Monthly edits on Wikimedia wikis still on the rise

This chart shows that the overall volume of manual edits by registered users on all Wikimedia wikis combined is still increasing, slowly but steadily. When we substract rookie edits (arbitrarily set to first 100) the rise of highly effective edits (red line) rises even a bit sharper.

EditsOnAllProjectsTillDec2012

This entry was posted in Nice Charts, Wikimedia Edit(or)s. Bookmark the permalink.

9 Responses to Monthly edits on Wikimedia wikis still on the rise

  1. Nemo says:

    Nice graph, however the difference may just be that registered users are reverting more vandalism; if only we could get updates for http://stats.wikimedia.org/EN/PlotsPngEditHistoryTop.htm and related graphs…
    The decrease of what you called “rookie edits” proves the constant decline of new editor recruitment since January 2007, doesn’t it? It would be extremely nice to have that orange line disaggregated by project (log scale).

  2. Erik says:

    @Nemo, What is new with this chart is that it uses merged editor data for all wikis. Before we had those we couldn’t know whether some user was new to our projects or only new to that particular wiki. Yes you can interpret the chart in different ways. As you say the bad news is we have less rookie edits, mostly indeed because we have less new editors. The good news is our experienced editors are still going for it, and seem not less motivated, or more distracted by (other) social sites. I agree, up to date revert stats would help to assess how much time veterans can spend to enrich our content (to do: make scripts use relatively recent checksums in stub dumps for revert detection).
    A big unknown I would say is how much more productive the average edit has become because of changes in UI, better templates, and what have you. As for your last comment I hope to add a chart for this to the summaries page for each wiki (time permitting).

  3. Nemo says:

    You’re right, I did notice that the smart idea of counting edits allows us to forget duplicates, which is the big problem of the active editors counts: as aggregated global metric, this one is more reliable. I just gave it for granted in that “nice graph”, sorry for my usual negativity. 😉

  4. Erik says:

    @Nemo, not problem at all. My initial comments were a bit hasty. Instead I should thank you for being one of the most engaged followers of Wikistats with always interesting comments. I highly appreciate it. So “Thank you!” 🙂

  5. Pingback: Wikimedia Research Newsletter, March 2013 — Wikimedia blog

  6. WereSpielChequers says:

    Hi Erik, nice graph, but I wonder what effect the edit filters are having on this? On the English Wikipedia and I suspect the other major ones the edit filters are getting steadily more sophisticated and preventing ever more vandalism from actually taking place, and if it doesn’t take place then we “lose” not only a lot of vandalism edits but also large numbers of reversions and warnings.

    I used to subscribe to the theory that the volume of vandalism, like spam, was related to our readership, hence a stable amount of editing and a rapidly increasing readership was a recipe for trouble. However I’m now moving to the idea that Smart phone users are much more passive – we are in a sense broadcasting Wikipedia to today’s smartphone generation. Some devices are much more suitable for editing on than others, and the rise of small screen devices could be the main break between our readership volume and our editorship volume. If it is true that teenagers have largely moved from PCs to smartphones then that might help account for the greying of the pedia and the relatively small number of vandalisms that I now seem to find on my watchlist.

    For similar reasons you might see a big change to your graph as a result of Wikidata and the move to a hub and spoke system of intrawiki links. On the old system an otherwise near moribund language version could be guaranteed a steady stream of bot edits adding extra intrawiki links to articles on Barack Obama et al as other wikipedias created such articles and that has now stopped. Of course if you are screening out Bot edits you will lose much of this, but not all of it.

    PS I’m trying to summarise the various theories for the community going off the boil at http://en.wikipedia.org/wiki/User:WereSpielChequers/Going_off_the_boil – if you have any time to run an eye over it I would greatly value your views and knowledge.

  7. Erik says:

    @WereSpielChequers your point of less vandalism that gets through nowadays makes me even more optimistic. You have a lot of interesting observations and hypotheses in your essay.

    Your point of the greying of the pedia is also reflected in my line for inexperienced edits. Around 2007 more edits were edits by inexperienced editors and probably many were fixed by that same editor who did not know how to use the preview button.

    Your point of scaring people away that we want to attract (gender gap) has a pendant where we should be careful not to demotivate existing editors. If we tell, even unprompted, at every interview that Wikipedia’s biggest problem is loss of editors, then old hands who are still around may start ask themselves “Why am I still around?” I believe in openness and bad news needs to go out. But alarmism is in my view totally unwarranted. If feeds sensation seeking press. Your term ‘going of the boil’ is exactly how I would phrase it. Thank you. I hope it sticks.

    Is your essay finished? I’ll tweet it then.

  8. WereSpielChequers says:

    Hi Erik,

    Thanks for that, yes having it tweated about by you would be an honour, but it probably won’t be finished till sometime next week. I will update you then.

  9. WereSpielChequers says:

    Dear Erik, I’ve slightly expanded http://en.wikipedia.org/wiki/User:WereSpielChequers/Going_off_the_boil%3F and added a question mark to rename it “Going off the boil?”, I’d forgotten about the shift to Vector replacing a skin that was “optimized for editor usage” with one that was designed to be more friendly for readers.

Leave a Reply

Your email address will not be published. Required fields are marked *