I'm continuing to use this sporadically to back up pages I can manually find, which is surely a small portion of the addressable space. A sizeable % of historically significant pads are not yet backed up.
For instance, found today: /covid19 and /Covering_George_Floyd .

+1 for experimenting with this. The outputs of a completed Polis question can be migrated onto Meta for archiving and defragmentation; it's a different tool that we don't have an equivalent to on-wiki.

@Zanahary great idea, openverse is fantastic. :) Maybe you and @kaldari could collab on a toolforge implementation that supports a few known image sources.

I don't know how the projected 90-day expiration will be implemented or made visible to editors. ( see T421316 )

Bumping this one, seems relevant to unblock. There are few enough thoughtful research projects that take care to create a Research: page on meta.

cmd+shift+a for [a]nnotations? In a solarpunk future where we all have rich marginalia, I could imagine setting a pref for what marginalia I want to see and this pops them out for an enhanced reading experience.

@bmartinezcalvo cool! If you do that I'd greatly appreciate expanding to popover the page text rather than forcing a reflow or moving the main text, which is visually jarring and would make me turn it off.

Looking at some other languages, is there a standard css file for displaying the print version? I see enough variation that there may still be problems in some languages, but spot checking 20 I didn't find a major one. Here's what I think was / continues to happen:

• the WP logo, in the "logo + subtitle" form, shows up once at the top of page 1 of a print version.
• there are some css settings that determine the size of that logo and the padding around it
• there is another css setting that determines the padding between that logo and the page title that shows up in the header on each page of the print version.

Oh look! This and many other formatting features have been fixed including rendering of the section headers and margin harmonization. Thanks, mysterious layout fixer :)

@Dzahn @Quiddity possibly of interest. Is a variation of this planned under different context?

Thanks for remembering this @Jdlrobson-WMF should be a quick fix and could be flagged as a good first task?

Existing article style covers a lot of this, and should be referenced as a starting point. We have a range of options that tend to localize the suggestion to the place where it can be applied, and only show you a few at a time.

I can't imagine a scenario where we would want to delete the current Wikinews projects. What would that accomplish?

In T415237#11737580, @Tkarcher wrote:

@Novem_Linguae : It would probably be technically possible, but at the same time quite risky: I know users who consider Etherpads with a long enough random title as completely "private" (though they never really are), so your approach could accidentally unveil *a lot* of personal data, passwords and other unexpected content.

Comment: You are getting padding on top of the (tight) line height in the menu, so it should be clear that a long menu item across two lines is still one item. I think the impact on accessibility (for a single line) is low but the impact on usability (the number of menu options that fit on a single screen) is high, especially for smaller screens, which is also an accessibility issue.

Bumping this, as I had to show someone today that there were in fact 79 languages and not the 8 that showed up in the dropdown

This seems important for an unconference session and perhaps also a wiki map of views, tools, and approaches. (for those not there)

Thanks @Snaevar. Updated versions of those lists (maybe just the gadgets list with usage stats) would address the use I had in mind.

Is it possible to generate a list of requested domains that are being blocked by the CSP, in some browsable way?

In T415237#11706951, @Tkarcher wrote:

If keeping a read-only version at least for some time is not an option, can you *please* at least put a warning on the front page (https://etherpad.wikimedia.org/) and ideally also in the header of the editor page that everything will be deleted in 6 weeks? I see people still creating new pads and updating existing pads which were already backed-up

@Tkarcher and @Pppery and @bd thank you! for your work on this.

In T415237#11663510, @Ladsgroup wrote:

(FWIW, everything that has been linked on any wiki or anywhere else that could have been found is already backed up in a separate toolforge tool. See T415237#11617021 (Thanks‌ @Tkarcher)

Posted in another thread; @Miriam suggested cross-posting here!

Maybe intelligent automation can help with point 1 :) I'll take a look at the other brief.

A definite improvement. Compare the UX now for searching through all gadgets in user preferences.

@Alaexis and I chatted off-phab, I would love to see more tools trying this out. (An advantage of these fully open tools is that anyone can run them locally or on their own cloud, even though chat.pai.co offers a hosted service as a courtesy)

NB as hinted above in T385749, Integraality now has a QLever endpoint, for which data growth is not projected to become a problem in the near future.

Just noticing this, hurrah :) Love to see it.

QLever performed the best in independent benchmarks of some leading candidates; you're right that MDB didn't have full SPARQL support but all of the considered dbs had made significant progress towards it. Here's an updated report which tested how WD would perform at 2x its current size, (including bibliodata).

So nice to see this in progress. I was reminded recently that #newpalmyra produced great 3D models of which we only have STLs on Commons.
They are available via Sketchfab but that's increasingly hard to use since its acquisition by Epic. So lots of freely licensed sketchfab models need a properly free-knowledge repository.

On clickthrough licenses and regular deletion: this is a good moment to intentionally change the social norms that rely on Etherpad, or choose an alternative.
We shouldn't make a change for purely technical reasons and update the norms as an afterthought. Questions of "risk budget" affect users of the system just as much as they do the hosts of the resulting pads.

There was recently a wide-ranging brainstorming session in Frankfurt about what Wikimedia should become.
One thing it should probably not become is "unreliable at storing the knowledge outputs of its communities".

Saving this knowledge is a fine challenge. Interesting that something that would be a no-brainer for the Internet Archive seems so hard for us.

@leila excuse me if there's a better place to discuss reader surveys, I wasn't able to find a specific space for that.

I. Could this become an annual survey? Having some process that runs every year would be helpful even if the annual core is simpler / fewer updates / lighter overhead and new Qs are only added every 2y

And we are making Apertus and Olmo-Think and other fully public models available via API for free (though with rate limits to prevent saturation) vua the Public AI Inference Facility https://chat.publicai.co

!! Thank you.

<3 nice solution.

@OpenDEM @GabrielLucas the demo wiki is very cool, thanks for sharing. And that's a brilliant open letter, full support! More stats about current usage + interest in specific new 3D changes would be helpful for some readers: i.e. how many educational / cultural heritage models and source collections exist that need a better home?

I see three separate issues here:

@Jdlrobson-WMF just a note, this still happens in Vector 2022. (and I've continued to run across it infrequently, mainly in Chrome / MacOS 12 up to OS 15)

Qlever is approaching being able to support real-time updates, and is being more widely used by various wikidata-related initiatives as well as those involving larger datasets. This would be excellent.

@Ottomata thank you kindly, I could just have looked at Pageviews. I wasn't sure how to refine that, presumably it's almost all logged-out but includes some spiders?
Some special pages / views like DownloadAsPdf don't exist in that dataset.

Renamed because I couldn't find this. The reported css issues are still present.

An update re: moving off of Blazegraph: one early 2025 benchmark of different backends suggests QLever and possibly Millennium are solid candidates for replacing Blazegraph. Which should stabilize WDQS more reliably than the temporary delay-tactic of splitting the graph.

Thanks @Jdlrobson would love to see a quick analytics eval (or just an explanation of how I could find that data myself)

Please, please. In addition to stijn's point about this being widely desired across a number of different community requests, I highlight this previous comment:

Removing this logic would actually throw away a lot of code debt.

Thanks Gehel for making this an explicit issue.

Noting here that @Hannah_Bast further updated the page above, and QLever now has basic support for SPARQL 1.1 Update -- love to see the velocity of improvement!

@SD0001 fantastic, I experimented a bit.

Fantastic. Thanks @Samwalton9-WMF !!

Noting here that QLever now has basic SPARQL 1.1 support. Hurrah :)

just a warm note to @OpenDEM @PantheraLeo1359531 and @GabrielLucas : thanks for the progress and keeping this thread live

<hopeful>

Aha, thanks @Legoktm. Perhaps we could

a) get stats on non-monitoring usage [to confirm that this is low], and 
b) turn off auth for a month to see how usage changes?

In T330525#10142939, @Hannah_Bast wrote:

@Sj Getting the updates in batches would be perfectly fine. But how do you want to verify that it works without having a reference endpoint to compare to?

@Hannah_Bast: Appreciate the update, great to see it!

Given https://github.com/ad-freiburg/qlever/wiki/QLever-support-for-SPARQL-1.1-Update and conversations around that in various foums, 'showstopper' seems too strong.

@So9q per your comment on Wikidata:WDQS - is this worth reopening as an option? Whether hosted by WMF or elsewhere, but linked to from WDQS and pointed to when getting a timeout?

Thanks @Krinkle. The most significant updates this year seem to be the dramatic speedups observed by @Pfps and @Hannah_Bast on this page
https://www.wikidata.org/wiki/Wikidata:SPARQL_query_service/WDQS_backend_update/WDQS_backend_alternatives and the more extensive benchmarking here https://github.com/ad-freiburg/qlever/wiki/QLever-performance-evaluation-and-comparison-to-other-SPARQL-engines

I see, thanks Chris.

Since people implementing the fully working OWD embed were asked to pause it, could someone please add scoping when and how to support that to the Charts lineup?

Yes, by 'ideal environment' I meant one that lets an informed user do what they want, which includes things currently prohibited by nested (possibly over-specified) privacy + security policies that protect me from myself :)

Thanks Peter, I commented on T368980; the description is fine.

By the time I paste a URL into an editing interface as a citation, I have generally already visited that URL in the same browser to read it.
So when I call on citoid to extract and format a citation, I would ideally not need to download the source again.

Aside: the link to DB2Rest should be db2rest.com (github)

Why are the details about the problems sites report so vague? 😅
Did T367452 resolve any of these sites problems? Every visit being a double-download might have been enough of a spike for some of them to block.

That looks great, thanks @Quiddity. Deserves more visibility than just your user-css :)
Perhaps there's also a css-only solution to T328847 ?

Still the most-tokenized open issue :)

I see, alas! As with T364545 and T334941 I find V22 less polished at lower-width windows on desktop; I regularly use two-up windows on a laptop, so I run into them. Do you get feedback / polled data from readers with smaller windows?

Ran into this in compiling a newsletter this week. :) Only realized on finding this ticket that someone has had such good + quantified success. https://phabricator.wikimedia.org/T18691#6019615

Thanks gwern. I can see a single classifier working for our purposes, if it supports an 'uncertain' middle ground. just need to tweak where the threshold is drawn.

A family of solutions here, or even one flexible one, would be very impactful.

+1 to this idea. If there were a canonical image associated with the audio file, as there is in most catalogs, that could be the default poster to display which would be overridden by this setting.

Has there been any recent progress on this? T348269 by one of the preeminent reusers of SDC, in both scale and impact, seems relevant; but that remains untriaged and this one is High but unassigned.

you might try a mixture-of-classifiers, which all have an 'uncertain' range between yes and no:
0 - is precise color the point?
1 - does inversion look good in dark context?
2 - is there a color key in the text?
3 - other (non-standard inversion?)

Many technical problems are also social problems, this seems like a definite antifeature + I hope it is ok to keep the ticket open while addressing the social component.
@valerio.bozzolan I also do not recall any global discussion about how range blocks should affect various trusted users, and it is indeed a regular source of difficulty.*

@Jdlrobson @Dwain_Zwerg renicing this -- what is the proximal obstacle to updating the default?

@Gwern wonderful application, and agreed that this should be by URL and not by chain-of-reasoning.

One reason is that citations are a large corpus with a fairly narrow range of schemas and uses, so it could conceivably be implemented with optimizations that can't be applied across the board. But I do think that after a split into a separate citations wikibase, we should then formally evaluate the split and consider whether re-merging after a migration to a new graph db makes sense [I would hope it does -- 1B entities is not too much for a single db in other communities and contexts]

@Michael forgive my confusion, isn't part of this being able to use one Repository with multiple Clients, or to enable WikibaseClient to draw from both a local Repository and the global Wikidata repository?

aliu: absolutely right. Thanks for continuing to engage on this, and your generally constructive approach :)

For en:wp, this seems to differ from WP:Statistics by a factor of two. (looks like the latter shows ~500 articles/day)

It seems Naypta is not around. Someone else has tentatively offered to take over bot maintenance, and could use guidance, see this wiki discussion.

it *is* important that the solution be secure and performant.