Google didn’t invent the long tail, but it certainly enabled it. With its original algorithm, Google enabled searchers to find information on the most obscure search terms with surprising accuracy. Adwords enabled advertisers to target obscure keywords, and Adsense enabled publishers to attract advertisers automatically for highly specialized content. Now, a new search engine, Mahalo, is aiming at short tail searches by attempting to human-edit the top 10,000 searches to produce the best possible results [Wall Street Journal]. According to founder Jason Calacanis, this modest number of search terms account for about a quarter of all searches. Searches for which no human-edited results are available will be filled by Google.
Initial reports are that the search results that have human editing are, indeed, better than what algorithmic engines serve up. That’s no huge surprise, I suppose. I suspect that the spammier the search terms (think travel terms, for example), the better the human results will look. The real question in my mind isn’t how the company will create its initial batch of common searches - reportedly, they have already created 4,000 - but how they will keep these entries current. It’s easy to imagine the human-edited links becoming stale, and rarely reflecting recently added content. Even in its heyday with its massive complement of volunteers, DMOZ.org was never able to keep up with the growth of Web content.
Perhaps, though, this approach could work for a particular kind of search - those searches where one is seeking basic resource information that isn’t likely to change quickly. And we do like the informative way some results are presented - a search for “iPhone“, for example, returns a “Mahalo Top 7″, Google News results, selected blogs and forums, a couple of shopping links, reviews, and even competitors - all in all, a very nice and useful result. (It’s not so cool if you have a great resource that doesn’t happen to get listed by the editor - and there’s no obvious link to “more search results” for the term. Unlike traditional search, for their edited terms what you see is what you get - you can’t page down through the results looking for more.)
This isn’t a volunteer effort - Mahalo is using paid editors to create their listings. They do try to tap the wisdom of crowds, though, by letting users suggest links and listing the top suggestions. All in all, we find Mahalo to be an interesting new venture. We’d like to see traditional search results easily accessible, though, even for human edited terms. We think having to go to another engine to get traditional SERPs is a disincentive to using Mahalo. Nevertheless, its simplified, distilled, and organized links may well appeal to users looking for easy access to the major resources on a topic.
Calacanis is a darling of the blogosphere, and needless to say the launch is gathering plenty of attention. Danny Sullivan has a lengthy and thoughtful analysis of Mahalo. SERoundtable has some sample searches that illustrate the difference in levels of human editing. David Berkowitz of SearchNewz has some issues with Mahalo, notably that the results seemed hard to differentiate due to lack of descriptions and the potential arbitrary choices of unknown human editors. Chuggnutt didn’t find the concept overly original, and doubted it would get much use due to its small number of search terms. Sam Huleatt of Leveraging Ideas thinks Mahalo has already succeeded simply because of the network effect of all the attention it has received.
Add this post to: del.icio.us - Digg it - Stumble it - Furl - Yahoo MyWeb No Comments so far
Leave a comment
Leave a comment
Line and paragraph breaks automatic, e-mail address never displayed, HTML allowed:
<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>