[Foundation-l] Regarding Berkman/Sciences Po study

[Kim Bruning]

On Fri, Dec 09, 2011 at 07:51:06PM -0800, Dario Taraborelli wrote:
> So what went wrong?

Local consensus does not override global consensus.

* The research committee failed to adhere to applicable consensus. [1]
* In lieu of consensus, the research committee failed to adhere to or point to any applicable policy permitting or denying their action. [2]
* In lieu of policy, the research committee failed to discuss or explain their actions on-wiki within a reasonable time-frame, nor was any subtantial corrective action undertaken within a reasonable time frame. (IAR, WIARM, BRD) [3][4]

Therefore, the decision to terminate the use of the study banners at
this time was correct, and could be (and was) validly carried out by
any meta admin. (In the end 2 different people drew the same
conclusion almost simultaneously)

It seems reasonable to suggest that the research committee take the
time to obtain an (at least slightly) broader consensus before
restarting their study[5]. It is reasonable to believe that the
research committee might be sanctioned individually or collectively,
should they fail to do so. The making of on-wiki statements is highly
recommended, as off-wiki statements do not contribute to consensus.

In future, before a person or committee starts or continues use of a
wiki-resource, it would be wise to:
* Research, interpret, and adhere to any applicable policy/guideline/essay documents.  
* Obtain and/or research and/or interpret relevant consensus, and adhere to it.  
* Plan sufficient time and resources for the correction of (inevitable) unforseen issues.

If this is done in a timely manner, this needn't take a lot of time or
difficulty. Fixing errors and misunderstandings post-hoc is more
costly.

sincerely,
	Kim Bruning


[1] AN* discussions are not merely "suggestions".  Depending on
consensus, statements made on AN*  can be actionable. Compliance might
be mandatory, failure to comply may be sanctionable.

The following consensus discussion is applicable:
http://en.wikipedia.org/wiki/Wikipedia:Administrators%27_noticeboard/Archive222#Researchers_requesting_administrators.E2.80.99_advices_to_launch_a_study

This discussion can be interpreted as denying the request to run this survey in the then proposed form, and discouraging the current form. The discussion also provides some minimal requirements to make the survey acceptable.  These requirements were only partially met.

[2] So far I've only found the following proposed policy:
http://meta.wikimedia.org/wiki/CentralNotice_banner_guidelines . Note
that this recommends that banners be approved by the community
beforehand. This was not done. 

[3] AFAICT, no input at
http://en.wikipedia.org/wiki/Wikipedia:ANI#Harvard.2FScience_Po_Adverts
. It may be open to discussion whether or not the research committee
was properly informed via their page
http://meta.wikimedia.org/wiki/Research_talk:Dynamics_of_Online_Interactions_and_Behavior#Discussions_about_the_banner 

[4] Jerome and I did try some last minute IAR-ish defence of the
project (and we convinced a number of people!). Obviously, our last-minute
arguments were insufficient to balance out the previous and continuing
issues at the time. (worth a try though! :-)

[5] WP:POLL suggests that simply discussing and then adhering to a
common position is potentially sufficient. Running an actual poll
might be counter-productive.