New Wikia Search Released

[Logan]

The alpha version of wikia search has been released.

http://alpha.search.wikia.com/

Of course, before we start, we have no user feedback data. So the results are pretty bad. But we expect them to improve rapidly in coming weeks

Well, I have to agree the results are pretty bad. lol. I'll give them the benefit and wait to see how this evolves as people get involved.

One thing I noticed is as you navigate from the 1st page of serps to others, the additional results are shown on the same page. I didn't care for that on first impression.

Any hope for this?

[asprat]

The interface is definitely a little odd...I have to keep scrolling down over and over on my end. I should run it against a comparable search engine and see how it does result-wise.

[RayGoneFishing]

Originally Posted by Logan

Well, I have to agree the results are pretty bad. lol.

Definitely. There's still some serious work to do. Although, the results seem to be a bit better with more specific search terms. For instance, results for [guatemala] are fairly useless, [antigua guatemala] is a bit better, and [antigua guatemala spanish schools] is almost acceptable.

Originally Posted by Logan

One thing I noticed is as you navigate from the 1st page of serps to others, the additional results are shown on the same page. I didn't care for that on first impression.

That irritated me too. Why not simply open up a new page?

Another thing I didn't like is that websites in various languages come up in the search results, with no apparent way to set a preferred language. A search for [guatemala], for instance, returns English, French, Spanish and German pages on the first results page.

Originally Posted by Logan

Any hope for this?

A bit early to say, but I'm skeptical. It may find its niche and become a useful research tool, but I don't think it will be competing with the major search engines anytime soon. But that's probably not their intention anyway...

[Sporkman]

Here's a good Forbes article on it:

http://www.forbes.com/2008/01/09/wik...rtner=yahootix

[Cratima]

I like the 'mini article" idea and the photo matching "keyword" idea.
I can't imagine a revolutionary design idea for search engines.
As for the relevancy of the results found, I think they are comparable with Google, yahoo and msn.
I like the fact that wen loking for example for "web design" I also find web design directories, not 5 pages with only web design companies.

[Sporkman]

Originally Posted by Cratima

I can't imagine a revolutionary design idea for search engines.

One idea would be, after entering the keywords you're looking for, the search engine would ask you a few intelligent questions to get better info on what you're looking for, then tailor the results accordingly...

[RayGoneFishing]

Originally Posted by Cratima

As for the relevancy of the results found, I think they are comparable with Google, yahoo and msn.

That was not my impression at all. I found Wikia's results way inferior to the major search engines. But maybe that depends on what you search for.

Originally Posted by Cratima

I like the fact that wen loking for example for "web design" I also find web design directories, not 5 pages with only web design companies.

Again, that's not what I'm seeing. If I search for [web design] on Google (regular search on google.com, not a local search), I see a Wikipedia article on web design at #1, an about.com article about web design at #2, a template site at #3, web design companies at #4-8, a directory of web design companies at #9 and a web design course at #10. So only half of Google's top 10 are actually web design companies.

But then, if you're only interested in web design and not design companies, maybe you want to change your query to [web design course] or [web design tutorial] or [web design principles]? The better the question you ask, the better the answer you get

[Logan]

I agree Ray, seems there is some consensus regarding that. I'll at least give Jimmy Wales credit for calling a spade a spade. Maybe lowering expectations, but I think what he is saying about the results being low quality is being pretty straight. Good article sporkman -

"I've been trying to tell everyone for months that Wikia would suck when it starts, but that it's just the beginning,"

As a result, the first required step, Wales argues, was to unwrap the engine for the public, raw edges and all. "The concept is to have the community providing the human intelligence. And you can't gather data until you have people searching," he says. "So it's kind of a chicken-and-egg type problem."

So, maybe an opportunity for us to see the 'raw edges'. I have to admit it would be interesting to see the raw edges of google, yahoo or other search egnines.

Wales says that more than 5 million Wikia searches have been performed in the last two days and 5 thousand users have registered on the site.

Those numbers do seem to reinforce this regarding community participation. I'm curious what percentage of those that register participate. A small percentage I'm thinking. A percentage who are heavy users, a percentage of spammers, and a percentage of those who don't contribute ... imo.

[RayGoneFishing]

Originally Posted by Logan

I'll at least give Jimmy Wales credit for calling a spade a spade. Maybe lowering expectations, but I think what he is saying about the results being low quality is being pretty straight.

Yes, definitely. At least he did not try to sell us the umpteenth Instant Google Killer.

As Wikia's concept depends heavily on community participation, we'll have to wait and see how the community develops. If Mahalo is any indication, though, it won't be easy for Wikia to reach a critical mass.

[Logan]

well that didn't work.

Wikia Search Closing

we will be closing the doors on the Wikia Search project (as of March 31, 2009) and will be re-directing and refocusing resources on other Wikia.com properties