[Wikimedia-l] Resolution: Media about living people

Sat Dec 14 16:54:22 UTC 2013

On Sat, Dec 14, 2013 at 6:08 PM, Craig Franklin
<cfranklin at halonetwork.net> wrote:
> Hi Jane,
>
> I am concerned about the issue surrounding the comment "the real BLP
>> problems happen when heavyweight (in edit count terms) Wikipedia users
>> swing their weight around"
>>
>
> I think the problem is that if you ask ten different people about the
> reason why we have BLP problems, you'll get ten different answers.  All ten
> would probably have some truth in them, but any one in isolation would be
> inadequate.

The list of problems becomes even longer for images.

The 2009 resolution on biographies of living people was about
identifiable people, given they were the subject of a biography.  This
new 'media about living people' resolution doesn't make any such
distinction for media, which I guess will result in lots of confusion
about whether the scope includes images of unidentifiable people.  It
should, but ...

This resolution appears to be asking for verifiability regarding
images of living people.  We are going to need some clarity around
what the board considers to be verifiability (how do we prove the
photo was taken at a public event and it is real? etc), and whether
that includes unidentifiable people.

"Ensuring that all projects in all languages that describe or show
living people have policies in place calling for special attention to
the principles of neutrality and verifiability in those articles;.."

On English Wikipedia we have some guidance regarding photos of living
people, but I can't find anything relating to verifiability or
neutrality.

https://en.wikipedia.org/wiki/Wikipedia:No_original_research#Original_images

https://en.wikipedia.org/wiki/Wikipedia:Biographies_of_living_persons#Images

Wikimedia Commons has a policy which rejects 'neutrality', and it
doesnt have a verifiability policy.

https://commons.wikimedia.org/wiki/Commons:Project_scope/Neutral_point_of_view
https://commons.wikimedia.org/wiki/Commons:Photographs_of_identifiable_people

Is there a discussion happening on Commons somewhere about the
implications of this resolution?

> My own point of view is that our policies and procedures are actually
> pretty good on paper, but they're just very unevenly and inconsistently
> applied in the real world.  The "Tier 1" biographies, such as those of
> Messrs Obama, Cameron, and Abbott are pretty safe from BLP hijinx, but
> there is a massive underbelly of poorly defended BLPs on minor celebrities,
> local politicians, and the like, which are not watched consistently and
> where hagiography or defamation can take root.  This is why, while things
> like the BoT's declaration are not unwelcome, I feel that they don't have
> any practical effect in fixing the problem.  All it takes is for one
> negatively written bio to slip through the net to do real harm to someone
> in the real world.

I agree with you Craig up to here ..

> My preferred way of dealing with this on en.wp would be to massively
> tighten the notability criteria where they related to biographies of living
> or possibly living people, but this would no doubt be met with cries of
> "deletionism!".

And agree your preferred approach could help.  On English Wikipedia, I
think we have an article/editor ratio problem, which is only getting
worse as articles increase and editors leave, and is meaning
watchlists are less useful to scan for problematic edits.

The test for this is what is the average length of time between an
edit of an old page (e.g. created in 2005) to the point in time that
the edit a) appears on a watchlist, or b) is viewed as a diff, or c)
is loaded as a page view, or d) leads to another edit.  Then compare
those averages with the averages from a year before, to determine
whether edits are slipping past watchlists and recentchanges. I'm
guessing that the length of time from edit to (a) or (b) is
increasing, while (c) may be decreasing as Wikipedia readership
increases.

A smaller Wikipedia scope means there are less articles, with more
editors watching and editing the pages the BLP problems appear on.

I think it is necessary to add here that FlaggedRevs (Pending Changes)
also helps, as any BLP problems are held in a queue.  The 'volume of
edits' can be a problem with FlaggedRevs in practise, but a) the
'size.

> Indeed, I don't think it's possible to adequately address
> the issue on large projects like en.wp or commons without a massive
> cultural shift and sweeping changes to policy that would cause immense
> disruption in the community; something the BoT is understandably reluctant
> to do.

Another way the board can get serious about this problem is to mandate
that each project write a BLP management strategy that needs to be
approved by the WMF board, which would involve some type of periodic
review of all content.  The strategy would differ for each project
based on their policies, scope and the size of the project.  e.g.
Wikisources would need to review only unpublished sources added each
year; Wikipedias using FlaggedRevs could do spot checks annually;
Wikipedias which have chosen to not use FlaggedRevs would be required
to come up with feasible alternative solutions to verify the existing
BLPs are clean of significant BLP problems.  Projects which failed to
complete their periodic reviews of the content would be put into
maintenance mode(s) until they have completed the review.  e.g. The
devs might be asked to disable 'creation of new pages in mainspace' on
the wiki as a first step measure to focus the community on the task.

More generally, we should have tiers in the notability system, by
which we agree that not everyone is as notable as Barack Obama, and
therefore their 'living' bio should not contain every detail that is
ever published.  The lowest tier is bios about people with
questionable notability or low notability and avoid publicity, such as
(most) referees, sports people who only played a few matches, most
academics, which should only include facts that are relevant to their
notability and their brief appearances into 'public life'.  On English
Wikipedia, those articles should all be put under FlaggedRevs, and
edits that increase the scope of the biography are
rejected/held/not-approved until there is consensus on the talk page
that the subject is notable enough that other aspects of their life
are of general interest to understanding their achievements or actions
which have become notable.

Perhaps not just yet, but Wikidata should bring new solutions to this
problem.  We may have more consensus to remove classes of living
people biographies from Wikipedia as the basic details of their life
can be placed into Wikidata.

For example, only a few of these referees deserve a proper 'biography'
- for the others, their bio exists on Wikipedia only because it is
useful to have a unique identifier for the person, and we like to
record a list of a person's public appearances.

https://en.wikipedia.org/wiki/Category:Australian_soccer_referees

In a few of those articles, there are unsourced claims that the
referee made a significant mistake.  Besides official honours awarded,
there is not similar commentary describing all of the times that
sports commentators spoke highly of the referees decisions.  i.e.
these articles are either BLP problems now, or will be in the future.
A referees decisions are usually only relevant within the context of a
match, and don't belong on their bio.

In almost every case, the details in those articles can be moved to
claims in Wikidata once a few Wikidata properties are created, and a
non-editable page could be automatically generated on Wikipedias to
describe the subject and list the events the person appeared in.

-- 
John Vandenberg