Steve Bennett wrote:
On 2/11/07, William Pietri william@scissor.com wrote:
At least in my professional life, I'd user-test all of the suggestions to see how they play in Peoria. Then the ones that passed that I'd to A/B testing with a modest percent of the traffic to see which ones were clearly better by the numbers.
I'd be wary of applying the results of testing on smaller wikis directly to Wikipedia. In this case, "Recent changes" probably performs much better on smaller wikis.
Sorry if I wasn't clear; I'm suggesting doing typical user testing, where you use the real thing or simulations as close as you can get. I agree that smaller wikis may not be good substitutes.
Historically, are either of those things done with Wikipedia?
Usability testing on Wikipedia? Not that I know of.
Interesting. Is that a conscious choice? Or more a historical habit?
Thanks,
William