I think this is an interesting idea, Pine.  Thanks for bringing it up.  I imagine that you could learn a lot from a few long-term, highly active Japanese Wikipedians by talking to them about how we have <problem/solution/process/policy> in English Wiki (and/or other wikis that are discussed often in the research lit) and learning where they see differences in their jawiki.

Resource-wise, this would be difficult to get on the plates of the WMF Research team.  Personally, I spend a lot more time operating as infrastructure for other people's research and tool development (see ORES[1]) than doing my own research projects.  I think that's generally true of people on our team.  However, I'd love to *help* someone else look into this.  I think I could do a pretty good job of connecting the English Wikipedia-focused peer-review lit to casual discussions of Wikipedia stuff.  I could also probably help pull in some other academics to help flesh out references to the academic lit.  Pine, maybe you could kick off the project by posting a set of questions about Japanese Wikipedia on a wiki page and trying to see if you can find a Japanese Wikipedian willing to give their perspective.  Here are some of my own questions:

* Do you do anything like subject-focused WikiProjects?  

* Why, do you suspect, that your new article survival rate is so high?[1]

* How do you help newcomers figure out how things work (rules, processes, jargon, etc.)?

* How do you find and address the most persistent cases of vandalism and other bad-faith activities?

* Do you have essays like we do in English Wikipedia?  If so, which are your personal favorites and why?

* What gadgets and 3rd party tools do you use?  Which tools seem to be popular?

1. https://meta.wikimedia.org/wiki/Research:Wikipedia_article_creation

-Aaron

On Sun, Dec 18, 2016 at 2:05 AM, Pine W <wiki.pine@gmail.com> wrote:

Hi Yusuke,

It's nice to hear from you. I remember you from our time on the IEG Committee together.

ENWP has its share of bots and sockpuppets too, but I still think Japanese Wikipedia has some good statistics.

Let's consider the 100+ edits/month users. According to https://stats.wikimedia.org/EN/Sitemap.htm, Japanese Wikipedia has 372 by last count, which is the 6th highest of all language Wikipedias. Now let's look at these ratios:

English Wikipedia: 3340 editors with 100+ edits/month, divided by 1500M prim+sec speakers, gives us a ratio of about 2.227 highly active editors per million prim+sec speakers.

Japanese Wikipedia: 372 editors with 100+ edits/month, divided by 132M prim+sec speakers, gives us a ratio of about  2.818 highly active editors per million prim+sec speakers.

So even by that measure, Japanese Wikipedia is doing relatively well.

Perhaps there are Japanese universities that WMF could contact about the possibility of doing this kind of research, and WMF could provide funding if necessary. (Here in the US, I believe that some similar work is funded by the National Science Foundation; perhaps there is a similar source of funding for research in Japan that could also be asked for funding for the project). I'm not sure who at WMF would need to support this kind of work for it to go forward, but I think that Dario is at least open to the idea. Perhaps he'd be willing to have a conversation with you about how to set up this kind of research project; I for one would be interested in in the findings.

Pine

On Sat, Dec 17, 2016 at 9:31 PM, Yusuke Matsubara <whym@whym.org> wrote:

Hi Pine,

> I'd be
> interested in learning more about what makes their community's edition
> of Wikipedia so successful in terms of a high proportion of Japanese
> speakers contributing to the site, that could be applied to other
> language editions.

Are you thinking of qualitative research or quantitative research?
Regarding the former, the question reminded me of a research project
on the history of the Russian Wikipedia and other wikis (lead by
Maryana Pinchuk in 2011) which might serve as a reference point.
[1][2]

>From the project's description:

> Across the globe, there are currently over 270 autonomous
> language-based Wikipedia projects, plus many sister projects such as
> Wikiversity, Wikibooks, and Commons. Why are the communities behind
> some of these projects still experiencing rapid growth while others
> are leveling off or gradually declining? How have different
> communities of volunteers overcome cultural, social, and technological
> obstacles to create the most up-to-date online reference materials in
> the world? What lessons can communities learn from each other in order
> to make every project more healthy?

I would be interested in how much resources (if any) the WMF provided
to the research project, and how much effort from the communities was
needed. An experienced Japanese Wikipedian and me floated the idea of
something similar for Japanese Wikipedia one or two years ago - at
least regarding the community resources, we were not sure how we can
secure the time and energy required for such a study, especially from
the few volunteers qualified to do so and who tend to be already busy
for other Wikimedia-related volunteer work.

Also, regarding "a high proportion of Japanese speakers contributing
to the site", this might not be as great as it might look like. The
number comes from the number of accounts with 5+ edits in a month.
Sockpuppets and throwaway accounts could easily skew this number (and
sadly they are incentivized to do so because of a 5-edit requirement
in the voting rules). If you look at the proportion of the 100+ edit
users, Japanese Wikipedia is not as high.

[1] https://meta.wikimedia.org/wiki/Research:WikiHistories_fellowship
[2] https://meta.wikimedia.org/wiki/RuWiki_History_(Doronina_and_Pinchuk)/English

-Yusuke

On Fri, Dec 16, 2016 at 5:22 AM, Pine W <wiki.pine@gmail.com> wrote:
> The topic of audiences was discussed at today's WMF Metrics and
> Activities meeting.
>
> Looking at https://stats.wikimedia.org/EN/Sitemap.htm, and sorting by
> editors (5+ per million speakers), there are some language communities
> that appear to have high participation rates on their language's
> edition of Wikipedia, but I hear very little from them in meta
> discussions. Japanese Wikipedia comes to mind in particular, with its
> large number of primary + secondary language speakers. I'd be
> interested in learning more about what makes their community's edition
> of Wikipedia so successful in terms of a high proportion of Japanese
> speakers contributing to the site, that could be applied to other
> language editions.
>
> Could WMF direct more resources to studying the successes on Japanese
> Wikipedia, and how information about those successes could be applied
> to other language editions of Wikipedia?
>
> Pine
>

> _______________________________________________
> Wiki-research-l mailing list
> Wiki-research-l@lists.wikimedia.org
> https://lists.wikimedia.org/mailman/listinfo/wiki-research-l

_______________________________________________
Wiki-research-l mailing list
Wiki-research-l@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l

_______________________________________________
Wiki-research-l mailing list
Wiki-research-l@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l

_______________________________________________
Wiki-research-l mailing list
Wiki-research-l@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/wiki-research-l