[Foundation-l] Vector, a year after

Erik Moeller erik at wikimedia.org
Fri Apr 1 02:24:48 UTC 2011


2011/3/31 Amir E. Aharoni <amir.aharoni at mail.huji.ac.il>:
> The Vector skin, the main product of the Usability Initiative, was
> deployed on Wikimedia projects in April 2010.
>
> Quoting usability.wikimedia.org: "The goal of this initiative is to
> measurably increase the usability of Wikipedia for new contributors by
> improving the underlying software on the basis of user behavioral
> studies, thereby reducing barriers to public participation."
>
> In the year that passed since then, did anyone measure whether the
> usability of Wikipedia for new contributors increased?

The usability initiative was accompanied by three qualitative studies:

http://usability.wikimedia.org/wiki/Usability,_Experience,_and_Evaluation_Study
http://usability.wikimedia.org/wiki/Usability_and_Experience_Study
http://usability.wikimedia.org/wiki/Usability,_Experience,_and_Progress_Study

Our studies validated that the changes we made did indeed by and large
have the intended effect of simplifying the experience of new users.
With that said, the aggregate editing trends continue to be troubling.
See, for example, this page for a comparison of active editors across
languages:

http://stats.wikimedia.org/EN/TablesWikipediansEditsGt5.htm

.. and, of course, the editor trends study and the New Wikipedians
numbers. But, these larger trends aren't purely technical trends --
they're social trends as well, and it's entirely possible that no
amount of technical improvement is going to even make a meaningful
dent unless/until we also make progress on making Wikimedia projects
more open and more welcoming.

We haven't deployed some of the last-stage features of the project
yet. These include an in-editor outline of the article headings, a
tabbed view of preview/edit, and a default collapsed view of
templates. Making template collapsing work cleanly in all browsers and
for all document operations turned out to be very hard (due to the
wrangling required to make the browser's rich-text-editor behave
essentially like a beefed-up code editor), so we may not ever add that
feature to a wikitext editor (as opposed to a visual editor). The
other two features are likely doable with some more effort, but we're
prioritizing them against other improvements and the visual editor
effort itself.

So, in sum, 1) our qualitative research has shown an improvement for
new users, 2) the quantitative trends are troubling, and it's not
demonstrable that we've made a difference either way in the larger
trends (which aren't purely technical but also social trends), 3)
there's still quite a bit of code that we may end up picking up again
but that's not currently running on WMF projects. I'm happy that we've
done Vector as a first step, but it's just that - a first step.

-- 
Erik Möller
Deputy Director, Wikimedia Foundation

Support Free Knowledge: http://wikimediafoundation.org/wiki/Donate




More information about the wikimedia-l mailing list