually quite fantastic. Want to see how many people are talking about a particular article or several articles on a specific subject? Topsy does it beautifully, because it ranks the results using some specific algorithms. Certain tweeters get an ‘influential’ rating, based on a series of criteria that surely includes the number of followers, the amount of re-tweets they garner, and so on. So as something new and trending pops up online, you can get an immediate sense of ‘authoritative’ search that some of the other real-time engines lack. I’d like to know more about how Topsy evaluates its ‘influential’ criteria, but they’ve definitely hit on a good idea.

The other thing I’m enjoying about their site is the following: it gives me the satisfaction of google’s indexed, archived, older searches, only applied to Twitter. I can go through Topsy and immediately find my own profile, and then get a sense of how many other people tweeted the same links I did. Authors can also use this to find articles and blog posts of theirs that other people have been talking about—while Twitter’s main search lets you do this too, Topsy makes it more elegant and far better organized.

In the end, when it came to my breaking news search (Xinjiang), Topsy was much like the others, giving me general links to big newspaper stories with lots of authority, and little else. Perhaps I picked one that didn’t have much original content outside of news agencies—it was coming from China, after all, where there’s still plenty of censorship. Breaking stories elsewhere have a much different feel to them—faster links to amateur video on youtube, and so on.

The Verdict:

Out of all of these, Topsy is the one search engine I can see myself going back to, and using it to supplement a google search. I’ve already used it about five times today, which is the only thing that matters, in the end.

Collecta

collecta_1247062969250-w595

Collecta is another startup (all of these are startups, of course) searching twitter, this time with an interface that lets you put into larger focus individual tweets, and lets you run several searches concurrently.

I liked both of those features, and the fact that it pulls comments from blogs, as well as blog posts and flickr photos. There was no smooth interface action where the newest tweets cause the old ones to elegantly slide down (something that FriendFeed does, which gives it a more ‘real time’ feeling, at least aesthetically), and there was still, as is inevitable, a ton of spam when it came to any ‘hot topic’.

Also, leaving the site gave me a message about disconnection, and going back to the site left me with a big wait before all my searches started again.

The Verdict:

Again, a nice interface, but nothing I’d use on a daily basis. I’m starting to notice a trend here—slight variations on the main twitter search engine—and I think a large part of this is because all of these search engines are just so damn new.

Almost.at

almostat-conflict-in-xinjiang-china_1247063025754-w595

Here we’ve got a site that isn’t really a search engine, as much as a way to follow a series of ongoing events through real-time chatter online. You can choose major events from the left toolbar, and you get a series of twitter updates, flickr photos and youtube videos, and various links displayed.

What’s nice is that the timing is broken down minute-by-minute, so you get a nice clean view of the traffic for a specific event. You can also head back in the timeline to see what the chatter was at a previous time. Once I did this, though, I found it tough to get back to the ‘present moment’ and started forgetting today’s date.

The timestamping for the youtube and flick results also seemed to be non-existent.

The Verdict:

A cool idea for following live events, and the actual event choices (at least during the beta) are moderated, so the selection is small and relevant. The execution isn’t anywhere close to perfect, but the concept is one of the most interesting. Following protests using only #hashtags or twitter’s main search can become a spam-fest or a messy confusion, and almost.at’s concept of choosing specific ‘events’ and following just those is a nice way of organizing the information.

To Sum Up….

Pretty much all of these services are in beta, and they all show it. In the rush to capitalize on the new flood of twitter data that’s out there, it’s inevitable there are going to be a ton of real-time search engines popping up this year, and if the rumours are any guide, Google will surely launch their own effort soon enough.

Lots of chatter about real-time search tends to paint Google as an old-fashioned search engine relying on caches and ‘yesterday’s internet’. Be weary about such claims—determining relevance when it comes to search results takes time, and for now, the best algorithms are ones that can’t get the job done instantaneously.

And What About Google?

Another cliche in the chatter surrouding these sites is Google’s lack of a real-time offering as of yet, and the fact that they’ve admitted they’re somewhat ‘behind’ on the issue. While it is surprising that so many competitors have moved into the field while Google has released nothing so far (but goes ahead and announces stuff like the Chrome OS months and months in advance, without any previews or substantial details), it’s all just a matter of tracking and indexing data, something we all know Google is very good at.

And most of the data being analyzed is simply coming from Twitter streams, combined with reading the timestamp of various other postings. None of this is exceedingly difficult when it comes to search, so what’s really going to matter is getting relevant real-time results, blocking spam, and creating an interface that’s easy to use but robust underneath.

So far, a site like that doesn’t exist (although I’d vote for topsy as the best one so far). Might Google jump into the market with the best implementation of real-time search yet seen? Although the field seems insanely crowded already, even just a cursory glance like we’ve done here shows that there’s massive room for improvement. Real-time search is still in its ‘alpha’ phase, and needs a big player to pull it out. One is probably going to arrive any day now.

(photo by flickr user fdecomite, used under a creative commons license)

Tags: , , , , , , , ,

Leave a Reply

Search website form
  • Start Live Chat: Available 24x7 - 365"

    Start Live Chat Button

    Chat Live with an internet marketing expert

Categories

© 2008-2017 SEO Design by LoveClients - The only 24x7-365 Live SEO Company Online.