Mahalo: The Long Route To Wikipedia

Sometimes you can’t help but want to see someone fail. I think I’m going to get what I want.

Yesterday SEO hater Jason Calacanis launched Mahalo (Hawaiian for Thank You), which he claims is the first human-powered search engine. Wait a minute hasn’t ChaCha been providing human-powered search for awhile now? And Mahalo may be a lot of things, but a search engine isn’t one of them.

The basic idea is you perform a search and are taken to an internal page on the site where ‘guides’ have “hand-crafted the best search results possible.” Of course they’re only hand-crafting those results for the most popular queries. At the moment that means 4,000, but the goal is to have results for the top 25,000 queries. What that ultimately means is the vast majority of searches will have no human-powered results.

The majority of results on Mahalo will default to displaying organic results from Google with a few sponsored results thrown in. Apparently AdSense will be the way the site will be monetized for now. Let’s see, a site that scrapes Google results and uses AdSense. Can you say MFA? The non-results pages are kind enough to list related pages. My search for ‘web design’ offered me a highly related page for Jeans and and a page for Chlo© which reads.

There is currently no text in this page, you can search for this page title in other pages.

Most recommendations for related pages are of the same high quality. And while ‘iPhone’ does yield results, ‘iPhones’ does not. It does at least tell me the ‘iPhone’ page is related. That’s one fine search algorithm.

Before getting on to the few pages that do have results I’ll mention a little bit of good. I actually like the interface. It’s clean and simple in the tradition of web2.0 and while I’ve seen some criticize the logo it works for me. I also like that the Guides have their own bio pages, which show the pages they’ve written and lets you know a little about them.

The pages with results do offer a Guide Note and Fast Facts in the right sidebar, which I admit is some good quick info. Lower down in the sidebar will be some user submitted and editor approved links. There aren’t many at the moment, but in all fairness it’s only been a day. Each result page will also link to a message board so users can discuss the search. Most boards are empty at the moment, but again it’s only been a day. I did find this gem in the iPhone message board.

Test message, just wanted to explore Mahalo..

I hope all the other messages are as helpful to my search.

So what about those hand-crafted results? Hard to call them hand-crafted. With very few exceptions every search I tried lists the Wikipedia entry as the number two top result. I wonder how long it took to find that link. The page for George Bush showed the Wikipedia page first. The home page for the White House is apparently less relevant. Wouldn’t the George Bush bio page on the White House site be most relevant? I guess not, though it is listed in the biography section along with George Bush’s biography page on IMDB (Internet Movie Data Base).

If you do a few searches you’ll find that the top seven results are mainly to obvious sites like the Wikipedia and IMDB. The kind of sites you would just go to in the first place without needing to search at all. Hand-crafted indeed. At least it will be easy to build the other 21,000 pages as it seems results are being pulled from just a handful of sites.

How do these finely crafted results stack up to the algorithm at Google? Five of the top seven results for iPhone are listed on the first page of results at Google. A sixth ranks #12 at the real search engine. I’m so happy I checked Mahalo first or I would have found the same pages.

You also have to wonder how they’re coming up with their top searches. Frank Thomas has a page. Frank Thomas is a great baseball player, someone who’ll be going to the hall of fame more than likely, but a top search he’s not. And while Frank Thomas does have a page of results, Thomas Jefferson does not.

Could anyone find the site useful? I suppose if you were back in junior high and needed to write a paper it would be nice to have a bunch of links all in one place. If you were new to the internet and had never heard of popular sites like the Wikipedia, Yahoo, or the New York Times you might find it helpful to have links to them all in one place. For everyone else probably not. If you’re going to present me the Wikipedia as a top result for most every search how long should it take me to start my search at the Wikipedia directly. These aren’t particularly good results. They’re simply the most obvious ones. Good to know the Guides are such experts when it comes to research.

As for the man behind the site, the one who has told the world that seo is worthless and that 90% of SEOs are snake oil peddlers, he’s hypocritically enough redirected URLs to search friendly versions and included ‘link to this page’ code you can copy and paste. Meta keyword tags are also included, useless though they may be?

Mahalo offers little in value from what I can see. For a limited number of queries it’ll point you to pages on some sites you already know. It strikes me as a directory of pages from some of the most popular sites online. Most of the time it will graciously present Google results and hope you click on the ads instead of the organic listings. The one truly remarkable thing about Mahalo is that it managed to convince VCs an MFA site was worth funding.

Here are a few more opinions of Mahalo. Most against, but a few lukewarm wait and see posts. My two favorites are Todd Malicoat’s “How Sequoia Venture Capital Will Pop the Bubble of Web 2.0 – Mahalo or MFAhollow?” and Brian Provost’s “Mahalo: That’s Hawaiian for 8 Visitors A Month.” Brian wins for the best post title, though MFAhollow is brilliant.

Download a free sample from my book, Design Fundamentals.

Leave a Reply

Your email address will not be published.

css.php