In T280798#7205364, @dom_walden wrote:

I can confirm this on beta. I am guessing it is because $this->getLanguage()->getFormattedNsText() returns '' for NS_MAIN.

@awight: I've +1'd https://gerrit.wikimedia.org/r/c/mediawiki/extensions/Popups/+/667897/. I'd really appreciate it if you could confirm that the ReferencePreviewsPopups instrumentation is still works as expected.

@ovasileva: Could you provide a working example that this has regressed from?

In T268782#7188384, @cjming wrote:

In T268782#7188374, @Jdlrobson wrote:

I think for now, the simple regex map will do the trick. In the future however, if more granular targeting for surveys is desired, we probably should consider using a proper parser lib.

In T280770#7186635, @MNeisler wrote:

@phuedx - Are there any other scenarios where this event context would be logged? I'm seeing event.context = 'content-language-switcher' events logged on the legacy vector (see data below). Is this also logged if a user switches languages after selecting the N More button in the sidebar?

In T281797#7181070, @Jdforrester-WMF wrote:

In T281797#7172769, @phuedx wrote:

I'd have thought this was a feature you'd want to flag on a language basis, but I defer to the team. :-)

As we discussed in yesterday's Web task sync I've run a query to see if the treatments are balanced:

@Jdforrester-WMF: Would you be able to do a concept review of https://gerrit.wikimedia.org/r/c/operations/mediawiki-config/+/699416/, which attempts to tag the relevant wikis (see the task description for the list). Specifically, is it necessary to create another file, disable_search_autocomplete_highlights.yaml say, simply to add a wikiTag line rather than add that line directly to the wiki config files?

@Jdlrobson: It looks like you've started reviewing @nray's patches so I'm being bold and assigning this task to you.

and issue has been filed

In T254977#7168282, @Jdlrobson wrote:

@phuedx does T245292 impact your analysis in any way?

@Edtadros: It should be sufficient to test that the survey appears as before (on both desktop and mobile) and that an event.eventName="impression" event is logged when it appears.

In T280982#7165594, @XanonymusX wrote:

I always have this happen on dewiki, where content language and interface language are both German!

Being bold.

Being bold.

Per T277534#7165409, I'm seeing the following in production when the interface language is different from the content language (e.g. https://en.wikipedia.org/wiki/Main_Page?useskinversion=2&uselang=he):

Generally, I'm not seeing this in production, locally, or on the Beta Cluster.

Firstly, let's address what we need to know in order to remove certain parts of the survey config payload ("the payload"):

In T280770#7157479, @ovasileva wrote:

Quick question on the current implementation - can we tell the difference between selecting a language from the list and changing the settings for the control group?

In T280770#7155428, @MNeisler wrote:

However, there is a caveat: it's quite difficult to thread through a value for event.context through a value so it'll always be the default ("interface"),

@phuedx To confirm, this means that both of the following events are recorded the same way, correct?:

• A user clicks the language button in the header and changes the display language setting
• A user clicks the language button in the header and switches languages

I've added a filter to the eventgate-validation Logstash dashboard (Logstashboard?!) that refers to this task. Given T283881#7152241, I'm inclined to either lower the priority of this task to Lowest or decline this task. Regardless, I'm moving this off of Readers Web's kanban board.

In T280770#7147624, @MNeisler wrote:

POSSIBLE BUG/MISSING INSTRUMENTATION:

• For sessions with clicks to the new language button in the header, I'm not seeing any language change events (event.action = 'language-change') for when a user changes language by clicking on one of the language links that appear after the user clicks the button. Was instrumentation added to track these events?

Tempus fugit

In T257695#7143016, @Jdlrobson wrote:

I'll set one up but will need help getting it deployed..

@bwang: Are you working on this? If so, could you move it into Doing.

The Message object doesn't guess at how parameters should be handled but needs to be told. Message::numParams() tags the passed parameter as a number so that it can be formatted when needed. AIUI we have to change

@ovasileva: @Jdlrobson and I are discussing how much of the survey should be visible on the screen before we log an impression. Currently, we log an impression when approx. half of the survey is visible. @Jdlrobson's patch supports changing that value if we want. Do we want to change this behaviour? If so, then how much of the survey should be visible before we log an impression?

Thanks!

I'm being bold and resolving this. If it can be replicated, then please do reopen this task.

This is also occurring for the following data streams:

Being bold and signing this off. I triggered a build and it passed as expected.

This is blocked on the next WVUI release /cc @Volker_E

In T274869#7120707, @Jdlrobson wrote:

What patch(es) need code review (I don't see any associated)?

In T282653#7092642, @hashar wrote:

Browsing again, the Use Legacy Vector option is still unchecked in my user preference and the pages show properly.

Being bold and signing this off and broader instrumentation QA and sign off is happening over at T280770: Instrumentation QA for language switching .

Being bold and signing this off and broader instrumentation QA and sign off is happening over at T280770: Instrumentation QA for language switching .

I discussed this with @ovasileva and we concluded that we can backport that above changes next week and then wait until the week after to enable the A/B test.

In T267256#7116550, @Jdlrobson wrote:

Were there issues implementing this or is this clear to implement it's just a matter of priorities?