[00:32:14] this applies only to extensions destined for the main prod cluster right? (re @wmtelegram_bot: yes, ongoing maintainance of extensions has been a huge problem recently, hence the not-new but not well documen...) [00:32:52] oh beta is wayyy worse [00:33:18] but yes no approval is required to write an extension, just to put it in WMF production [00:34:59] I meant vs. e.g. wikia, miraheze, single wikis (not farms), etc. [00:35:14] would love to hear beta rant somewhen :) [00:36:06] https://www.whyisbetabroken.com/ [00:36:28] omg [00:38:03] I wonder how much is maybe addressed by a future with https://wikitech.wikimedia.org/wiki/Catalyst [00:38:32] that's part of the solution, yes [07:37:11] ah yes, volunteers have to put up with storing code in hard-to-translate and hard-to-review wiki pages, copying and pasting it between hundreds of wikis and making edit requests on the wikis where they don't have enough permissions, while the people who get paid get to use things like git and translatewiki, don't have to worry about wikis constantly getting out of [07:37:11] sync, can edit [07:37:12] anything, *and* they get paid (re @albertoleoncio: I have the impression that creating extensions was something that was commonplace some time ago, but was "disincentivized" in fa...) [07:41:19] and gadgets are also a pain to test if you try to use any of the gadget features, since they're only available to gadgets, not user scripts [08:58:53] It's probably not a consolation, but it also happens to affiliates. WMSE even got tens of thousands in funding by the WMF to create an extension and still have trouble getting it reviewed. (re @doguabaris: So I think it's also okay to post this here: [08:58:54] Open Letter to the Wikimedia Foundation: [08:58:55] I am writing to announce my decision to ...) [09:02:42] What’s the background (re @doguabaris: Hello, I’d like to consult your opinion on a matter. As the author, can I request a halt and the deletion of the repository for ...) [09:03:06] Don’t. I like your extension very much. [09:14:49] https://phabricator.wikimedia.org/T355150 (re @cvictorovich: What’s the background) [09:17:15] Please (re @cvictorovich: Don’t. I like your extension very much.) [10:46:26] that sounds ridiculous and badly organised :( (re @Jan_ainali: It's probably not a consolation, but it also happens to affiliates. WMSE even got tens of thousands in funding by the WMF to cre...) [10:48:39] And not a chance error happening once, as they have had renewed funding at least five times. (re @Nikki: that sounds ridiculous and badly organised :() [10:51:23] I feel you, even keeping an extension maintained is almost impossible. (re @doguabaris: So I think it's also okay to post this here: [10:51:24] Open Letter to the Wikimedia Foundation: [10:51:25] I am writing to announce my decision to ...) [10:51:27] so bizarre 🤨 [10:53:27] But i would like to point out that there are reasons. I notice several ppl saying “get paid”, but wmf doesnt even have the manpower to maintain their own stuff. Even internal trams share these frustrations I have noticed [10:56:09] trams? (re @djhartman: But i would like to point out that there are reasons. And I notice several ppl saying “get paid”, but wmf doesnt even have the m...) [10:56:21] "teams", I assume [11:22:55] I mentioned getting paid because people are generally more willing to do or put up with things they don't like if they're getting paid for it [11:22:55] I can't imagine many paid developers would stick around if they were expected to do mediawiki development by writing code on wiki pages, do code reviews by posting edit requests on talk pages, create branches by copying and pasting the code into a new page, use code in another project by copying and pasting all the files, etc, even if they were still being paid for [11:22:55] it, because wo [11:22:57] rking like that really sucks, yet that's what people who work on modules/gadgets are essentially expected to do for free (re @djhartman: But i would like to point out that there are reasons. And I notice several ppl saying “get paid”, but wmf doesnt even have the m...) [13:59:19] The venue for next year's hackathon, the Renaissance Istanbul Polat Bosphorus Hotel, already has a Wikidata entry (Q111400225). However, it currently lacks images. This presents us with an excellent opportunity to improve the item by adding photos during our visit in May. https://www.wikidata.org/wiki/Q111400225 [14:12:33] We could crop https://commons.wikimedia.org/wiki/File:BarbarosBoulevard.JPG 😉 [14:13:59] That’s not the hotel. That’s the other one (re @EdwardBetts: The venue for next year's hackathon, the Renaissance Istanbul Polat Bosphorus Hotel, already has a Wikidata entry (Q111400225). ...) [14:16:33] There’s two similarly named hotels. One with Bosphorous in it, one without [14:18:17] It's like Hotel Diament all over again [14:20:17] https://tools-static.wmflabs.org/bridgebot/caf99342/file_66284.jpg [14:20:18] https://tools-static.wmflabs.org/bridgebot/dd4145d2/file_66285.jpg [14:22:09] https://www.wikidata.org/wiki/Q111857339 [14:32:35] @sjoerddebruin Seems we had the same thought in terms of "different from" editing :D [15:15:51] It's funny how a ~7kb article is now a ~40kb [15:23:05] Yeah, I added a _few_ statements [15:24:54] activity policy in this place (P5023) [15:24:55] You learn something new every day (re @sjoerddebruin: Yeah, I added a few statements) [16:22:21] Thanks for pointing out that I'd got the wrong hotel! (re @tehreedy: That’s not the hotel. That’s the other one) [18:03:43] Doing hundreds of #wikimedia-site-requests on Phabricator to get an extension installed on hundreds of wiki is also a pretty laborious workflow. Any situation where something needs to be deployed to hundreds of wikis is going to take a lot of work, I think. (re @Nikki: ah yes, volunteers have to put up with storing code in hard-to-translate and hard-to-review [18:03:43] wiki pages, copying [18:03:43] and pasting it b...) [18:04:23] Don't think extensions like these will be forced on hunderds of wiki's though, probably opt-in [18:27:14] Hi, I get an error when I make more than one request quickly in php, what is the best solution for this problem? [18:27:15] GET https://stream.wikimedia.org/v2/stream/page-create,page-move` resulted in a `429 Too Many Requests [18:29:00] make fewer requests… [18:29:07] and also make sure you’re setting a user agent, see https://foundation.wikimedia.org/wiki/Policy:User-Agent_policy#PHP [18:29:13] (generic user agents sometimes have lower limits) [18:29:47] but the general recommendation is to make requests in series (one after the other) https://www.mediawiki.org/wiki/API:Etiquette#Request_limit [18:33:07] How long do I have to wait to make a request when I receive a 429 error? [18:35:43] check if there’s a Retry-After header in the response [18:35:46] otherwise I’m not sure [19:05:30] Oh @lucaswerkmeister are you coming to the hackathon?? :3 [19:50:20] maybe :3 [20:11:03] It's another nice long train journey. [20:11:36] I was looking into that, it looks pretty tricky to reach without planes from germany [20:11:42] google says there are some bus options [20:41:03] Shipping container? [20:41:59] take a slight detour to the south, block up the suez canal… [21:25:19] Oh no, there's night trains to Istanbul in the summer but probably not in May: https://www.seat61.com/Turkey.htm#london-to-istanbul-via-bucharest [21:53:26] On another page, it sounds like there’s a year-round Sofia to Istanbul night train https://www.seat61.com/trains-and-routes/bucharest-and-sofia-to-istanbul-by-train.htm (re @tuukkahastrup: Oh no, there's night trains to Istanbul in the summer but probably not in May: https://www.seat61.com/Turkey.htm#london-to-istan...) [22:08:46] Yes - then the problem is how to get to Sofia. (Possibly by bus from Budapest.) [22:14:52] If someone was to rent a bus and organize a Wikiexpedition from around Hungary to Istanbul, there would be a lot to experience on the way