Inside Architecture

Notes on Enterprise Architecture, Business Alignment, Interesting Trends, and anything else that interests me this week...

Does anonymity promote ill-informed consensus?

Does anonymity promote ill-informed consensus?

  • Comments 4

There are a spate of new Social Media apps that have emerged lately, all of which allow people to post comments and ideas anonymously.  They are being quickly adopted, especially among the very important 13-18 year old “adolescent market.”  They are also being quickly banned for promoting cyberstalking, cyberbullying, and otherwise cruel behavior.  Does anonymity protect cruelty?  And what does that say about more established Anonymous sites, like Wikipedia?

Normally I don’t comment on Social Media.  My regular readers know that I tend to focus primarily on enterprise architectural concerns like business model viability and strategic alignment.  But there is an interesting cross-over between Enterprise Architecture and Social Media, especially anonymous social media: the creation of community consensus.

The state of anonymity

For those not keeping up, there is a spate of new social media apps that have emerged lately, from Whisper to Secret to Yik Yak, that allow smartphone users to sign up and then post messages unfiltered and anonymously.  When in anonymous mode, users tend to say things that they feel uncomfortable saying on Twitter or Facebook (where their friends, family, and coworkers may discover a side of them that they may not agree with). 

YikYak especially is troubling because it uses a geolocation filter… you can see things posted by people within a certain distance of you.  Sounds innocent, right?  After all, young adults filtering through Bourbon Street festivities in New Orleans could share that a particular bar was playing really good jazz, or that drinks are strong and cheap across the street.  But you may quickly see the problem when I use two words: middle school.  Already, some High Schools and Middle Schools have had to ban the app because it became a platform for bullying and cruel comments.

The effects of anonymity

But what does it mean to be anonymous?  What are these comments that the guy next to you would like to send to “the world” without anyone knowing it was him?

You can look for yourself at Whisper.sh.  I spent a few minutes browsing through some of today’s messages.  Most were simple secrets… many were sexual or related to dating.  Some were work related.  Most had responses from equally anonymous people, and most were fairly benign.  Of course, there could be some judicious editing going on for the sake of casual surfers like me that own a Windows phone (and therefore can’t use the app).  Secret and Yik Yak don’t even make an effort to show any of their messages on their website.  It’s all in the app (once again, only for IPhone or, in the case of YikYak, android).

Of these, I think Yik Yak is the most interesting from a consensus point of view, because it is the only one that attempts to filter according to a community.  GeoLocation, especially when it comes to universities or even small towns, is sure to limit the reach of a message to people who share something in common with you.  That sense of “sharing something in common” is really what defines a community, and consensus only really matters in a community.

Anonymity and consensus

Does anonymity work to create consensus?  Sure.  Think of standing in a large crowd.  If one person yells something, you don’t normally turn to them and identify the source before considering, and possibly agreeing with, the content.  This is the very essence of a political rally or a protest march.  Taking in unfiltered ideas and deciding on them, on the spot, is part of how consensus is built.  Of course, there is no good way to take in ONLY good ideas when you are in a crowd.  We count on the crowd to do that for us.  If someone in a political rally yells “Death to the other guys!” we would expect the folks standing next to them to react, possibly causing the rabble-rouser to back down.  (Unless your protest march is in Karachi or Tehran or Cairo… but that’s another post).

In that sense, standing in a crowd is only “partially” anonymous.  There are still people who can see you, and if you do something really outrageous, there are people who could react by hitting you.  This is why you won’t find many people who will go to a crowded Yom Kippur (Jewish) service and stand up in the middle of the crowd and yell “Hitler was right!”  Pandemonium. 

But consensus and anonymity online is very different than standing in a crowd, and I think we need to be aware of the differences. 

The perils of anonymity online

Online, you can make claims that are difficult for another person to dispel, without consequence at all.  There is no one next to you ready to elbow you when you use name calling, or circulate unfounded rumors, or simply make things up!  Even when we use our actual names, we may participate in a discussion where we are not in the same room, or even the same continent, as our peers, and this can cause problems.

I cannot count the number of times I’ve witnessed this on LinkedIn.  A person will ask a question about frameworks, and I may point them to PEAF (a framework created by Kevin Smith).  No problem.  But if Kevin himself gets on the thread and mentions PEAF, his messages are blocked and he may even be kicked out of the discussion.  Why?  Because someone somewhere made a spurious charge (that he makes money when you use PEAF, which is not true).  Since the administrators of most LinkedIn Groups are anonymous, they can make bad decisions without consequence.  There is no good way for Kevin to clear his name of these charges because he does not know who the administrators are, and they appear unwilling to consider the possibility that he is not, in fact, using the platform to promote his own self interests.  Rumor rules the roost.  Not good.

I believe that the same thing applies to Wikipedia. 

Wikipedia, with its millions of articles, has emerged as one of the chief sources of encyclopedic content on the Internet.  It is widely respected, and most search engines make a point of returning Wikipedia entries near the top of their search results.  However, the administrators on Wikipedia are mostly anonymous.  (They use pseudonyms to do their editing work). 

This causes the same problems to occur in Wikipedia that occur in any other setting where people can be anonymous… mostly benign behavior with occasional outburst of bad behavior (with nearly no consequence). 

There is an essay (not a policy) on Wikipedia that says “Only Martians Should Edit.”  This policy says that some topics are so controversial that anyone associated with the actual content would be too biased to edit the content in a neutral manner.  Therefore, topics dealing with such things as State or Provincial politics, or national boundary disputes, or whether specific historic events should be counted as a genocide.  These things trigger strong emotions, so having people edit the articles as though they are “from Mars” can be a good policy.

On the other hand, for some topics that are very narrow, it is not possible to edit the article without knowledge of the subject.  If you are not an expert in African pop music, you may not do a good job discussing Azonto music and dance from Ghana.  In this case, an editor with no grounding in the subject is likely to make mistakes. 

The problem is that Wikipedia is based on consensus, and you may find yourself editing a page on Wikipedia where you have to build consensus among anonymous people, people that may or may not have ANY understanding of the subject matter.  And those people can be nice, or cruel, with no consequence.  There is no one in the crowd next to them ready to elbow them for making an outrageous statement… because the other editors don’t know if the statement is outrageous!  You can build credibility on how well you enforce the rules, and then use that credibility to attack someone, and no one else can tell the difference.

Anonymity: Handle With Care

I’m of the opinion that anonymity on the Internet has to be handled with care.  There are times when it is necessary, especially when attempting to avoid governmental or organized oppression to free speech.  On the other hand, there are times when it is a license for ill-informed people to promote nonsense as a consensus.  After all, one third of Louisiana Republicans have been misled into thinking that Obama is to blame for the poor response to Hurricane Katrina.  I can think of a other examples of an ill-fated consensus among the ill-informed, but rarely one so laughable.

I believe that Sites and Apps should not leverage anonymity as a feature.  I make exceptions for Tahrir Square and Occupy Wall Street, etc, where rumor may be the only information you can trust, but that is not what these apps do. For normal social interactions, anonymity is actually a problem.  On Wikipedia, I believe that anonymity has outlived its usefulness. 

  • There's a careful distinction that you're failing to make between anonymity (no identity revealed) and pseudonymity (a false or proxy, but consistent, identity revealed). Pseudonymity offers a middle ground between the anarchy of anonymity and the chilled environment of "real names". Wikipedia administrators, for example, are not anonymous because there's a consistent identity necessary to earn the trust to be elected to the position.

    People have a strange idea that enforcing "real names" will somehow improve behaviour online. While it's somewhat true in that it'll *discourage* the worst behaviors, it fails in two key respects: it is either not enforceable or not scaleable, and some of the worst actors will be undeterred by use of real names. If enforcing real names is strict and strongly confirms identity, it'll be next to impossible to scale to the millions on larger sites. If it's not strict, people may simply stop being "PseudonymousUser423" and start being "John Q. Smith", which is a situation that's worse than anonymity because it promotes a false sense of security.

    Identity policy is not a silver bullet or any substitute for real moderation. Bullies have been bullies for hundreds of years before the Internet under "real names": why would they change now?

  • Fair criticism, Nihiltres (which I assume is a pseudonym)

    It appears you are concerned primarily about my criticism of Wikipedia.  Realize that my article was not centered on Wikipedia.  It was centered on social media like Secret and Whisper.

    As for the relatively minor concern of Wikipedia, I explicitly covered the concern I have: that uninformed people can create a consensus without any reference to knowledge or understanding.  The "rules" of using reliable third-party sources in Wikipedia entries helps to relieve the need for the editors themselves to actually be aware of a subject area.  However, editors create words, and words frame ideas.  While reliable sources are required, there is no doubt that the translation of words that goes on between reading a source, and citing it, creates opportunities to either focus on a narrow aspect of the source material, or to ignore the original point altogether and draw unfounded conclusions.

    We have seen this a thousand times with the most heavily cited book: the Bible.  How many people with little or no liturgical understanding have used the Bible to justify things that the subjects in the bible would clearly NOT condone?  Let's be clear here: people who take from a source and build an article are people who decide what is said, and what is heard.

    So these people should be vetted.  We need to check the sources of our editing, not just our content.  We should listen to all, but we should listen to knowledgeable people MORE than non-knowledgeable people.  This is the model of the Royal Society and it works.  The opinion of an editor on Wikipedia, even one that has made 10,000 edits, is not more important or more relevant than the opinion of an editor that has made 10 edits but is more knowledgeable about the subject.

    Yet, that is the result of pseudonym identity.  A person can hide from their actual position in the world.  They can pretend that their contribution stands apart from the character.  

    It doesn't.

  • I find the concept of schools banning individual application scary.  There is already a power imbalance in a school, and banning these apps simply shuts down a manner in which an individual could expose an abuse of power safely.

    On Wikipedia, my view is that  rather than a simple consensus, a particular group-think had emerged like it does in many open source projects leading to the establishment of an intellectual hegemony that governs the site.  My impression of the people I have met in person that are highly active on Wikipedia is that they are mostly likely those people who display the Dunning-Kruger effect.  Acceptance on Wikipedia equates to corroboration of their "high" intellect.

    When it comes to EA fora, you don't have to accept the status quo, you could start a forum yourself and publish the identity of all the moderators as a matter of course.

  • Hi Robert,

    I have to admit that I had to look up the Dunning-Kruger effect.  Fascinating.  Of course, I looked it up on Wikipedia. :-)

    Methinks my fellow Wikipedia editors would be reject the notion that they are so incompetent that they are unaware of their own incompetence.  Well... I'm sure that the incompetent ones would reject the idea.  The competent ones would probably ask follow-up questions.  (grin)

    As for school, there is a "boundary license" that we need to be aware of.  Students can be somewhat restricted in their speech within the boundary, but outside the boundary they have license to speak freely.  

    Therefore, I have no problem with limiting their speech (especially when we are talking young adolescent age range of 11-14 years of age) because you get a combination of ill-informed consensus and a mob-effect.  The result can range from disrupted learning (at the minor end) to all-out destruction of property and danger to staff and students (at the major end).  

    Abuse of power on Monday will still be an abuse on Tuesday, after the students have gone home, discussed the situation with parents, and shared their insight on non-anonymous social media channels.  Waiting a day to protest is probably a better model for adolescent response to the mistakes of teachers and the principal.

    Lastly, I'll finish with a quote from an essay that is currently posted on the front page of the site Wikipedocracy (a site that criticizes Wikipedia) --

    "The basic activity of Wikipedia is flawed. Wikipedia uses people with scant subject knowledge accumulating sentences from other sources produced in the traditional way, using careful research and review. Unless the source text is in the public domain, each sentence is manipulated to avoid close paraphrasing, for fear of copyright lawsuits. Yet even this rule is sometimes ignored or misapplied. Because the Wikipedia editors are mostly unskilled, the sense of the original is often lost or, worse, distorted. When the errors are picked up by experts, they come under fire for being “critical” or “not assuming good faith” or just not being nice."

    The essay's author (who goes by the pseudonym 'anteater') makes a case for "traditional methods" which includes the notion of having experts write articles while editors and other experts review them.  He talks about the idea that Wikipedia is a challenge to that model, but comes to the following conclusion:

    "Far from undermining the idea of traditional methods of knowledge production, the example of Wikipedia supports it."

Page 1 of 1 (4 items)