arghyle

Considering how new and unproven Mahalo is, there has been a lot of news about it. It’s a cool idea, they apparently have a lot of money, and Jason Calacanis is a pretty well connected guy, so it’s no surprise that it’s looking very strong despite how new it is.

If you are not familiar with the site, you can go there and figure things out pretty quickly. Basically it is a human-powered search engine based on a wiki. In retrospect the project should be no surprise since Calacanis (on his blog at least) has been extremely interested in Wikipedia and SEO, plus from what we saw of his work on Netscape, we know he likes to employ people-power when possible. Add up all of those, throw in some cheesy Hawaiian graphics, and you have Mahalo.

Right now the site is still in alpha, which means that the editors have only been able to put together a limited number of SERPs. They seem to be using a top-down approach and getting the top search terms and filling in pages about those (I think I read that they have about 4000 right now with a goal of 100,000) and with 28 “guides” and user recommendations, this number should be increasing rapidly (I would not be surprised if they were doing a 1000 a week. 28 guides x 5 days x 10 pages/day/guide = 1400…). Even so, if you search for something that is not popular right now (Apple, Paris Hilton, etc) you probably won’t find it.

It is very interesting to see how they did the results pages. Here is one mentioned in their blog- Paris Hotels- that should give a good idea of what Mahalo’s goals are. Going from the top down- we have:

• Top 7- the best search results. Basically what you want to get from Google once you filter out the crap…
• Specific hotels separated by price (luxury/moderate/affordable/hostel) with phone number, recommendation and Google maps link
• Related searches within Mahalo
• Also- you can recommend a link or email someone the page

So while you can get everything here by going to another search engine, Mahalo puts in all in one page, throws you some good links, and should save you some time because the results are pre-filtered by a (hopefully) capable human.
What are the problems with this?

1. The lists are far from complete. For example, I stayed at a great hotel in Paris a few months ago, it’s not listed, and just looking at the list I can see a lot more that are missing. [To be fair it says the page is 75% done, but all the pages are that way.]
2. The guide (Rupak) may not be an expert in Paris hotels. Sure he can get links from Google or read Frommer’s online and get some recommendations, but what he is listing is nothing more than the site’s he likes and filtered results from other sites.
3. How fresh is this page? Mahalo is brand new, so this page is probably just a few weeks old, but once Mahalo has thousands more pages, how will things stay updated. In fact check this out- the page has not been touched since 22 March. That’s over two months…
4. The discussion page is dead, and the functionality here is limited.
5. This isn’t a problem, but it’s worth mentioning that I can’t edit it- http://www.mahalo.com/Paris_hotel&action=edit, so we’ll have to trust the guides.

Number three seems especially crucial to me. Wikipedia stays up to date because thousands of people are overseeing pages and put lots of time into making sure they are updated. Plus anyone can go in and make a quick change if they notice something that needs reworking. This is not the case with Mahalo and if all the resources are going to new pages I don’t see how they can keep up with old ones.
So, as a user, what can I get from Mahalo? I am fast and capable with other resources, so probably not much. It can save me time by aggregating popular links and showing recommended pages from other users, but is that it? If I was not as technically proficient (for example, “what could my parent’s get from it?”) then it would clearly be a bigger help. They would be able to get quality links without spam, but their search would be limited to the biggest sites. What about smaller blogs about Paris that could offer great firsthand experiences and recommendation? Those will probably never make it because Mahalo is all about getting us to the big resources and filtering out the rest.

At this point it’s hard to say whether Mahalo is optimizing or dumbing down web search, but it’s clear that something had to be done. There is more than enough room for a few more players in the search engine business and even if a site is getting only a small percentage of search traffic they can still be highly successful.This is, if nothing else, a novel approach and one that will be a big hit with some people. It’s like Cha-Cha done right and without having to directly deal with someone else (which is the whole reason we are on the internet in the first place, right?).

Trackback URI | Comments RSS