By Tim Davenport /// “Carrite” (Wikipedia username) /// “Randy from Boise” (Wikipediocracy username) with some modest assistance from Yerucham Turing
One of my pet peeves is the way that the WMF bureaucracy conceptualizes Wikipedia participants. They see the world as a potential drone army for them to manipulate into editing Wikipedia through “social networking” devices (such as their failed “Rate This Article” initiative) and artificially-sweetened raspberry-flavored software solutions (Media Viewer, Flow, Visual Editor).
There are billions of people in the world, after all, and golly, they should all be editing “The Encyclopedia Anyone Can Edit” because, ummmm, it’s an encyclopedia anyone can edit, after all. It’s just a matter of the Kumbaya San Francisco Friendly Spacers making it happen with good vibrations and their software programming brilliance! (Hurrah!!! Hurrrah!!! Hurrah!!!)
Of course this assumes that the community of volunteers that actually built the encyclopedia and governing apparatus behind the encyclopedia, are nothing but the Most Highly Perfected editing drones created by their Bay Area masters — who hold all the cards and call all the shots.
This is a matter of fundamental importance.
I was thinking during the Mediaviewer/Superprotection fiasco that it had finally sunk in with the WMF circle jerkers that the Wikipedia Volunteer Community was indeed a real entity, to be dealt with on the basis of partnership. No such luck. Get a load of the following slide from a presentation at the WMF Sept. 25 Mobil Metrics meeting held at Club Headquarters in San Francisco:
Once again we see WMF accepting as axiomatic the dangerous and erroneous model of linear “editor engagement” — casual visitors via Google become regular readers straight to the site; who become casual editors; who become more regular editors; who VOILA!!! become very active editors!
“It’s all a matter of getting more readers, you see, and the rest will take care of itself, thanks to us.”
This is not the way that Wikipedia’s core volunteer cadre is recruited and built and it is especially not the way that Wikipedia’s need for core content writers in specialized subjects needs to be recruited and built. We need specialists and experts, not an influx of random “crowdsourcers” tricked into making edits 1 through 6 with magic software beans…
“Very Active Editors”
|On the other hand, the notion of “Very Active Editors” (100 edits/month) is very useful: unless one uses high levels of automation, 100 edits/month takes a day or two of hard work each month, or sustained small efforts over the course of the whole month. It is also a metric that has been carefully charted for years. That’s not a bad means to estimate the size of the WP volunteer community — certainly the best metric extant. One venerable Wikipedia-watcher observed:
There is some limited academic research on this topic, which was mentioned in a recent edition of Signpost. The summary lays on the jargon heavily; it is at root an attempt to analyze the question of editor turnover at WP by dividing WP participants into discrete categories and analyzing the process of change in participation of each category.
The paper assumes that WP editors have identifiable “life cycles” and indicates that one possible benefit of understanding the likelihood of continued participation would be the enabling of “task routing software” to assign specific jobs to volunteers most likely to complete them.
Reference is made to a 2011 study by Weser et al. which identified four primary groups of Wikipedia participants: “substantive experts, technical editors, vandal fighters, and social networkers.” Another 2009 study by Panciera et al. found that long term volunteers tended to “start intensely, tail off a little, then maintain a relatively high level of activity over the course of their career.” (pg. 2)
The study examined the activity of Wikipedians from the start of the project until August 2013, divided by three-month “quarters” of the year, tallying edits in each Wikipedia namespace (mainspace, mainspace talk, project, project talk, user, user talk, etc.). The study found that an “overwhelming number” of WP editors (4,468,352 out of 5,749,590 — 77.7%) were active only for a single quarter.”
One obvious trend is that the number of users who stayed active for longer time periods is becoming smaller and smaller, indicating that Wikipedia experiences high levels of member withdrawal,” the authors of the study noted. (pg. 5) Indeed, about 60% of registered users stayed only one day. (pg. 6)
The paper reveals that social networking elements have been consciously introduced by WMF in an attempt to “attract and retain user participation,” which it assumes is linked to editing activity. (pg. 6)
But this is all in one ear and out the other with WMF, if they bother to read it at all. WMF hasn’t made any effort whatsoever to understand who these people are and how they are subdivided. There is much more they could and should be doing in terms of building databases of Very Active Editors and surveying their needs.