In T306018#8142946, @Sgs wrote:

In the event details mentioned in the description what are the expected values for:

• block type: I see there are 5 types in the Block.php interface. Are these the expected values for this field? Should they map to a meaningful string?

In T306018#8143015, @mpopov wrote:

Oh it just occurred to me that the http.client_ip fragment thing perhaps doesn't work as expected for server-side events?

Reopening this because we've learned that the pulled data doesn't contain Add a Link rejections where multiple reasons were given. Add a Link was updated in T292592 to switch to checkboxes and allow for multiple rejection reasons. The schema documentation doesn't reflect this. I'll file a phab task and get the schema documentation updated.

We've decided that we want to get a short report of the findings up on Mediawiki, which I'll time box and work on next week.

A first pass of this analysis is completed and shared with the pilot group.

We have a draft report that's being reviewed by @KStoller-WMF and the team. If we find that we have what we need, we'll write up a summary either in this task or on wiki.

In T310320#8068399, @Tgr wrote:

In T310320#8023299, @nettrom_WMF wrote:

I think using the HelpPanel schema makes sense for this as it's already used to log post-edit dialogue events.

What happens if the help panel is disabled on that wiki? It's unlikely to happen in practice, but theoretically possible.

I guess we can just skip logging in that edge case.

The roughly 286k page views reported for the landing pages on mobile were roughly 50k lower than what we'd expect based on reports from our social media provider. I dug into various resources to gain an understanding of possible sources for this difference, paying particular attention to how we use wmf.webrequest as our data source since that also affects two other analysis that are currently ongoing (T308774 and T308784).

I'm reopening this task as we're looking into explanations for differences in counts reported from the social media provider and our analysis.

This is waiting for T312686 before data gathering can start. Moving it off the Growth sprint board until it gets unblocked.

@Isaac has made a preliminary investigation into this for English Wikipedia from May 2022. Adding the query he used and the results here so we can build upon it.

Declining this for now. We have a rolling 90-day window of data available, which should suffice for a lot of use cases. The task can be reopened if needs change, of course.

I'm going to decline this for now as we 1) haven't gotten around to it, and 2) there hasn't been any new developments indicating that we need to prioritize it. If this changes @MRaishWMF , please let me know and we'll pick it up and schedule it.

Moving this off the PA Kanban board for now while we wait for it to unblock.

Moving this off the PA Kanban and Growth sprint boards as this work has to wait for a bit until I can pick it up again. It's important that we make these results available to the community, and we'll need to schedule this work together with other things for the current quarter.

This analysis has been presented to the Growth team and so it's time to wrap this up as resolved.

This analysis has been presented to the Growth team and so it's time to wrap this up as resolved.

In T306845#8030799, @MShilova_WMF wrote:

@nettrom_WMF , same question here - should we mark it 'Resolved' on the Growth board?

The first pass off this analysis is now complete. We've gathered data from the Spanish Wikipedia from May 3–31 for the two campaigns social-latam-2022-A and social-latam-2022-B. In this analysis we focus on actions taken on the mobile platform. For more details about the data gathering methodology and why desktop platform actions are disregarded, see the notes below.

In T310320#8022721, @kostajh wrote:

A. Allow new accounts that were created mid-edit to complete their edit before showing a dialog to go to the homepage.

@nettrom_WMF for instrumentation purposes, should interactions with this dialog use the existing analytics/legacy/helppanel schema? That's what we use for post-edit dialog events. It might be a stretch to include events for this in that schema, but creating a new schema seems like a lot of overhead for this one component.

I needed something like this for T309036, so I wrote a version that does this on a per-day basis. This can then be modified to do hourly queries by adding freq = 'H' to the parameters of the pandas.date_range call. If we want the right end of the range to be exclusive we'd also add inclusive = "left" to the parameters. I'll file a pull request for this, in the meantime here's the code for the day-based version with an example call and imports that make it work:

In T304718#8015995, @kostajh wrote:

@nettrom_WMF is there a public report that could be linked here?

This analysis is complete and the results have been shared with WMDE.

We've completed the analysis report and published it on MediaWiki, and it'll be incorporated into the Add a Link project page by @KStoller-WMF. I've also gone through the Jupyter notebooks and they're uploaded on GitHub in this high-level-metrics folder. Closing this as resolved.

Reassigning this to @Sgs while he works on implementing this, I'll support that work as needed. Also moving it off the Product Analytics kanban board and into Tracking on our team's main board while it's not assigned to me. If things change, we'll move it as necessary.

@Miriam : I'm trying to gauge what the current needs are with regards to this data. We never set up a recurring export of image decisions and have a revolving 90-day window of data available through the image_suggestion_interaction schema. If you or someone you collaborate with needs this kind of data, will you grab it yourself from the Data Lake (perhaps working with me to identify the right events), or should we be looking at getting a more permanent export set up so we have data over a longer period of time? Asking so that I can understand whether I need to discuss prioritizing this with @KStoller-WMF.

In T301652#7976433, @kostajh wrote:

@nettrom_WMF @KStoller-WMF so I think the least time consuming plan would be:

• stop the add image experiment
• define multiple new variants for add link as defined above

Otherwise, we would need to do T288022: Support multidimensional user variants in GrowthExperiments, which is probably a medium sized task. How would you like to move forward on this?

I filed this task and am notifying @KStoller-WMF about it so she can fill out the necessary information. I don't think SSH access is needed at this time as Superset & Turnilo provide access to similar data.

In T286816#7976460, @MShilova_WMF wrote:

@nettrom_WMF , are you still working on this task?

I've completed a first pass of this analysis. As the task description mentions, the approach here was to gather data from Arabic, Bengali, and Czech Wikipedia prior to the bug going into effect, and compare that with a similar dataset right after deployment. In this case, I chose to use two weeks of data post-deployment, as that's roughly the amount of data we have for the prior period.

This presentation has been held, and I've uploaded the slides to Commons.

@MShilova_WMF : thanks for reminding me about this, and apologies for dropping the ball on this one.