<CPOV> Wikipedia as an Alternative United Nations-Like Forum

[andrew.famiglietti at lcc.gatech.edu]

----- "Jon Awbrey" <jawbrey at att.net> wrote:

> Andy,
> 
> The situation is not quite as you say, since the use of multiple
> accounts
> by a single person is not strictly verboten as some people seem to
> think.
> This is probably one of the most frequent topics on The Wikipedia
> Review --
> incidentally, WR staffer Herschel Krustofsky has just written an
> excellent
> white paper on the State of the Sockpuppet Issue, as the natives know
> it:
> 
> http://wikipediareview.com/blog/20100614/the-duck-test/
>

This is indeed an interesting piece! I have encountered the sort of, ah, problematic, use of behavior profiling as sock-puppet indicator described here in my investigation. Of course, I'm a little concerned that the author of this Wikipedia Review piece seems to take what looks to me to be a satirical critique of the "duck test" as a defense of it. The important questions this raises for me are "how prevalent is this sort of activity on Wikipedia?" and "what are its consequences?" The author of the Wikipedia Review essay seems convinced that the behavior is widespread, and serves to disenfranchise some positions at the expense of others. I'd be curious to know which positions Wikipedia Review folks feel are systematically discriminated against within Wikipedia, since that might help us design a study to look at the editing activity on related articles in a systematic way.

I don't, however, see any evidence here that contradicts my sense that Wikipedia editors are at least outwardly hostile to the practice of sockpuppetry. 
 
> It is an empirical fact that the mapping of persons (or corporate
> entities)
> to user accounts is both one-to-many and many-to-one (the latter being
> known
> as "public" or "shared" accounts), and Wikipedists have no way of
> controlling
> either with any degree of effectiveness, no matter how much blue smoke
> they may
> blow about the CheckUser mystique.
> 

I don't disagree that Sock Puppets and public accounts certainly exist on Wikipedia. Perfectly policing such a practice in such a large, active community is almost certainly impossible. I suspect it is exactly this impossibility that leads to the wide-spread paranoia of Sock puppets that I've seen on-site. However, the presence of sock-puppets does not, in my opinion, mean that they are a substantial influence on editing activity. It certainly does not tell us what the nature of that influence, if it does exist, is.  

> If you follow the discussions at The Wikipedia Review, you will know
> that there
> has been a constant stream of investigations into flagrantly abusive
> and fraudulent
> exploitation of multiple accounts by onetime "trusted" administrators
> and their "pets",
> many of these investigations being conducted in ad hoc, if somewhat
> implausibly denied
> coordination between "Good Samaritan" WP users and the members of The
> Wikipedia Review.

I'll take a look at this, thanks! 
 
> There is a wealth of data and a host of savvy informants there if
> anyone
> is really up to pursuing the truth about Wikipedian culture and
> practices.
> 
> Jon Awbrey

You've given me a lot to think on, Jon. Thanks for this. 

- Andy
 
> andrew.famiglietti at lcc.gatech.edu wrote:
>  >
>  > Jon,
>  >
>  > Interesting. Are you implying that many Wikipedia accounts are not,
> in fact,
>  > representative of actual persons? Within Wikipedia, as I'm sure
> we're all
>  > well aware, the practice of a single individual creating multiple
> accounts
>  > is called "sock puppetry" and is considered a very serious form of
> fraud.
>  > In my research, I've often found it very difficult to establish the
> extent
>  > to which Wikipedia is successful at defending itself from Sock
> Puppets,
>  > since the data on the IP address identities of registered users is
> only
>  > available to those admins with access to the CheckUser tool. 
> Obviously,
>  > this data cannot be freely handed out for privacy reasons.  I've
> often
>  > thought about trying to talk to the Wikimedia Foundation about the
>  > possibility of getting IP data for a proper study of the issue,
> thought
>  > the hassles of IRB clearance, negotiating with them, and my own
> mediocre
>  > skills at quantitative methods have dissuaded me from doing so. 
> However,
>  > my time in the community has left me with the impression that they
> take
>  > the matter very seriously, there may be no better way to piss a
> Wikipedian
>  > off than to call him or her a Sock Puppet!  Anyway, what I'm
> getting at is,
>  > because this is such an interesting issue so clearly in need of
> more research,
>  > I'm very curious to know what data lead you to believe that
> identity fraud is
>  > rampant on Wikipedia.  This is clearly an area where any new
> evidence could
>  > shed a lot of light on a confusing situation.
>  >
>  > - Andy
> 
> -- 
> 
> inquiry list: http://stderr.org/pipermail/inquiry/
> mwb: http://www.mywikibiz.com/Directory:Jon_Awbrey
> knol: http://knol.google.com/k/-/-/3fkwvf69kridz/1
> oeiswiki: http://www.oeis.org/wiki/User:Jon_Awbrey