I'm failing to find a simple answer to the pageviews thing. I did discover that the Popups pageviews include Special page and Edit page impressions, but these alone shouldn't add up to the 2.5x disparity found in the real data. I'm going to wrap up this phase, and put a warning on the graph with suspect data. If anything, the difference will *under*-estimate the impact of Reference Previews, and we're already getting much higher popup view rates than baseline footnote views.

I think the confusion is caused by a late failure. The original error is from composer install, where one dependency requires wikimedia/assert:v0.4.0, and then wikibase-lexeme requires wikimedia/assert ([>= 0.2.2.0-dev < 0.3.0.0-dev]). Composer reports,

14:48:37 [500.2MB/9.66s] Your requirements could not be resolved to an installable set of packages.
14:48:37 [500.2MB/9.66s] 
14:48:37   Problem 1
14:48:37     - The requested package wikimedia/assert ~0.2.2 is satisfiable by wikimedia/assert[v0.2.2] but these conflict with your requirements or minimum-stability.

Indentation could be split out into its own patches, if that makes the implementation easier.

Demo in beta:

Something's wrong here, for example I see percentages well over 100% for the anon group. Can you spell out what the math is supposed to be doing?

There are many more Popups pageviews than I had expected (or many fewer from Cite). Look into this, and tweak the tracking to be idempotent in case that's the issue.

The dashboard is ready for review. One think I feel uneasy about was my decision to calculate the references viewed for RP like this:

I'm moving into the current workboard since this is a user-facing crash.

Step change is a step change!

Draft dashboard can be found at https://grafana.wikimedia.org/d/aHQQ_20Wz/reference-previews-usage?orgId=1

In T233108#5643443, @awight wrote:

I'm debugging why baseline data isn't landing in Grafana yet. We're averaging c. 1.5 events/second on the baseline topic, which should come out to about 140k events per day. That was what we were seeing until Nov 1st, at which point the counts dropped to a few dozen.

I'm debugging why baseline data isn't landing in Grafana yet. We're averaging c. 1.5 events/second on the baseline topic, which should come out to about 140k events per day. That was what we were seeing until Nov 1st, at which point the counts dropped to a few dozen.

In T236886#5640321, @JStrodt_WMDE wrote:

In T236886#5640239, @awight wrote:

Needs grafana follow-up.

Is there a ticket for this already?

We're discussing rewriting this ticket without the proposed, intermediate implementation. We would go straight to indented and sub-numbered references.

We need to document the decision taken and methodology.

Needs grafana follow-up.

Test edit here: https://en.wikipedia.beta.wmflabs.org/w/index.php?title=User:Adamw/sandbox&veaction=editsource

Need to tweak the ReferencePreviewsCite report...

In T237063#5638384, @DLynch wrote:

Thanks for looking into it! I'll just mention that, the schema can be changed to make the fields optional, which will then give you proper metrics showing how often this condition occurs.

I'm pretty sure they're already optional, or at least they don't have required: true. It's just that EventLogging doesn't treat a null that makes its way into the event as being equivalent to the value not having been provided.

In T228792#5636215, @Tsevener wrote:

@awight Sorry for the late response - the endpoint was implemented here.

NHarateh_WMF moved this task from Tasks from Product Backlog to Needs Code Review on the iOS-app-v6.5-Squid-On-A-Tandem-Bike board.

Thanks for looking into it! I'll just mention that, the schema can be changed to make the fields optional, which will then give you proper metrics showing how often this condition occurs.

@GoranSMilovanovic Some last notes from a conversation with @AndyRussG: WMF is still using the beacon/impression numbers, so we can rely on them for this year's campaign.