[Foundation-l] Pending Changes development update: September 27

[Risker]

On 28 September 2010 23:19, Michael Snow <wikipedia at 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