In T220456#5131546, @Nuria wrote:

@Neil_P._Quinn_WMF the private wikis are not included on the labs replicas and that is intentional, if you notice we also do not report pageview data for those wikis either. This is at the request of users of those wikis and seems like the right policy as there are wikis of delicate nature there, these wikis will continue not to be included.

In T220507#5110485, @JAllemandou wrote:

In order to be as explicit and precise as we can, I will add a field for users named userRegistrationTimestamp, and we'll have 3 fields:

• userRegistrationTimestamp - user_registration from user-table
• userCreationTimestamp - Timestamp of user-create event in logging table
• userFirstEditTimestamp - Timestamp of the oldest revision associated to the user (by user_id), whether in revision or archive table.
  
  Then in the event-lineage, we'll put the minimum value of those three.

@Nuria, @JAllemandou, this is the most significant issue that I identified in T218819 (T218463 is also major, but you've already fixed it 👏).

I have now identified pretty much all the discrepancies; they're listed in the description, along with the link to the full notebook of my results.

We have decided that, sadly, we don't have the capacity to finish phases 2 and 3 during this fiscal year (by June), so we have deprioritized them. We would still like to do them someday, so they live on as separate tasks (T221563 and T203033).

This was part of the wiki segmentation output from the 2018-19 annual plan (T188391), but we've decided to deprioritize it for now. Hopefully, we'll be able to take it up again within the next year.

I've updated this task to reflect the fact that we have decided to deprioritize it. We (sadly) won't be doing it during the current fiscal year, although we do hope to take it up at some point in the future when we have more capacity.

Thanks, @mforns! Looks good to me as well.

In T211173#5116048, @mforns wrote:

If you're OK then, I will move on with CRs and deploying this version.

Thanks, @JAllemandou! Sounds like a good plan.

In T206894#5022937, @Nuria wrote:

@mforns restarted job, this one shoudl have correct e-mail: https://hue.wikimedia.org/oozie/list_oozie_coordinator/0147256-181112144035577-oozie-oozi-C/

I just checked this out. Since the experiment started, we haven't seen any clear change in the number of mobile visual editor sessions or their completion rate in either the treatment or the control group. The number of mobile VE users in each bucket is very close to even. So, everything looks very sane. Full numbers in this spreadsheet.

In T196839#5101379, @Esanders wrote:

It's slightly higher among editors at 1.6%:

In T196839#5108112, @Esanders wrote:

Deployed on prototype server: https://visualeditor-prototype.wmflabs.org/wiki/Pride_and_Prejudice

There is still a slight flicker whenever the cursor is placed anwhere past the first page of the document, but it is usually almost unnoticeable.

In T196839#5099862, @ppelberg wrote:

4. Caveat to user agent breakdown data: as best I can tell, we don't know what browsers, operating systems and devices contributors are using. Although, maybe we can assume FF usage is equally small considering this issue has persisted for >1 year without our noticing/re-engaging with it.

In T218851#5076149, @ppelberg wrote:

The Section Editing A/B test will run through 14-May-2019.

In T218349#5066859, @Neil_P._Quinn_WMF wrote:

I agree with @Jdlrobson that there's no point rewriting the database just to change the tag from "mobile web edit" to "mobile web action". For such a minor change, updating the display version of the tag (which can differ substantially from the database name: see w:en:Special:Tags for an example) is sufficient.

Closing since fiscal year 2017-18 ended 9 months ago.

Closing since fiscal year 2017-18 ended 9 months ago.

Closing since Closing since fiscal year 2017-18 ended 9 months ago.

Closing since fiscal year 2017-18 ended 9 months ago.

Closing since fiscal year 2017-18 ended 9 months ago.

Closing since fiscal year 2017-18 ended 9 months ago.

Documentation added to mw:Wikimedia Audiences/Data dictionary#Contributors.

The bucket field is now whitelisted, so I think this is done!

We actually don't need this since VisualEditorFeatureUse relies on EditAttemptStep for all the session-related data.

Yes, it is!

@Ottomata, @JAllemandou it sounds like the long-term vision you have is using ISO strings for EventLogging and similar data streams and Unix timestamps for "Hive-native" datasets like mediawiki_history.

In T219542#5066946, @Nuria wrote:

I have no problem myself on making those guidelines public, if you want to move the document to wikitech please do so. Now, they really give the wrong impression that anyone can file a ticket and get access to this data (this is a regular request) while that is not the case.