XTools/Authorship/diq

The Authorship tool provides editor authorship attribution, measured by character count, excluding spaces. It is powered by the WikiWho algorithm.

From the form, you can get authorship stats from the latest revision of a page, or provide a revision ID or a date.

How it works

To get these statistics, we use an external service called WikiWho. It uses a token-based system in its algorithm. A token is generally a word or group of syntax. For example {{foo}} can be broken into tokens for the brackets ({{, }}) and the word foo. Each can have its own author. So if the original author added {foo}, forgetting the outer brackets, the second editor gets attribution to the {{ and }} tokens, but the first editor retains attribution to foo.

It sounds like could result in inaccuracies, but a gold standard test showed it to have 95% accuracy for English Wikipedia, and should in theory work for all wikis.

To make the system easier to understand, XTools incorporates the number of characters in the tokens in our figures.

Xulasa

The Summary section lists the article in question and the revision that was scanned, as well as the number of unique authors and total characters they have added.

Authorship

This section lists each user by the number of characters they've added to the page. Only the top 10 are colorized in the graph, the rest appear as a gray color.