@JKatzWMF are we still using the MobileWebMainMenuClickTracking table/data? The code hasn't been touched in years and apparently is not working on desktop Minerva.

@ovasileva is a better judge of current use than me. I can't speak to the original intent, but I used it to gauge the popularity of the various features on the site, which seems like a valid thing to know. It doesn't change much though, so its not a very interesting dashboard. Its funny that you mention it though, because I was just complaining that we don't have a similar metric on desktop and therefore don't know how to trim the infamously lengthy left-hand nac.

@ovasileva can we get rid of this schema?

@Jdlrobson - I can't find the actual task, but didn't we identify some issues with this schema when we were looking at the settings changes?

That was a different schema (MobileWebSettings). This schema in MobileWebMainMenuClickTracking - which has been neglected, probably doesn't work and appears to be unused.

Unless you or @Tbayer have used this schema in the last 6 months, I propose we remove this code from mobile.

In T193505#4416514, @Jdlrobson wrote:

That was a different schema (MobileWebSettings). This schema in MobileWebMainMenuClickTracking - which has been neglected, probably doesn't work and appears to be unused.

A schema named MobileWebSettings doesn't seem to exist, did you mean something else?

Unless you or @Tbayer have used this schema in the last 6 months, I propose we remove this code from mobile.

Thanks for the ping! I don't recall having used it so far. That said, an EL schema doesn't have to be used regularly to be valuable. It's often very useful to be able to resort to an existing schema for a new data question instead of having to build a new instrumentation. How likely is it that the instrumentation is broken at this point?

As far as I recall, usage of Minerva on desktop is still comparatively very low, does anyone have recent numbers? (I might have some soonish in the context of T189554.) So in principle I would be skeptical of removing instrumentations or features just because they don't work there. I agree though that data integrity/code rot could be an argument.

How likely is it that the instrumentation is broken at this point?

I dont seem to have access to it but we would be able to determine if it's still working quite quickly by checking if new entries are coming in for every single item in the main menu.

A quick (totally non-exhaustive) check of the current data:

SELECT event.name, COUNT(*) AS events 
FROM event.MobileWebMainMenuClickTracking 
WHERE year = 2018 AND month = 7 AND day =11 
GROUP BY event.name ORDER BY events DESC LIMIT 10000;

event.name	events
home	34902
random	34304
settings	15970
login	5767
nearby	4215
watchlist	1771
profile	1001
contributions	901
logout	341

Thanks for checking @Tbayer ! So it looks like this is working as it's returning results for all the things I'd expect it to be. Note, schema is quite flexible so it could also be used in future for logging any new items we add to the menu.

I've updated the two options we have here. @ovasileva I don't mind what we do here, but we should do one of the options. Please delete as appropriate and move it to estimation!

Ping @ovasileva and @Tbayer we are in the process of modernising this code in T204584. Several patches will be needed.
I'd appreciate a speedy decision here - whether to remove it or keep it - with reasoning. It will help us think about how to think of this code going forward. Option 1 would also simplify T204584 a lot if we decide to do this...

Thank you!

@Jdlrobson - I feel like it will be very useful because we'll want to do similar instrumentation for the new navigation so I would say let's go for option 2 (unless reusing this for the new navigation ends up being harder than building it out again).

Thanks.
I've pulled out "Consider lowering the sample rate to reduce the number of events we see." - we can deal with that later.

@Jdlrobson added projects: patch-welcome, Google-Code-in-2018.

@Jdlrobson: Testing for GCI, I fail to reproduce this. I go to https://www.mediawiki.org/wiki/Phabricator?useskin=minerva&debug=true and clicked the hamburger menu, notifications bell, and language selector but no such output in the console of Firefox 62 on desktop Linux. Am I doing something wrong? Is this still an issue?

@Aklapper did the description change help?

Jdlrobson: Yes, thanks! Imported as https://codein.withgoogle.com/tasks/5542909697851392/

Unassigning from @ovasileva. I think this was a mistake.

Change 475917 had a related patch set uploaded (by Pmiazga; owner: Pmiazga):
[mediawiki/extensions/MobileFrontend@master] Bump the MobileWebMainMenuClickTracking schema revision

https://gerrit.wikimedia.org/r/475917

Change 475922 had a related patch set uploaded (by Pmiazga; owner: Pmiazga):
[mediawiki/skins/MinervaNeue@master] Pass mode property to MobileWebMainMenuClickTracking schema

https://gerrit.wikimedia.org/r/475922

Change 475917 merged by Jdlrobson:
[mediawiki/extensions/MobileFrontend@master] Bump the MobileWebMainMenuClickTracking schema revision

https://gerrit.wikimedia.org/r/475917

Change 475922 merged by jenkins-bot:
[mediawiki/skins/MinervaNeue@master] Pass mode property to MobileWebMainMenuClickTracking schema

https://gerrit.wikimedia.org/r/475922