Posted January 7, 2008 11:25 pm by with 6 comments

Tweet about this on TwitterShare on LinkedInShare on Google+Share on FacebookBuffer this page

Mahalo, everyone’s favorite human-powered search engine and Jason Calacanis venture, has announced a few more “socially” improvements to its site. One of the major updates leverages other social media sites; the other is frankly patterned after another.

Mahalo has expanded its Mahalo Follow tool to enable its users to post directly to several other social sites, including Delicious and Twitter. Also in the Mahalo Follow Sidebar are tips/tricks that accompany selected sites, including Gmail and Twitter. Vanessa Fox samples the sidebar’s suggestions and shares her results, which aren’t particularly helpful.

The second update is adding stub articles to Mahalo. If you’ll recall, Mahalo has always had articles and link lists for its most important subjects, created by its (supposedly vetted) editors. But now, admittedly patterned after Wikipedia, Mahalo has added the capability for anyone to add short articles and recommend links for terms that have not yet had full articles written on them.

Now I know what you’re thinking: this is the chance all the black hatters have been waiting for. now we can all start gaming Mahalo, the much-vaunted search engine that was supposed to kill SEO.

But before we all start our evil laughter, Jason Calacanis has already thought of this:

Now, you’re of course saying “won’t this lead to tons of spam?” and the answer to that is “yes it would *if* we didn’t have 50 full-time people watching the incoming links.”

Considering Mahalo’s current market reach, I’m not sure if that will be enough. As soon as someone figures out a bot for this, Mahalo may be inundated. My blog, which has far less traffic, still sees thousands of spam comments a month.

Then again, considering Mahalo’s current market reach, they probably won’t have that problem.

  • You can check for youself to see how much spam is getting in at the “user created stubs” category page at Mahalo here:

    Nothing horrible is making it through, although some of the stuff is lame or “just ok.”

    Our thought is we can watch these stub pages and either kill them over time if other folks don’t add to them, or we can beef them up if they are good.

    Looks like we’re going to get 100-200 of these a day. That’s fairly easy for us to handle with one or two people. So, unless we get to like 10,000+ a day we’ll be fine.

    Also, if we do get to 10,000+ a day we’ll have a really big service and be able to afford to expand the program. 🙂

    Also, we can ban folks by IP, email address, and by URL. The reason why blog spam is such and issue is because a) people don’t require registration and b) people don’t ban IPs/URL/emails. If you do ban folks they give up quickly and find some other town to vandalize.

    At least that’s what I learned at Weblogs, Inc.

    best jason

  • Jordan McCollum

    Actually, lots of people that I know (including me) do ban by IP etc. But the vast majority of spam sent at the blogs that I write for comes from bots at a plethora of IP addresses, rather than actual people making multiple attempts from single IP addresses.

    Blog spammers are getting smarter.

    But like I said in the post, I can see that Mahalo probably won’t have a huge problem with spam for a while yet.

  • Well, when you require email validation AND you ban the links that folks are posting that REALLY helps a lot as well.

    If I’m a spammer and I have to buy a new domain name/start a new blog over and over again I’m gonna probably give up.

    Also, we can put a catchpa on the submit forms if a “spam storm” comes in.

    We’ve got it covered.

    The harder issue is folks coming in and making 50 good suggestions and THEN submitting a spammy site that looks like it’s good. Those are the super slick SEOs that we’re gonna have a harder time with!

  • Looks liek a lot of people are going to try to misuse this. Lets hope their spam protection measures to live up to expectations. Blackhat marketers are mostly a step ahead.

  • At this point, 50 people can most likely handle everything but as the project grows, I can definitely see a lot of people abusing this one, or at least trying to and getting away with it for a while, which will, of course, have a negative impact as far as user experience is concerned.

    Alan Johnson

  • Mahalo. Yawn.