Page MenuHomePhabricator

[SPIKE] Model Impact of IP masking on datasets
Closed, ResolvedPublic3 Estimated Story Points



The Anti-Harassment Tools team is leading the cross-department program for implementing IP Masking. The IP Masking program kicked off in 2018 and has been recently reprioritized as an important workstream as concerns around data use and privacy are growing around the world with various governments introducing data laws. It is aimed at improving our users’ privacy, ensuring that they can contribute to our projects effectively and safely, as well as to address legal risk around publishing and retaining IP addresses.

The Data Products team needs to understand the impact that changing anonymous users to temp users will have across our datasets. Notable datasets that will be impacted are:

  • MediaWiki History
  • Geo Editors
  • and Data Dumps

Acceptance Criteria

  • Starting list of datasets that will be impacted
  • Break the work down into epics based on known compatibility issues within affected data sets with rough estimation of time to complete
  • Findings report that can be used to prioritize and estimate work required to mitigate the effects
  • Decision record outlining the thinking behind a final go / no go decision


Event Timeline

VirginiaPoundstone renamed this task from Model Impact of IP masking on datasets to [SPIKE] Model Impact of IP masking on datasets.Aug 25 2023, 5:08 PM
WDoranWMF reassigned this task from dan to Milimetric.
WDoranWMF moved this task from Sprint Backlog to In Process on the Data Products (Sprint 00) board.
WDoranWMF added a subscriber: dan.
Milimetric set the point value for this task to 3.Aug 28 2023, 2:11 PM
VirginiaPoundstone updated the task description. (Show Details)

@Milimetric I added a link to the work breakdown spreadsheet in the acceptance criteria list. Please add your rough complexity and time estimates.

Re-adding subscribers that I accidentally removed! No idea how I managed to do that. Apologies.

@Milimetric thanks for adding time estimations to the workbreak down. Very helpful. Please also add complexity scores (fibonacci). Thank you.