Page MenuHomePhabricator

[REQUEST] anonymous editors data request
Closed, DeclinedPublic

Description

What's requested:

Research questions for background
Anonymous editors: who are they, how many of them are there, and how should we engage them?
Who are the anonymous editors?
How many are there?
What share of productive edits (scale and impact/productivity overall) are made by anonymous editors? (across wikis)

Please help brainstorm other ways there may be to break down this data!

How many are there?

  • Unique users/devices from year 2019

How else can we break down the data we do have?

  • Platforms being used (mobile, desktop, other?)
    • Tools being used if applicable, indicated by edit tags
    • Size of users' screen
  • Geocoding
  • Browser family (chrome, safari, and associated types/versions)
  • Referral class (internal -- internal page v external -- search engine)
  • Time ranges (daily, weekly, monthly)
  • % of total edits on all wikis
    • How to measure ‘productivity’?
      • Non reverted?
      • Size? Is typo-fixing sufficiently ‘productive’?
      • What other ways?

Why it's requested:
Toby requested this research and Growth team (and others) will utilize the findings to inform product direction. This data will provide more context of the scale of IP editors and their impact. Will also inform design research interview protocol.

When it's requested:
Connie has already provided some Superset data on anon editor edits, the above are remaining questions that require sql querying.

Other helpful information:
Research brief on the project.

Event Timeline

dchen triaged this task as High priority.Feb 25 2020, 7:19 AM
dchen created this task.

@dchen : For the question "What share of productive edits (scale and impact/productivity overall) are made by anonymous editors? (across wikis)", I did an investigation into that for AHT last fall. The phab task for that is T231605, and the output from that project is this huge page on meta-wiki. The code behind that analysis is in this GitHub repo.

It's a lot of numbers because we did it for all the wikis, and we looked at both the extent of anon edits as well as whether they were reverted, revdeleted, and whether the IP was subsequently blocked.

kzimmerman added subscribers: cchen, kzimmerman.

@dchen we don't have extra bandwidth to tackle this right now, unless this is a blocker/drop everything kind of thing (however, team members can help point you to existing resources, like Morten did above). Can you provide more information about the context/urgency?

@MMiller_WMF : could you speak to the Growth team's prioritization of this, ref @kzimmerman's comment above?

@kzimmerman hi all - this is a request that came from Toby and the analytics portion is important to answer some of his questions directly and also to inform the design research interview portion of the research.

@nettrom_WMF can we do a walkthrough meeting of 30-60 min to discuss the linked meta page? how much of the various breakdowns initially requested is addressed within your research?

thanks all!

@dchen : Sure thing, grab an open slot on my calendar! The work answers the questions of what % of edits are by non-registered users, and of those, what % get reverted (within 48 hours, only counting identity reverts), what % get revdeleted (at some point, AFAIK we don't know when it happened), and whether reverts/revdeletion leads to the IP getting blocked (within 6 hours).

I reached out to @MNovotny_WMF and haven't heard back about whether we should elevate the priority of this above current work. Related to Toby's ask: Connie is adding count of anonymous edits (but not an estimate for anonymous editors) to monthly key product metrics.

Moving to the backlog for now, as this falls under Growth and @MMiller_WMF has higher priority work for Morten.

kzimmerman lowered the priority of this task from High to Medium.Mar 17 2020, 6:41 PM

Hi, the Due Date set for this open task is more than a month ago.
Can you please either update or reset the Due Date (by clicking Edit Task), or set the status of this task to resolved in case this task is done? Thanks.

No reply by anyone, hence resetting Due Date.