Sanity check: Can Mahalo save us from Google, Digg, and Wikipedia?

While the rise of Internet tools has democratized the publication process and made information more freely accessible, that process has also had negative side effects, as seen when you look critically at Google, Digg, and Wikipedia. See why Mahalo could help reverse some of the damage.

Let's get one thing straight from the start -- before you click send on any hate mail -- I am not going to predict that Mahalo will unseat Google as world's top search engine, or drive Digg out of business, or replace Wikipedia. What I will argue is that Mahalo has an opportunity to save us from the increasing ineffectiveness of Google, Digg, and Wikipedia in one critical area.

How IT professionals use the Web

Before there was Internet search, IT professionals typically had a library of books and multiple stacks of magazines that they would flip through -- sometimes frantically -- to find the critical information they needed to do their jobs.

However, if the Web has done nothing else, it has drastically reduced the amount of paper and shelf space that IT pros need, because since the late 1990's, they have been tossing out most of their books and magazines (hopefully into recycling bins) and turning to the Web as their first stop for research and technical content.

Today, most IT pros primarily use the Web for two critical job-related activities:

1.) Problem solving

2.) Information gathering

When it comes to solving problems, one of the most common scenarios is to copy and paste error messages or keywords into Google, which returns a list of discussion threads where other users who have encountered the same problem have posted and received solutions. In fact, the TechRepublic forums get a couple million pageviews every week from Google users doing these types of searches.

The other critical type of search for IT pros involves information gathering about technologies, products, and concepts that can potentially help them improve the business. I'll give you three quick examples: SOA, ITIL, and WiMAX. Today, if you enter any of those three into Google, the first search result you get is a Wikipedia entry. The rest of the first page of results is made up of official sources (which most users could have found on their own), a handful of semi-useful sites that have gamed the Google algorithm, and paid search ads. That's not a very compelling mix of content, and that's where Mahalo's human-powered search engine offers an effective alternative.

The problems with Google, Wikipedia, and Digg

For problem solving by looking up error messages or keywords for common tech snafus, Google is amazing because of its reach across so many millions of forums and pages on the Internet. This is where you see the power of the Google algorithm at work. No human-powered search engine can achieve that kind of scale. For IT professionals, this is Google's greatest value, and nothing is going to replace it anytime soon.

However, when you need to do research and information gathering for tech topics, the quality of Google search results has always been a gamble. Sometimes you quickly find useful stuff, while many other times you have to flip through multiple pages of results to find only mildly-helpful stuff. Plus, Google's results are arguably getting less effective with time. The USC Center for the Digital Future found that only 52% of users trusted the information from search engines in 2007, down from 62% in 2006. For 2007, 49% of users trusted Google's results.

Also, as mentioned above, Wikipedia often dominates the top of Google's first page of search results. Wikipedia can be useful if you don't know what WiMAX is, for example, and you just want a quick definition. However, if you want to do any additional research (like finding white papers and in-depth articles), Wikipedia is usually not much more useful than Google because it is plagued by wild inconsistencies and inaccuracies. Even founder Jimmy Wales has admitted to Wikipedia's serious quality control issues.

Wikipedia is wide open for anyone to edit, but it is also easy to dominate by a few self-appointed rulers, some of whom have a clear axe to grind or simply don't know the subject matter but have the time to follow and control the edits to the pages. The result is a body of work on Wikipedia that is presented as facts but is riddled with errors.

For one quick example, take a look at the TechRepublic Wikipedia entry. It lists TechRepublic as a "technology news site," when it's actually an online trade publication and social community for IT professionals. Wikipedia lists Stephen Howard-Sarin as the "creator" of TechRepublic. In fact, TechRepublic opened its doors in 1999, and Stephen didn't have any direct association with TechRepublic until 2005. It also lists Stephen as the "leader" of TechRepublic, when he's actually the VP of product development at CNET Networks for the group that includes TechRepublic, ZDNet, and BNET. These inaccuracies have been up on Wikipedia for over a year. I could have edited them myself but I didn't, because I wanted to wait and see how long they would last. I'm still waiting.

I also lump Digg with Google and Wikipedia in this category of information gathering. When IT pros go to Google and Wikipedia, they already know what they're looking for and they type it in the search box. When they go to Digg, they go to see what other users are talking about, what's hot, and to find information about technology that they didn't know they needed to know.

Back in 2006, I regularly checked Digg. For the past 12 months, I can count my visits to Digg on one hand. It used to be a great place to find the top stories in tech from various publishers, large and small. But Digg no longer surfaces enough relevant stories to be useful to me, and it's become clear that it's very arbitrary which stories will make it to the top and which ones will get buried. For example, the same story will get posted four times and three of them will get less than five diggs, while the other one will get 400 diggs and make it to the front page of Digg. There's obviously a small group of users who control the Digg algorithm, and they don't do a very good job of picking which items to push to the top, or at least they don't do a good job of picking stuff that's timely and relevant for IT anymore.

Can Mahalo make a difference?

Like many users, I've been losing faith in Google, Digg, and Wikipedia for research and information gathering. Google is still great for looking up specific problems or finding specific things that you already know exist. Wikipedia is still fine for getting quick definitions. And Digg, well, I'm not sure what Digg is good for anymore. Unfortunately, none of them are consistently effective enough for the serious information gathering that IT professionals need.

Mahalo is a tool that has the potential to succeed where these three are currently failing. Mahalo is a search engine powered by people rather than algorithms. It was founded by Jason Calacanis, who previously ran Engadget and Silicon Alley Reporter. He launched Mahalo during the summer of 2007, with the initial goal of creating human-generated search results pages for the top 50,000 search terms on the Web.

I'm not sure what Mahalo's time frame is or how they are measuring the "top" search terms, but Mahalo editors and users have already created over 25,000 pages. From what I've seen so far, they are producing some quality pages that can serve as reliable resources.

For comparison sake, let's take a look at the pages for WiMAX from Google, Wikipedia, and Mahalo:

Google has what you'd expect -- Wikipedia is the first entry, the WiMAX Forum (the official governing body for WiMAX) has the second slot, and then it has a mix of other publications and articles. The Wikipedia entry is quite detailed, but as with any Wikipedia piece, I always wonder how many inaccuracies are contained within all of that text.

The Mahalo WiMAX page (see the screenshot below) has a quick synopsis of WiMAX at the top of the right column and editor-selected "Top 7" links at the top of the left column. The number one link is the Wikipedia article, and then it points to six more links from established media sources. The Mahalo page then goes on to include links for news, background info, blogs and discussions, and related companies.

By the way, Mahalo originally had its entry spelled "WiMax," so on Saturday afternoon, I made a quick note about the misspelling in the discussion to the entry. Later that evening, an editor spotted my comment, replied, and quickly fixed the page. This is evidence of both the "greenness" of Mahalo and the fact that it is currently run by some diligent editors.

The reason why I think Mahalo has the potential to succeed from this information gathering perspective is that it taps into the power of the crowds on the Internet while also providing clear, transparent leadership for its entries -- the latter is where Digg and Wikipedia fail. The Mahalo formula provides strong search results pages that aren't manipulated by a small group of users or by gaming the algorithm. As such, Mahalo will rise or fall on the strength of its editors and the level of participation that it can inspire in its community.

Ultimately, this whole issue comes down to trust and whether you trust algorithms, editors, or the masses. I trust algorithms for finding needles in haystacks. I trust the masses to contribute and tweak content. And I trust editors to organize, point out opportunities, and manage quality control. That's the formula that Mahalo appears to have adopted, and it has already given the product a leg up on Google and Wikipedia for some search terms.

Mahalo has also shown the ability to respond quickly. For example, this weekend when news broke that Yahoo was going to reject Microsoft's takeover bid, Mahalo was the first place where I saw an alert about it -- in a link just under the search box on the Mahalo home page. This is where Mahalo has a chance to compete with Digg. If it can quickly create pages like the one for Yahoo rejects Microsoft offer and surface them prominently, then it could become a place that I go for not just information gathering, but to take a quick glance at the latest buzz in tech.

Right now, Mahalo gathers 5-7 recent links in each category and puts them under each of the category names on its front page, but when you actually click the category (e.g. Technology) and expect to see a a full list of the hottest and newest entries, instead you get a general category page (see below). If Mahalo could use one of the columns on that page for the hottest and newest pages in that category, then it could create a Digg-like effect for these pages and give them some broader usefulness.

I would encourage IT professionals to check out Mahalo, but keep in mind that it is still in its infancy and it has very few business technology entries. I'd like to see entries for hot biz tech terms such as Software-as-a-Service, ITILv3, blade servers, and WAN acceleration, just to name a few off the top of my head. If you sign up for Mahalo Greenhouse, you could write entries for these types of topics and possibly even get paid for it.

If you don't want to write whole entries from scratch, you can also sign up for a Mahalo profile and then recommend links. I created a profile on Mahalo under Digitalcowboy9 (see below). If you set up a Mahalo profile, feel free to add me to your friends list. I'll be interested to see if we can make this tool useful for those of us interested in IT and business technology.

Do you think a human-powered search engine like Mahalo could have a positive impact on information gathering on the Web for IT pros? Join the discussion.