On Wed, Jan 9, 2013 at 1:15 PM, Jan Kučera <kozuch82(a)gmail.com> wrote:
Yes this was quick (and maybe little ugly) mockup, but
the idea is simple.
Someone from E3 team around here for comments?
Hey, E3 team member here.
I agree with Sage for sure. The real value of A/B testing this particular
change or a variation of it, is to help everyone decide what the ideal
change might be (e.g. red vs bold vs an icon) and what impact each of those
options have in comparison to each other and/or to a baseline. That way we
can still make judgement calls with the input of design expertise and even
consensus, but conversation can form around data, rather than just opinion.
If anyone wants to write a description of a test like this they want to
see, then you are always welcome to post it to
MediaWiki.org or Meta, and
get our attention. Being bold and adding it to the list of planned
experiments
https://meta.wikimedia.org/wiki/Editor_engagement_experimentsis
a good way to do so. :)
Even if we may not always have the bandwidth to do these experiments
ourselves immediately, we are always happy to gather ideas for future work,
since we regularly have time to run smaller experiments on the side. We can
also help you set up an experiment and work through the data
collection/analysis cycle.
Steven