[00:09:13] I wonder if all the individual requests were sent simultaneously when the original bulk page request was made? If so then refreshes just uncover that almost all of those simultaneous requests failed. (re @u99of9: Actually, the retries perform badly even if they are pressed individually.) [00:22:19] " This powers the script that refreshes each allowed article in each language every Monday, to keep articles up-to-date. " I presume that improvements and refreshes during editing will also update the article content directly? (re @vrandecic: Abstract Wikipedia Newsletter #255 [00:22:20] * Integration of Abstract Wikipedia on Test wiki [00:22:21] * The year ahead of us [00:22:23] * Recent Changes in ...) [00:30:05] " To maintain the demonstration and encourage people to write better articles about their own cities, it might be helpful to keep the Paris abstract article very short and simple for now. " I won't touch it. But I don't understand your objective here. Is it so that you can meaningfully improve it during the Wikimania workshop? Or so that all content works in all [00:30:05] unlocked language [00:30:05] s? Or so that editors writing about their own city will see a one sentence template? (re @vrandecic: Abstract Wikipedia Newsletter #255 [00:30:06] * Integration of Abstract Wikipedia on Test wiki [00:30:08] * The year ahead of us [00:30:09] * Recent Changes in ...) [01:06:37] They don't currently seem to: our AW article for Paris has a picture, but on TestWiki it's not there. Will we have to wait a week to fix vandalism?? (re @u99of9: " This powers the script that refreshes each allowed article in each language every Monday, to keep articles up-to-date. " I pre...) [01:10:42] These two are quite different. I suggest that the software should check which one is the case, and provide different messages in each case. : https://tools-static.wmflabs.org/bridgebot/71a690ad/file_81451.jpg [05:35:15] T430898 (re @u99of9: Actually, the retries perform badly even if they are pressed individually.) [05:36:58] Do we have something to mark the parameters as required and optional, by the way? [05:40:05] No, they are all required. A workaround is to have functions without those parameters as alternatives. (re @Winston_Sung: Do we have something to mark the parameters as required and optional, by the way?) [05:58:45] T430899 (re @u99of9: These two are quite different. I suggest that the software should check which one is the case, and provide different messages in...) [06:55:22] Then I guess it would be some kind of confusing as the "try function" button is not disabled and editing compose implementations doesn't warn about null/undefined inputs, or we should mark optionals/nullables? (re @vrandecic: No, they are all required. A workaround is to have functions without those parameters as alternatives.) [08:00:19] those are indeed good points! I mean, sometimes we can't tell if there's a value or not (e.g. for strings), but sometimes it is clear. Would you mind filing a bug? (re @Winston_Sung: Then I guess it would be some kind of confusing as the "try function" button is not disabled and editing compose implementations...) [08:42:12] Hi all! Our next *Volunteers' Corner* will be held Monday, July 6 at 17:30 UTC. (https://zonestamp.toolforge.org/1783359000) [08:42:12] If you have questions or ideas to discuss, or you want to get in touch with the dev team, please join us! The link to the meeting is https://meet.google.com/xuy-njxh-rkw [08:42:14] Hope to see you there! [08:42:16] (re @Sannita: Hi all! Our next Volunteers' Corner will be held Monday, July 6 at 17:30 UTC. [08:42:17] If you have questions or ideas to discuss, or you...) [09:43:00] I think that is a fair point if we long-term are thinking that they should be added one by one. But I imagine that they will be added in the many many thousands at a time. And that could perhaps be carried out after a community discussion by someone with extended rights. (re @Feeglgeef: I don't think limiting it to sysops is a good decision long-term) [09:46:17] I think the community discussion should be about whether to enable it and what the import criteria should be. But thousands at a time implies you can evaluate those criteria automatically. I expect a key criteria will often be something like "reads well in our language", which I think humans are best placed to check. (re @Jan_ainali: I think that is a fair point if we [09:46:17] long-term a [09:46:17] re thinking that they should be added one by one. But I imagine that they will be ...) [09:54:10] Perhaps I am thinking about this in the wrong way, but I imagine something like this: [09:54:11] 1. We create some great functions so that we can create a decent abstract article for a set of concepts, for example, soccer player bios. [09:54:12] 2. On Abstract Wikipedia we bot copy this to all we through Wikidata can identify as soccer players, perhaps tens of thousands. [09:54:14] 3. On a local Wikipedia they decide they want to use all this and agrees to import them all. [09:54:16] 4. Import either fails or content is not rendered if for a certain soccer player it doesn't render as expected in that language. (re @u99of9: I think the community discussion should be about whether to enable it and what the import criteria should be. But thousands at a...) [10:11:32] I suppose my expectations are in the nearer term. Indeed AW bots may change my thinking. Taxa in the spirit of lsjbot perhaps? (re @Jan_ainali: Perhaps I am thinking about this in the wrong way, but I imagine something like this: [10:11:33] 1. We create some great functions so that ...) [10:46:51] Yes, something like that. (Early on I thought that we would create "template articles" for this, but the current implementation doesn't have this in the roadmap as I understood it recently and we are left to do bot work instead if we want an approach like this.) (re @u99of9: I suppose my expectations are in the nearer term. Indeed AW bots may change my thinking. [10:46:51] Taxa in the spiri [10:46:52] t of lsjbot perhaps?) [10:51:06] On yesterday's topic. I've tried this idea on some Natural language function lists. For example, Swedish is here: [[Wikifunctions:Catalogue/Natural_language_operations/Swedish#Configurations]] (sorry to write the blurb in English, feel free to replace it). I imagine annotating the missing ones if the default is satisfactory. (re @u99of9: I got something working. If [10:51:06] you visit [[Us [10:51:06] er:99of9/Sandbox3]] you should see whether each config is configured (in your language...) [10:56:35] The blurb in English is okay. The short list in the top is actually showing the name of the function, would that be possible for the list of configurations too? (re @u99of9: On yesterday's topic. I've tried this idea on some Natural language function lists. For example, Swedish is here: [[Wikifunction...) [10:58:09] I don't think so. Unless an embedded function can executable wikitext which calls the {{Z|}} template. (re @Jan_ainali: The blurb in English is okay. The short list in the top is actually showing the name of the function, would that be possible for...) [11:02:12] Oh, I'm wrong, there's an easier way... give me a minute (re @u99of9: I don't think so. Unless an embedded function can produce executable wikitext which calls the {{Z|}} template.) [11:05:06] Done, see if you like it now. (re @Jan_ainali: The blurb in English is okay. The short list in the top is actually showing the name of the function, would that be possible for...) [11:05:31] Thanks! Very helpful!