@jhathaway, that's correct, thanks! Nothing additionally needed beyond that access at the moment.

Approved, end date for this access1-October-2022. The 'non-private' access level probably suffices - assuming the topic datasets are accessible there, which is the area of primary interest (to be joined on other non-private tables, for example). So, let's start with that level of access, and we'll request additional access if necessary.

I don't see a need for production shell access at this time.

Thanks @Volans, thanks @Jdforrester-WMF.

Thanks all. This is all good and well. Thank you for the support and discussion! The access to some of the things around observability and metrics is part of this, and also for the occasional submit tasks on Gerrit (with less need for per-repo perms).

Pending the implementer's confirmation of the signed NDA (please let me know in case you need me to also reach out to Legal instead), I approve.

Hi all. Following up from email thread. The request here is for delegated owner access to be granted to @SCherukuwada's Google Workspace account via the following:

That's correct. I'll follow up under separate cover.

@sbasset just wanted to say thank you!

To answer the question on the creds, no, they don't need to be shared. But delegated access will need to be established. An SRE with access to the creds should be able to visit https://www.google.com/webmasters/verification/details?hl=en&domain=wikipedia.org and grant @SCherukuwada delegated access.

Approved.

Thanks all. @MatthewVernon I'm delegating responsibility on research and response on this to my direct report, @SCherukuwada (Senior Engineering Manager, Web team), who is also familiar with similar things from GSC land and who I've also asked to help streamline things in the Yandex system as well. Thanks again!

Ah yes, thanks @Jdforrester-WMF.

I think if I've heard correctly there may be some interest in a generalized solution that can service the needs of multiple use cases (e.g., things that are like site notice or other sorts of banners, as well as the content). Is that right?

Before we touch no-transform I'm requesting that @Maryana and team make a connection with the downstream orgs to see if we can get a read on things.

@TheDJ I'm emailing out of band to coordinate days/times that may work for a Meet. Thanks!

@TheDJ if the basic functionality remains intact and it doesn't introduce breakage and is equally maintainable (which my impression is that it would be the case from what you've said and in regular discussion with @brion), then from an engineering director perspective I'm fine with releasing. FWIW, the new stuff is so much better for me from a UX perspective.

Approved, conditioned on the data remaining inaccessible except for those with NDA, need to know, and suitable strong authentication requirements where identities are under SOPs for access revocation at cessation of work arrangement (less wordy: Google Workspace 2FA for the org). Although the top 1,000 search query data are not surprising and have a high degree of overlap with actual site traffic, the resolution of the data can be more fine grained than what we typically aggregate on a geo basis, especially once actual country level filtering is applied or if clever inferences are drawn based on other dimensions and the nature of queries, hence this explicit conditioning.

For product UX and as an expectation for mobile user agent constraints especially (but it applies on classic desktop, too), I have the expectation that we won't have jarring FOUCs or other layout shifts as part of the page load and reading experience as a consequence of this. @greg cool?

Approved.

Good catch @TheDJ. Yeah, I think that was a stopgap until stuff is more broadly rolled out. @egardner @Cparle @MarkTraceur heads up - any insights to document here in this ticket re: MediaSearch appreciated for future selves.

For the sake of completeness and for future self memory, I was able to reproduce from Debian Buster the same Z802 behaviors. Here's the docker-compose.yml I used both with a host OS of Raspbian based on Bullseye and with a host OS of Debian Buster (the orchestrator and evaluator Docker images were based from node:12-bullseye in both cases).

Looks like the Z802 example directly available from the API sandbox Example {} section works just fine, whereas the old material on wiki was the thing faulting as listed in the Etherpad from the Description. @DMartin-WMF updated the wiki at https://www.mediawiki.org/w/index.php?title=MediaWiki-Docker%2FExtension%2FWikiLambda&type=revision&diff=4931967&oldid=4909538 and so maybe there's nothing to be done?

Approved

Approved

@jlinehan heads up.

@ovasileva @SCherukuwada @Jdlrobson @SWakiyama @CBogen @MarkTraceur @DVrandecic @CBlanton @Jdforrester-WMF probably good to have in at least a watching column. I got a ping to raise awareness. Others such as TProgM (hi @LGoto I saw you were already triaging a related task) or Product Analytics (hi @kzimmerman and others on task !) may broach this as well, but doing my part and raising awareness in case there are UX or feature detection or instrumentation pieces requiring attention.

Approved

@jcrespo we have not allocated a wikimedia.org email address. Is that required? If so, I'll ask ITS to provision one. I'm out the rest of the day, heads up.

@jcrespo Yes, confirmed on the contract with This Dot (Simone is a This Dot consultant working on WMF projects).

Following up from off thread discussion: we'll plan on the following based on what we know today:

Thanks all. I will approve of the logstash access on a new ticket to be filed by Simone. I gather if Simone specifies that it's Logstash, then no LDAP group needs to be specified (I see that LDAP membership an Optional field).

I though that macro was going to give a little emoji, but I guess it gives a big seal of approval literally! Thank you, in any case :)

Approved.

Would it be possible to make the Superset dashboard also browsable by additional filters for edit namespace and country?

@Gilles I touched up the Description some. As emailed, we'd like review from the 1st of June (Tuesday) to the 14th of June (Tuesday), 2021, inclusive.

@Aklapper do you think it would make more sense to get a separate board for those tasks and put them in Stalled mode?

Hi all - I think I can help. What's the ordered set of nameservers that should be used in MarkMonitor?

Approved.

This and performance review timing.

Yes, please.

Yes, it works. Thank you!

IIRC the Webmaster Central URL at https://www.google.com/webmasters/verification/details?hl=en&domain=wikidata.org ought to make it possible delegate owner on that parent domain without conferring other access given that site verification seems to be in place via a TXT record already. No need for the NOC password.

Thank you for the access. I see the "owner" access for those domains but am unable to add users. Would you please grant me access on the wikidata.org parent domain? I think that's what will be needed in order for the DNS verification of the domain to confer the actual permission to add users on the subdomains.

@brion thoughts and pointers on VLC libs?

It'll be a bit before TMH has the updates for things to Just Work™ on the mobile web. And that's still not the app. We're hoping some day a universal file format acceptable on Commons will be supported natively on iOS, but it may be a while yet and it's possible it will require higher end hardware.

Bumping this task. We would like to have a performance readiness review performed in Q4 FY 2020-2021. Exact projected requested start date to be determined and agreed upon within the Abstract Wikipedia team.

Not Phabricator tasks, but something Denny recently shared:

Here's what we discussed early this morning for a web based dashboard, in terms of the inputs and defaults. This is a rough outline, and so it's expected things may diverge somewhat for practical reasons or because of insights gained in further use and development of the web based dashboard.

@Quiddity (Nick) see the attachment on this task for data analysis so far. I'm out the next couple business days, but for our next weekly meeting my goal is to have a good "paper prototype" of a simple web dashboard for being able to drill down to modules based on some heuristics / filters.

@LostEnchanter I think I understand. I believe what you may be witnessing here is that no page uses Module:yesno (lowercase 'y') indirectly (e.g., a page using a template #invokeing some module which in turn requires yesno) on hiwiktionary, and therefore no MediaWiki parser hooks (and manually executed or scheduled link refresh jobs) run in a context where a determination is made to insert an entry into templatelinks.

@tanny411 I think your intuition is probably correct: better to skip iwlinks for now if it's causing too many problems. You can always come back later and see if there's anything of high value there. The other tables queried definitely contain some interesting stuff, we know that much!

Approved

Thinking out loud:

I was able to reproduce the new behavior observed by @Ckoerner on a number of articles on enwiki. Additionally, it seems if one shrinks the Safari window as narrowly as possible the UA is changed so one gets mdot. I was the one who originally reported the bug to Joe. Closing.

Approved.

Thanks @tstarling and @Joe. It sounds like this isn't the right fit for Wikifunctions. I appreciate the advice!

@tstarling, I was wondering if you could please weigh in on my questions above?

Not sure if it's my call, but dropping this to "Low", because although it's true that docker-compose up without the -d flag will keep repeating the error message over and over, if launching in the background with the -d flag (like DEVELOPERS.MD suggests using), I can then run docker-compose exec mediawiki composer update and docker-compose exec mediawiki /bin/bash /docker/install.sh, which will generate the LocalSettings.php.

Hi @tanny411, the Microtask is the same for all. You'll want to start with the Microtask first, and we'll select the intern based on the Microtask entries.

In T263678#6532649, @Udoka_Ugo wrote:

Okay thanks
Here is the link to my microtask, https://github.com/UdokaVrede/Wikimedia-Microtask
Please let me know of any modifications I would be required to make in this report.
Thank you

Hi @Udoka_Ugo. Feel free to post your solution to the microtask to GitHub. Thanks!