On 28 September 2010 23:19, Michael Snow <wikipedia(a)frontier.com> wrote:
On 9/28/2010 4:41 PM, Risker wrote:
Aside from the point already made regarding the desires of projects
other than the English Wikipedia - I guess I struggle to see what's so
demotivating about the prospect of a feature being "permanent" in the
sense of being written into MediaWiki code while the English Wikipedia
community still has the full ability to decide not to implement it on
that project. Is it the potential of having to withstand continued
political battles seeking to have it activated? That would implicitly
acknowledge, at the very least, that there is some need not being met,
meaning that alternative solutions are required.
Further improved trials might get us closer to such solutions, and we
should keep experimenting where we can. I'll reserve comment as to
whether we have the right balance between urgency in tackling serious
problems and exercising patience to maximize our chances of success.
This trial was poorly organized and badly timed, as it was the third major
deployment of new software into the UI in just over a month, and the dust
hadn't settled on either of the other two when this came along. Rushing a
second trial in to place before the data is analysed from the first one, the
bugs are properly fixed and tested, and agreed-upon criteria are
established, will likely wind up with exactly the same result we had this
time. Remember, my first post asked for more time before the next trial so
that it can be done properly. Give us enough time to construct a proper
trial that can genuinely explore this tool, and help us to line up timely
analytical resources needed to make an informed decision about the tool's
value. "Test early, test often" doesn't really work in a project where 98%
of the regular users have no idea what a bugzilla is, let alone how to file
one.
I don't
often write to this list, and I realise that I sound fairly
negative
in this thread. The fact of the matter is that I
personally entered more
articles into the first trial than any other administrator (20% of all
articles involved), that I actively and strongly encouraged other
administrators to do so as well, that I pushed hard to ensure that the
largest number of editors possible received reviewer permissions, and I
was
one of the few people who trialed the version on
the test wiki in the two
weeks before it went live, finding a significant number of problems (some
of
which were addressed in advance of the release).
I was also the person
who
made sure that the WMF spokesperson with respect
to the trial was in
agreement with the prior stated position of the community, and that the
feature would be turned off if there was not clear and unambiguous
support
for it at the end of the trial, just to make sure
we were all on the same
page.
So, yes...right now I (and several other administrators who were very
active
in this trial) are very disturbed at what has
happened here. We felt
there
was a clear criterion for continued use of the
tool, which was worthy of
our
collective time, energy and powers of persuasion.
With that in mind, it's
almost impossible to consider developing a second trial, since it doesn't
seem like it will matter what criteria for continued use the project
determines.
From this characterization, my impression is not so
much that there is
a conflict between the community consensus and the developers; much
more, it strikes me that the extent of adoption and publicity for this
feature remains tremendously limited, so that it's extremely difficult
to say it's been adequately evaluated or speak of a consensus about it.
If the Wikimedia Foundation has fallen short, then, it's not by
disregarding the will of the community, but in a responsibility shared
with community leaders, of gaining attention from a wider group of
participants. I would guess that the vast majority of people actively
involved in the English Wikipedia still barely know any of what's going
on with this. That may be somewhat surprising to those of us who have
been involved in Wikimedia projects for a long time and think of this as
a perennial proposal for addressing longstanding issues. But I think not
only do people see this proposal through very different lenses, but for
many the lens is focused elsewhere anyway, and they are watching
different trees in the forest. Part of the challenge is figuring out
when and how it's appropriate to interpose "corrective" lenses to guide
people's energy in certain directions.
The on-wiki communication in advance of this deployment was pretty abysmal,
although no doubt part of the problem there is the lack of any consistent
communication process within the project as a whole. (In fact, posting
something on a half-dozen diverse but highly watched user talk pages is
probably the most effective way to get a message out to the largest and most
diverse group of editors.)
But yes, the fact that there are such divergent perceptions of the
objectives to be met by this software has definitely played a role, and the
software does give the impression of having been produced in accord with a
design developed by a very large committee. I rather doubt that most English
Wikipedia users, even those who've been involved long term in the flagged
revision discussions, realise that this software is being deployed with
variations all throughout the WMF empire. I'm not sure if knowing that
would make things better or worse. :-)
Risker/Anne