Page MenuHomePhabricator

Analytics & Data information Documentation
Closed, DeclinedPublic

Description

We need more robust and useful documentation regarding how to use Data from an analysts perspective (or anybody writing queries).. like creating a Data/Product Analytics FAQ or one documentation that’s more clearly organized by topic.
A page with things like why some fields use structs, the fact that the fields are mapped to from JSON types, what controls that mapping, general table layout, how the whitelist works, how to do historical queries (where you'd put an infobox or note about hey, if you're using fields that got sanitized, then there will be NULLs) etc.

It would be helpful to start collecting the questions we get (when we get questions), from PMs and others, and figuring out what to do with it. It would be good to start talking about a portal: good to have one place to remember, and be able to access all relevant documentation for there. Like on a Product Analytics Wikitech page..

Event Timeline

Mayakp.wiki renamed this task from Product Analytics Documentation to Product Analytics data information Documentation.Nov 19 2019, 7:52 PM

Neil had opened T193296 back in 2018 with the intention to create a data portal

mpopov added a project: Epic.
mpopov moved this task from Needs Investigation to Epics on the Product-Analytics board.
kzimmerman triaged this task as Medium priority.Feb 24 2020, 8:51 PM
kzimmerman renamed this task from Product Analytics data information Documentation to Analytics & Data information Documentation.Jun 30 2020, 12:26 AM
kzimmerman updated the task description. (Show Details)

@nshahquinn-wmf and @Mayakp.wiki are thinking of taking a stab at this task for Wikimania Hackathon this year (Aug 13, 2021)

@nshahquinn-wmf and @Mayakp.wiki are thinking of taking a stab at this task for Wikimania Hackathon this year (Aug 13, 2021)

It's official! T288680: Improve the documentation on research, data, and dashboards on Meta-Wiki

The focus is on documentation useful to the movement as a whole (but if you want to hang out and work on documentation about our internal data infrastructure, that's cool too!).

I think this is defunct.