Wikimania editor retention meetup notes

Notes from the lunch meetup about editor retention that happened at Wikimania 2019.

2 Likes
  • main reasons for losing experienced editors: harsh debates, changing life situation / priorities
  • Wikia editors are younger than Wikimedia (readers average 18, editors are not much older)
    • realtime communication is very powerful for engagement (many communities use Discord which is popular with the demographic but using proprietary SaaS software is a risk)
  • friwiki has a talkpage-based pseudo chat system; it takes lot of effort to get statistics out of it but otherwise works well
  • gamification would help (esp. for retaining experienced editors with reduced free time)
  • you need a critical mass in your community
  • depending on the size of the community, you might need different types (experience levels) of editors most
  • organizer retention is also a problem
  • microcontributions for engaging new editors (citations, suggested edits)
    • are they useful for growing people into content editors though?
    • starting an article and asking an expert to improve it could be a kind of microcontribution
    • dewiki: potential microcontributions (e.g. typos) get fixed too quickly by experienced editors
  • Icelandic Wikipedia deleted too much on notatibility grounds; users wrote encyclopedia articles in Wikibooks instead, some of those have since been moved back to Wikipedia
  • project office: collect all tasks/tasks suggestions on one page
    • would be useful for Growth task lists
  • if you want to teach editing skills, you need “minicontributions” (larger than micro), e.g. fix article summary
  • idea: wikishootme for nearby red linked sites
  • what do people do at editathons? (there was a fair bit of discussion on this but I repurposed my laptop as a clipboard at this point so I have no notes)
3 Likes

Has this been supported by any evidence other than anecdotal one?

For example, 10 or even 15 years ago the debates were way more harsh, but the editor decline started more recently. It must have to do smth with motivation (like that most easy article titles are there etc)

These are notes from a discussion over lunch where people involved with some form of editor retention work (mentors, event organizers etc) shared their experiences. Collecting anecdotal evidence was pretty much the goal.

(I’d be happy to start a discussion about quantitative evidence in a separate topic, if you are interested.)

I am in principle interested, but it would be good to have more people interested, not just me.

To write http://mediawiki.org/wiki/Talk_pages_consultation_2019/Discussion_tools_in_the_past I talked to several editors who were active in 2004 (15 years ago). They left me with the clear impression that the debates 15 years ago were much smaller and not as harsh (mostly enwiki and dewiki). There was more interest in getting stuff done, and having the occasional bad idea wasn’t a social disaster. As long as your idea didn’t directly conflict with someone else’s idea, you could usually proceed.

It might be helpful to remember just how long ago this was. “Old” ideas, like the Five Pillars, didn’t exist. MediaWiki itself was new. The idea of having separate namespaces for articles and discussions was about a year old. The Wikimedia Foundation was new and had zero staff. RFA meant that you sent an e-mail message with a reason why you wanted admin rights, and if nobody objected, you passed.

According to the stats at https://en.wikipedia.org/wiki/Template:Wikipedia_editor_graph_(100_per_month), 10 years ago (2009) was an era of significant decline for active editors at the English Wikipedia; the numbers there parallel the overall movement’s numbers (although a few individual communities had different results). That decline might have had as much to do with a reduced need for simple edits (you don’t need to revert a lot of vandals when ClueBot is doing it faster than you can) as with social changes, but 10 years ago, we were in a well-publicized decline.

By contrast, in the last couple of years, the number of highly active editors seems to have stabilized overall, and is going up in some places. While I am continually surprised to discover that people I knew from back in the day are now blocked or inactive, and I’ve had the difficult experience of writing obituaries for two editors I admired, I don’t think we can claim that there is an editor decline underway right now. If the discussions were actually more harsh 10 years ago, then maybe we’re improving, and seeing the results in the end of the decline.

2 Likes

This really varies wiki by wiki. For example, enwiki is mostly static, dewiki is declining, jawiki is growing. In any case, improving editor retention and increasing growth are high-impact goals even on wikis where there already is some growth. There’s also the issue of churn (if your wiki community is growing, but every new user comes at the cost of chewing up and spitting out 99 others, what happens when you run out of new users? getting ex-newbies with bad experiences to come back will probably be harder than helping new users to stay).

We are really getting off topic though :)

1 Like

@Tgr_WMF I made similar notes, so I won’t copy them here.

Two additional notes:

  • friwiki is frwiki (French Wikipedia)
  • We discussed, that microcontributions (fixing a lot of similar problems) in the form of games is better for experienced editors, but newcomers cannot really learn from them (and arguable if they enjoyed). For them varied small contributions are better for learning how Wikipedia works (technical side and editorial rules and guidelines).
1 Like

@Ymblanter you are not alone. :wink:
We had 14 participants on the first meetup on Wikimania, and 20 people indicated their interest until now.

I will invite them to this platform soon*.

(*: I am waiting for the Wikimedia login feature, which is almost here…)

3 Likes