Judging the Reliability of Wikipedia Through Colour

Submitted by Dean Giustini on August 22, 2007 - 13:22

[img_assist|nid=72|title=Wiki-colour|desc=|link=node|align=left|width=61|height=100]My spies tell me that Wikipedia will be testing a color-coding tool to mark questionable content on articles. The tool - tested within Wikia first before scaling up to Wikipedia entries - is not the first time advanced statistical analysis has been used to assess the reliability of articles within Wikipedia.

The color-coding scheme will give Wikipedians a sense of how many times sections of articles have been revisedas a means to see what sections are the most disputed and controversial; by extension, entries virtually unchanged over longer periods of time will indicate consensus of the crowd.

Color-coding will be a way for Wikipedia to address concerns regarding content reliability. With tools for marking questionable content, the reputation of Wikipedia should improve. Articles with the least amount of revisions will have a higher trust or reliability factor, or will they?

What do you think Open Medicine readers? 

Other relevant information:

Wiki-related developments include Search Wikia and Similpedia. See: The Wiki Toolbox: 30+ Wiki Tools and Resources

In a Chronicle podcast, Virgil Griffith, a graduate student at the California Institute of Technology, talks about WikiScanner - used to spot deletions of potentially unflattering information about groups, or people trying to hype themselves, organizations or bash their enemies.

Commenting on this Blog entry is closed.


In Wikipedia May Add Color-Coding to Mark Bad Entries (by Kristen Nicole, 8 August 2007), it's stated that those Wikipedia entries "with the least amount of revisions will have a higher trust factor". The USC Wiki Lab has a page (last edited on 17 August 2007) about a new "Wikipedia trust coloring demo". The demo contains only a few hundred pages. One can click on "Random page" to get to a page "based on the Wikipedia dump dated February 6, 2007".
On August 5, I selected a sequence of 10 successive "Random pages".
Although only a very small sample of 10 pages was examined, one
interesting finding was this one: the pages that showed the most
obvious (orange) highlighting were those for key words that had
multiple meanings. See: Trustworthiness of Wikipedia entries (August 6, 2007).

That sounds pretty cool. They have been taking a lot of flak due to their nature. I just wonder how well they are going to pull this one off.