Last modified: 2014-11-01 16:56:34 UTC
The Italian Wikipedia wants unregistered users to have same editing permissions on it.m.wikipedia.org as they have on it.wikipedia.org. As the feature is not yet considered ready for being default (bug 53069), the Italian Wikipedia has decided to set $wgMFAnonymousEditing = true for a period of two months, in order to identify the main pain points, bugs and feature requests; it.wiki previously pioneered Wikidata and CirrusSearch as well, among others. For simplicity of communication and understanding (both during and after the period in question), the configuration change will take effect on 2014-11-01 00:01 (UTC) and end on 2014-12-31 23:59 (UTC), so that it matches calendar months and doesn't overlap "busier" months like January. The configuration will be conditional to this time period and will deployed in the usual shell requests/ MediaWiki train windows, 28th or 29th. https://wikitech.wikimedia.org/wiki/Deployments#Tuesday.2C.C2.A0October.C2.A028 The decision has been taken with an official "sondaggio", the ultimate decision-making process of the Italian Wikipedia, with a 92 % majority. <https://it.wikipedia.org/w/index.php?title=Wikipedia:Sondaggi/Sperimentazione:_dare_agli_anonimi_gli_stessi_permessi_nel_sito_mobile&oldid=68864246> The "sondaggio" doesn't affect how further permission changes will be decided after this 2 months period. The WMF was aware of this intention since June 2014 and is of course free to conduct additional/parallel activities like scientific experiments to assess unregistered users' editing on MobileFrontend.
Change 168915 had a related patch set uploaded by Nemo bis: Set $wgMFAnonymousEditing = true for Italian Wikipedia in November-December 2014 https://gerrit.wikimedia.org/r/168915
Assigning to Reedy per standard permission changes process. Reedy, please update this report if you can't deploy the change; others, feel free to assign this to yourself if you want to deploy it earlier.
"Highest" applies because there are only 4 days left to do this.
I'm not sure how the itwiki community chose the duration for this test, but just so you're aware, there were 1,558 distinct logged in users who made a mobile edit on itwiki between September and October of this year, and they made 8,213 mobile edits in total. This is just to give you an idea of the scale you should anticipate with this test. You should also be aware that when we've looked at the quality of mobile web edits to date, there's a slightly higher rate of vandalism and test edits. It's not substantially higher, just a few percentage points greater than what you see coming from new users on desktop, but again this is something to keep in mind throughout the test period. I'm curious how you plan to evaluate the success or failure of this experiment. Are you simply interested in seeing what happens, or do you have specific metrics in mind (e.g., a specific volume of edits, quality of edits, or volume-to-quality ratio)? And if you're interested in quality, how do you plan to measure it? I ask because I've done some work on this in the past (for this paper, among others: http://www-users.cs.umn.edu/~halfak/publications/The_Rise_and_Decline/halfaker13rise-preprint.pdf) that might be of interest. Hand-coding edits is a laborious process, so it helps to have a categorization scheme/code-book in advance :)
(In reply to Maryana Pinchuk from comment #4) > I'm not sure how the itwiki community chose the duration for this test, but > just so you're aware, there were 1,558 distinct logged in users who made a > mobile edit on itwiki between September and October of this year, and they > made 8,213 mobile edits in total. This is just to give you an idea of the > scale you should anticipate with this test. You should also be aware that > when we've looked at the quality of mobile web edits to date, there's a > slightly higher rate of vandalism and test edits. It's not substantially > higher, just a few percentage points greater than what you see coming from > new users on desktop, but again this is something to keep in mind throughout > the test period. Yes, this was considered in preparatory discussions. > > I'm curious how you plan to evaluate the success or failure of this > experiment. I'd summarise as "did the wiki explode? yes/no", judging from the discussion, but I might be misreading: this wasn't decided yet. As I said in comment 0, this is not a scientific experiment nor a test to verify a certain hypothesis, just like Wikidata pioneering and CirrusSearch tests were not.
Change 168915 merged by jenkins-bot: Set $wgMFAnonymousEditing = true for Italian Wikipedia in November-December 2014 https://gerrit.wikimedia.org/r/168915
