Formerly SpringBlog

Sunday, March 20, 2011

Stumpedia

Morville's "Sociosemantic" web is a way to look at the internet in a bottom-up fashion. The website Stumpedia is a perfect example of what Morville envisions when talking about social-semantic webs and folksonomies. On Stumpedia, instead of having an algorithm-based result provider (Google for instance), the users decide their own results. When you use the site to search for something a list of results will appear and you will be allowed to either bump or bury each result as you see fit. If something completely irrelevant appears, bury it! If you find exactly what you want, bump it! The idea behind the website is that before long, users, not algorithms, will be deciding the search results and the ability to find relevant information will skyrocket!
Human powered Search at its finest? Let's take a look:

Google uses an algorithm to base its search results. Often times this works great, however it is also easily manipulated and isn't always the most reliable. If everyone searching "Apple Sauce" is clicking on a link that takes them to a website to buy apple sauce instead of researching it, then the first result is going to be a website to buy apple sauce. However, on Stumpedia, you would find that people click on the site to buy apple sauce, click to "bury" it, and then "bump" a site they were looking for. This provides users with a more refined search experience (supposedly).

The truth is, with unmonitored "user" based search results, its even easier to manipulate the results. The main problem is that Stumpedia has practically no web traffic. It's interesting, but the number of people who use Google every day pretty much dwarf those visiting Stumpedia. When you search, you will find that many sites have only been given one bump. If you bury it.. it's gone. If you search "Evil Badgers" and bump sites that all pertain to the food guide pyramid, the next person searching "Evil Badgers" is going to have a hard time finding anything that actually relates to badgers while they are surfing through pages of "bumped" food guides.

There are positives and negatives to both approaches, a social-semantic, user based web; and your run of the mill top-down view. However, I believe that if user-based searching sites such as Stumpedia had more traffic to their sites (which would give them more user-based monitoring) then they may be much more useful and powerful in the future (like Wikipedia).

2 comments:

  1. Frankly, I see more weight behind the potential for exploitation than behind the potential for innovation. If you leave EVERYTHING in the hands of the users, you have a search engine based on anarchy; you have to assume everyone has everyone else's interests in mind if you want to assume that the system will work, and that's just not going to happen. I know I talk about Anonymous a lot, but I can just imagine what Stumpedia will look like after they set their sights on it... and they WILL attack it if it gets popular.

    ReplyDelete
  2. I agree with Todd (A Large Trout). You can't have the user decide everything unless each user has the interest of every other user in mind. It is a great concept don't get me wrong, but its easy to see people going in there and burying good info and bumping the bad and messing the whole system up. I'm not very informed on the website, and maybe they do have some security in place to prevent that, but you can't always trust the user to have good intentions.

    ReplyDelete