In T433768#12336923, @Ladsgroup wrote:

I think one way to reduce the confusion is to wrap the magic word with a template the beside adding the magic word, explains also what's going on. This is exactly how hidden cat is done: https://en.wikipedia.org/wiki/Template:Hidden_category

So I created https://en.wikipedia.org/wiki/Template:Catsort-by-reverse-timestamp and https://en.wikipedia.org/wiki/Template:Catsort-by-timestamp

HIDDENCAT, NOTOC, and NOEDITSECTION are all very different key words since they are immediately obvious in their impact in a way that CATEGORYSORT is not. We don't explain how we're sorting categories, so having some randomly ordering them by date will confuse people. I like the original idea of having this as a drop-down menu choice.

I'm missing the context here then - where was this proposed on-wiki? Only using this in maintenance categories sounds reasonable to me - is there a way that could be enforced? I'm more worried about where it might be randomly applied to content categories.

I like this as an option that users can turn on when it's useful to them, but having it as an arbitrarily-implemented default in some categories but not others will probably lead to confusion. It sounds more like a user preference for readers than it does a magic word that editors that can force on everyone accessing the category.

Thanks! I think I saw this happening back then but hadn't realised the connection with this query. I had to change 'cllt.lt_it' to 'cllt.lt_id' (just typo), otherwise this query seems to be working nicely, and Pi bot's happily adding infoboxes again, thanks!

How does this work on Commons - can it sort by the photograph date? Also, what happens if pages are bulk-removed and then re-added to a category?

... of course ... because nothing can be simple, sorry ...

The same applies for other multilingual sites, though - inter-wiki links to Commons, Wikidata, Meta, etc. - so I'd have thought Abstract Wikipedia links would just go through that system?

It was fixed in the Infobox by excluding Abstract Wikipedia from the list of wikis that the Infobox adds interwiki links for (e.g., if the Commons category sitelink is on a category item on Wikidata, but the Abstract Wikipedia sitelink is on the corresponding topic item). Which is not the ideal resolution...

In T424403#12032979, @Olea wrote:

Dear @Mike_Peel: is there any place where I can read the different tasks Pi bot can do? Like recurring maintenance tasks, EXIF importing into SDC, etc.

All updated to log in, and all running again. Closing this now.

Current status:

• Pi bot is reinstalled, now using the latest Raspberry Pi OS and pywikibot version
• I'm modifying scripts to ensure that they are logged in, and moving them into https://github.com/mpeel/wikicode/tree/master/pibot
• My IP has changed, so I don't want to restart non-logged-in scripts until they're updated as that will likely quickly result in an API block...

This is a follow-up from a discussion with Effie Mouzeli at the hackathon today.

Need to investigate why Pi bot is not always logged in, and ideally also reduce the number of queries it is making, per https://meta.wikimedia.org/wiki/User_talk:Mike_Peel#Pi_bot_may_be_affected_by_rate_limits ....!

duplicate of T421955?

I'll be attending May hackathon, if that might be a good time to work on this? (I can't make the NWE one.)

MediaWiki already checks hashes of new uploads against existing files and prevents uploads of duplicates from what I've seen when using UploadWizard. I've also heard of ongoing projects to find duplicates using hashes, but I can't remember who presented about that at recent hackathons... There's also https://github.com/bio-guoda/preston/issues/239 .

Hi, I run the "New Wikiquote article and category matches", "New Wikipedia article and category matches", and "Commons category matches" Wikidata games (connected to Pi bot activities). It's great to see this project taking place, the games definitely need some love and expansion!

Thanks Matthew, I'll see how it goes!

I'm getting 503 errors again today while uploading to Commons.

Thanks for this significant improvement! I suspect most people will be familiar with spreading things over multiple pages, unless they've young enough to have only encountered endless scrolling?

Maybe @Alchimista can help with the .pt domain?

In T401438#11069994, @BCornwall wrote:

Emails sent to @Mike_Peel and @geraki and brazenly subbed them here too :)

At the start of the session we talked through the examples of:

• https://commons.wikimedia.org/wiki/File:At_Wikimedia_Hackathon_(MP)_2024_139.jpg - we have many photos of hackathon attendees like this on Commons
• Such as at https://commons.wikimedia.org/wiki/Category:Wikimedia_Hackathon_2024_-_Closing_session
• And https://commons.wikimedia.org/wiki/Category:Wikimedia_Hackathon_2024_-_Opening_session
• Some of these are categorised as showing people: https://commons.wikimedia.org/wiki/Category:People_at_Wikimedia_Hackathon_2024
• Some have named people at the event, e.g., https://commons.wikimedia.org/wiki/Category:Andrew_Lih_at_Wikimedia_Hackathon_2024
• Some Wikimedians have Commons categories like https://commons.wikimedia.org/wiki/Category:Andrew_Lih - and Wikidata items like https://www.wikidata.org/wiki/Q1113332
• Andrew's notable, so he has no escape from a Commons category and Wikidata item
• What about less notable attendees? E..g., https://commons.wikimedia.org/wiki/Category:Valerio_Bozzolan & https://www.wikidata.org/wiki/Q125953097
• Should all attendees have Commons categories and Wikidata items? Or should only some have them? What about descriptions?

That would be great @debt, thanks!

"The MediaWiki error backend-fail-internal occured: An unknown error occurred in storage backend "local-swift-eqiad"." ...

See https://www.mediawiki.org/wiki/Outreachy/Past_projects#What's_in_a_name?_Automatically_identifying_first_and_last_author_names_for_Wikicite_and_Wikidata for the project outcomes. I've replied over at https://www.wikidata.org/wiki/Wikidata:Bot_requests#Request_to_add_first_and_last_name_to_items_about_humans_%282025-02-11%29 -

This will be great! There's a bunch of issues over at T366607 that need resolving, and that also links to the start of the community discussion about this (at least one more round is needed before the community will be happy with the change, I think).

I retired tasks 17 and 20 back in December, and haven't spotted any complaints since, so am marking this as resolved. If I get time at some point, Pi bot could probably still do more to use mul, but that's in the longer grass at the moment.

In T316797#10410059, @Arian_Bozorg wrote:

Hi Mike, can you give me some details of the browser and the width of the screen?

In T316797#10394206, @Arian_Bozorg wrote:

This seems to be fixed now

In T62928#10230819, @SD0001 wrote:

To turn on syntax highlighting, click the highlighter icon in the toolbar.

This is quite frustrating for dvorak users ... any timeframe for it being reverted / help pointing out how to turn syntax highlighting on? Surprised this wasn't mentioned in yesterday's Tech News.

Thanks! Let's follow up there.

In T371774#10040734, @Bugreporter wrote:

They may be copied to mul labels instead if one does not exist.

oh, this is an old ticket. Nowadays the infobox automatically shrinks when you access it on mobile (most fields are just hidden), so I'm not sure this is still needed. Unless it would be better to autocollapse it to a single line and have the full infobox available on demand?

Uploads have been going much better today, thank you. :)

Uploads have been going much better today, thank you to whoever fixed this. :-)

Also getting errors while uploading:

Following the merge, can this be broadened to include all devices (e.g., meta quests), not just iOS?

Is the restriction on the page width relevant for Wikidata? I thought it was aimed at large amounts of prose, like Wikipedia articles. Could it just be turned off by default?

Sorry for the slow response. I'm not sure, it's an area I'd like to investigate more, but then there is T365208 and T246901, so I'm not sure if this ticket is also useful to keep active.

Work that @Mike_Peel can help us with:

• We need to update https://commons.wikimedia.org/w/index.php?title=MediaWiki:Filepage.css with the following:

table.commons-file-information-table, .fileinfotpl-type-information, .fileinfotpl-type-artwork,
.fileinfo-paramfield {
      color: #202122;
}

"hack stylesheet" - can you be more specific? That doesn't sound good for the long term.

I'm happy to make edits to fix this on-wiki (I'm an interface admin on Commons), I'm just not entirely sure what needs changing. :)

I'm still working through the processing of the photos I took, hope to upload them soon.

useful links:

• https://www.khronos.org/gltf/
• https://www.linuxfoundation.org/press/announcing-alliance-for-open-usd-aousd

https://developer.apple.com/augmented-reality/quick-look/ exists - not quite VR, but AR. Could work for MediaWiki? Possibly needs USDZ (for apple) and GLTF (for Android/Quest) versions of the 3D models?

Trying to figure out ways of linking to existing 3D content on Commons. For example, there is https://commons.wikimedia.org/wiki/File:Parthenon.stl - a 3D model of the Parthenon. It's possible to view it in 3D on Commons, if you click the '3D' button top-left, then you end up at https://commons.wikimedia.org/wiki/File:Parthenon.stl#/media/File:Parthenon.stl . Testing on Wikipedia though, you get https://en.wikipedia.org/wiki/User:Mike_Peel/Sandbox2 - clicking 3D takes you to commons, and you have to click again on 3D to actually view it in 3D... The 3D models are linked to in Wikidata items, though, using Property:P4896 - so there is a way of finding them from a Wikipedia article fairly directly. They are also in the Wikidata Infobox on Commons.

Copying over the report from the other ticket, in case that helps resolve this:

Oops, I'd forgotten about the previous ticket, thanks for merging. Also apologies for accidentally changing subscribers, restoring those.

In T209310#9715995, @Nardog wrote:

Not sure what still needs to be done about this technically. The deprecation of the old module can be done by volunteers on wiki.

In T209310#9715656, @Taylor wrote:

Working link: https://www.mediawiki.org/w/index.php?title=Extension:Scribunto/Lua_reference_manual&diff=prev&oldid=6470970

Indeed the global deprecation of "Template:=" "Template:!" and "Module:No globals" should be stronger.

The discussion at Commons is indeed silly. How complicated can it be to remove a blatantly outdated comment line from a module?

After being abruptly told to make an edit on Commons to remove Module:No_Globals from WikidataIB (https://commons.wikimedia.org/w/index.php?title=User_talk:Mike_Peel&diff=prev&oldid=868602169), I finally traced the logic back to this ticket. I've added a link to this from the documentation [https://www.mediawiki.org/w/index.php?title=Extension:Scribunto/Lua_reference_manual&diff=prev&oldid=6470970] to help others find this ticket, and will make the edit, but there needs to be way better outreach to migrate all the uses described at https://www.wikidata.org/wiki/Q16748603 to this new solution.

In T360360#9675009, @Bugreporter wrote:

Note if (in the future) one day we add chapter wikis to Wikidata, this will be embarrassing since they are not "multilingual" sites.

Thanks for working on this @Amire80! It looks like a good way forward. Maybe good to subscribe those following T102417 to this follow-up ticket, to keep people in the loop?

Migrated.

Many thanks @taavi - I managed to figure it all out today, particularly using that last link. It seems it's basically a complete change from the old system. I'll keep an eye on it, but hopefully it will run daily using the new system from now on.

The setup works fine with jsub, how do I just use the same environment with this new setup?

Can't I use my local environment, which loads when I log in?

Ah, well spotted, thanks! That moves the problem on a stage, where it now says:

So I must be missing something, but I was running this at /data/project/pibot/wikicode:

Oh, me again. On it...

I'm the operator of Pi bot, I'll check into this as soon as I can.

In T345783#9333011, @Legoktm wrote:

Switching to path-based input is something that tool developers would need to support in each tool

I'm (personally) not sure this is something that would only be solved with a research paper, it would probably also need the development of a generally available tool to enable future analytics in this area - although those probably aren't decoupled things.

Thanks @Legoktm - that seems to work for Scholia. However, it doesn't work for Resonator - try https://iw.toolforge.org/reasonator/Q1513315 vs. https://reasonator.toolforge.org/?q=Q1513315 . And then there are more complex cases like https://glamtools.toolforge.org/glamorgan.html?&category=South_Pole_Telescope&depth=1&month=last or https://wikishootme.toolforge.org/#q=Q1513315&main_commons_category=South_Pole_Telescope etc. (and other links at the bottom of https://commons.wikimedia.org/wiki/Category:South_Pole_Telescope - or the infobox in most commons categories).

In T14019#9299831, @Umherirrender wrote:

Some base work in core has started and is linked with T268526: Use a dedicated mechanism to track page dependencies
But without support for Special:WhatLinksHere, the data are internal to mediawiki.

Please do make sure to submit your final application via the Outreachy website before the deadline! That is the only way we can formally receive applications to consider for the project.

Just to be clear - make sure to submit your contributions and final application on the Outreachy website before the deadline! It's fine to still include contributions there if they haven't been reviewed or if you've made changes since you last got feedback on them. But do make sure you submit your final application before the deadline!

Just to be clear - make sure to submit your contributions and final application on the Outreachy website before the deadline! It's fine to still include contributions there if they haven't been reviewed or if you've made changes since you last got feedback on them. But do make sure you submit your final application before the deadline!

OK, I think linking is fine, I just wanted to avoid a situation where you submitted something here but didn't fill in a final application on Outreachy, which would be bad. :-) Normally I've just seen written timelines posted on Outreachy (along the lines of 'Weeks 1-2, X, Y, Z activities', with weeks on different lines), I guess the Outreachy system isn't so friendly for tables.

Hi @Shashi.coding Please make sure to post this on the Outreachy website directly, since we can only accept applicants through that process. You must make a final application on the Outreachy website. I'm closing this phabricator task as invalid.

Hi @Anne you need to submit this on the Outreachy website, not here. Closing the task as invalid.

In T313467#9271763, @ItamarWMDE wrote:

We can definitely release it a day or two later. This will give us a chance to review the release before it goes out.