Wikipedia:Village pump (proposals)/Archive 166

The Wikimedia CentralNotice is disruptive, heavily overused, and has a negative impact on usability. The CentralNotice process on Meta routinely approves of campaigns that receive no support, in contravention of a prior (Meta-based) RfC establishing that notices must have consensus in advance. The guidelines restricting the use of the CentralNotice are ignored, as are the requirements for notifying local communities of CentralNotice proposals. With that in mind...

I propose that the process for approving English Wikipedia sitenotices, including those using the CentralNotice, be relocated to the English Wikipedia. Sitewide banners would require community consensus in advance of being run. Exceptions to this process would be made for fundraising, regular Wikimedia-wide elections (stewards, trustees), and technical messages. No other sitenotices not approved by local processes should run on this site.

Implementation: If there is consensus for this proposal, I fully expect the CentralNotice administrators on Meta to be willing to comply with editing CentralNotice campaigns to exclude this wiki when a banner has not been locally approved. On the off-chance that consensus is not followed, local interface administrators are able to hide unapproved banners via Mediawiki:Common.css as necessary.

--Yair rand (talk) 17:54, 19 January 2020 (UTC)

@Yair rand: there are seldom "English Wikipedia" only CN's. To put this in better perspective can you give specific examples (link to the actual banner unless it was deleted), with dates, of the last three Central Notices that appeared for our readers/editors that would be impacted by this proposal? — xaosflux ^Talk 19:15, 19 January 2020 (UTC)

@Xaosflux: To clarify: I'm not talking specifically about English Wikipedia-only notices, I'm talking about all CentralNotices that would run on the English Wikipedia, regardless of their other targets. While the process here wouldn't influence whether any CN ran on any other project, it would decide whether it would run here. --Yair rand (talk) 20:47, 19 January 2020 (UTC)

@Yair rand: oh sure, I meant that part as just a comment - same question though - can you list the last 3 notices, with dates, that did appear here in any way, that would be impacted by your proposal? — xaosflux ^Talk 21:52, 19 January 2020 (UTC)

@Xaosflux: Impacted in the sense that they would have had to get approval on the English Wikipedia before running? I'm having difficulty reading the CentralNotice logs directly, but looking at the calendar entries and filtering out those that didn't run on ENWP... It looks like the three most recent were all (geographically) local events: There was a month-long photo contest in Spain ("Wiki Loves Folk", 11-01 to 11-30), a banner in Switzerland advertising a "Wikidata training and Hackathon" session (10-20 to 10-31), and a banner in the Czech Republic and Slovakia promoting a Wikiconference (10-14 to 10-29). Before those, we had a banner in Spain advertising a conference (10-01 to 10-17), and a banner across Northern and Western Europe from 08-26 to 09-08 bugging several hundred million people in order to try to get 17 volunteers to show up for a two-day hackathon. (Okay, that's five notices, but...) --Yair rand (talk) 06:58, 20 January 2020 (UTC)

@Yair rand: OK, I started looking (and yes the CN interface is confusing - I try to stay out of it as much as possible!). I'm not sure all of these actually ran, but it does appear that ones that have have been set by both staff and volunteers. Staff side, most of these are managed by User:Seddon (WMF). I'll go message Joseph to see if he can comment on this thread here. — xaosflux ^Talk 13:35, 20 January 2020 (UTC)

Hey all, I want to give a quick note. CentralNotice is a vital tool and it powers some of our biggest community led initiatives (Wiki Loves Monuments, Wikipedia Asian Month, Wiki Loves Earth). Originally it was mainly developed as a fundraising tool and it's currently the only real tool the movement has to engage with readers at scale. This is a role it has had placed upon it rather than one it was built for.

Prior to the current process being brought in, realistically there was no real process. You had to call in a favour from your best friend and there was limited transparency. It's a process that has evolved since I brought it in and I fully acknowledge it's by no means perfect. But what is being proposed, I don't think this is the long term solution. If every wiki did this it would be chaos and it would severely impact some of our biggest most successful programmes (Wiki Loves Monuments, Wikipedia Asian Month, Wiki Loves Earth). I think there are two things that need to happen:

• Process: I fully acknowledge it needs to be tightened up and the guidelines become more like rules and we need to get better at enforcing them. We need to be clearer about what types of projects get what support. What is overuse etc.
• Reduce reliance: We need to ensure that people need to use the less sledgehammery communication tools that already exist. Mailing lists, talk pages, social media, irc etc etc.
• Technical improvements #1: Improvements to CN have been limited and mainly focused on fundraising but this is slowing changing. As it's usage has grown, the tool has had new features built in to reduce the impact on end users experience such as impression limiting. We should look at prioritising functions that allow us to focus the outreach usage on newer users, or allowing targeting of specific topics rather than sitewide notices being used for topics with a narrow focus.
• Technical improvements #2: I really think we need a way for users to opt out, allow individual users to permanently opt out of some or almost all types of CN banner much like what you get with email preferences with a company.

There is room for improvement but CN definitely supports the creation of great content and we don't want to see that disrupted. Reduce the impact of CN, focusing its use, improving project leaders understanding of their responsibilities in using CentralNotice and get some improvements made to the system. I want to collect some data from across the movement on how people perceive CN and it's usage and what people want to see if it used for. If y'all willing to worth with me on this I think we can see some genuine improvements some in the short term and some in the long. Seddon (WMF) (talk) 06:19, 21 January 2020 (UTC)

Thank you for the update Seddon (WMF) - and I agree that it is time to discuss this issue more before thinking we are at any sort of impasse! Where do you think is the best place to coordinate this so that anyone who would like to join a working group for it may contribute? — xaosflux ^Talk 15:47, 21 January 2020 (UTC)

• I'm tempted by this, I probably wouldn't be so in favour of it if the actual rules were being followed, but we seem to be getting a number of site notices with seriously marginal relevance. Nosebagbear (talk) 09:39, 20 January 2020 (UTC)

To clarify for any future closer, my preferred method is my comment below, asking for an ability to opt out of non-administrative sitenotices. A serious firming up of the rules is 2nd choice, this would be my 3rd choice. Nosebagbear (talk)

@Yair rand: Why exempt fundraising notices? There were a lot of comments/complaints on the Help Desk this year about fundraising (more than I recall for past years). I think the community should at least have a say in how the messages are worded, and the messages should include the Wikipedia e-mail address where users can send comments/complaints about fundraising. (ie: fundraising complaints go to donate wikimedia.org. and not to the Help Desk) RudolfRed (talk) 02:23, 21 January 2020 (UTC)

@RudolfRed: just from the banner alone the donate at email address is provided twice in our FAQ, our problems donating page on four occasions ^{3 4 56}, twice in the donor privacy policy and a third via the WMF contact us page ¹⁰. That email address is also provided on the contact wikipedia page ¹¹. While I'm definitely not opposed to providing that email address in more locations, sticking it in a banner would likely place the donor services teams under immense pressure and reduce how effectively we can respond to queries so I don't think that's a solution.

If there are ways you think I can help support the help desk in responding, triaging and referring issues that come in via the banners I am definitely happy to work with you on that. We have an ongoing the relationship with affiliates who get similar queries, the OTRS teams who forward issues to the donor services team as well as to me directly as well as feedback that comes in via social media.Seddon (WMF) (talk) 05:40, 21 January 2020 (UTC)

As far as why exclude fundraising? This goes back to the WMF owns these servers, and manages the process to solicit and collect money from donors to keep them running and this is one of their main tools for doing so. We have already made it very easy for editors to opt-out of seeing these, but getting reader donations is still considered important. — xaosflux ^Talk 15:47, 21 January 2020 (UTC)

• I very much support this. Wiki Loves Monuments, Wikimania Stockholm Call for Submissions etc are irrelevant to practically all readers and they should not be displayed so prominently on every single article. The fundraising efforts were unusually aggressive this year (judging by the amount of internet jokes made about it). I would further suggest either: 1) Only display these banners on the Main Page, 2) display any necessary banners at the bottom of articles. – Thjarkur (talk) 11:42, 21 January 2020 (UTC)
• This is your occasional friendly reminder that if you don't want to see banners, then you can go to Special:Preferences#mw-prefsection-gadgets and choose "Suppress display of fundraiser banners" and/or "Suppress display of CentralNotices". (Please remember if you set these, so that you don't embarrass yourself by complaining that you didn't see the announcements you hid. It has happened a time or two before.) WhatamIdoing (talk) 19:13, 21 January 2020 (UTC)
• @Seddon (WMF) and WhatamIdoing: - as Seddon notes, it was an additional use-case placed on top of the adminstrative/fundraising side. Seddon, I'd be happy enough for it to just be tightened up if, like fundraising banners, I gained the ability to turn them down selectively. There are the occasional administrative ones I need to know about (site-wide elections, the Talk Page consultation), but I don't want the cost to be the need to see a large number of campaigns/projects that have relatively little relevance to me. Give them a tag that lets users opt-out, and I imagine the complaints will heavily dwindle. Nosebagbear (talk) 09:14, 23 January 2020 (UTC)

I would support Nosebagbear's suggestion of opting in just to seeing key administrative central notices. (Put your hands up, those of you at the back, if you can't hear me!) Nick Moyes (talk) 12:37, 23 January 2020 (UTC)

Tbh, I'm much more concerned about the endless spam sent out to the IPs. We have a pretty constant stream of notices. Over the next month, we're going to get a global notice for yet another photo contest (global, logged-in and IPs, all through February). The logged-in users are also going to see a month of Wikimania scholarship ads (02-10 to 03-09), a bunch of "Wikipedia day" banners throughout North America, a global notice promoting translating some human-rights related articles, a week-long notice from the Desktop Improvement team to show off their new prototype, and a lengthy notice for the steward elections. This is all on the heels of our extensive annual fundraiser which just finished. The normal appearance of the site for readers or editors should not be one with a banner on it. --Yair rand (talk) 18:29, 23 January 2020 (UTC)

• Support proposal, and advocate for documentation The status quo is that the Wikimedia Community is the arbiter of what is and is not acceptable for a CentralNotice. Wikimedia Community decision making process has overriding power, and this is how things always have been, and this is the social contract on which basis the Wikimedia Foundation entered the conversation, and this is the environment of expectation. As conversation proceeds it should go forth from that norm.

That said - if anyone wanted to purchase advertising with the reach of the CentralNotice from another website, no media organization (Facebook, Google, etc) would sell this kind of media reach for less than billions of dollars. Any time there is a billion dollars of value up for grabs there will be conflict and insane behavior. I advocate for financial investment in Wikimedia community organization now which would make any future decisions about the CentralNotice meaningful. If the Wikimedia Foundation neglects to invest in infrastructure which empowers the Wikimedia Community in this conversation, then that lack of investment profoundly undermines any Wikimedia Foundation legitimacy in the decision making process of what to do with this most astounding of Wikimedia community controlled resources. There might come a time when the CentralNotice can manifest career-capping, infinite amounts of revenue forever so encouraging a Wikimedia-community based ethical conversation about conflicts of interest would be wise. Blue Rasberry (talk) 19:43, 23 January 2020 (UTC)

• Support more banners, more often, with more Wikimedia community control I might be misunderstanding this proposal or what is proposed here. The Wikipedia community needs more banners with lower barriers to get them published and out. Wikimedia community groups, including Wikimedia chapters, WikiProjects, and other affiliate group models should be the authority in deciding when and how to use banners. Banners should generally come at the behest of organized community groups, because understanding when and how to use them requires long-term group planning with year over year documentation and lessons learned. I am in favor of empowering the English Wikipedia community to manage banners. The administration of that power should be mostly in the hands of community outreach groups which organize themselves, with less weight on unorganized ad hoc Wikipedia commentators who are not regular particpants in the broad discussion of banner policy and who are not stakeholders in the organization of Wikimedia outreach. I am a bit confused about this proposal because it raises lots of variables, like is it for or against banners, for or against meta, for or against random commentators in control, or for or against wiki community organization. I want more banners, less of the restrictions which happen from meta, less random commentary from people unaffiliated with banner campaigns, and more power to Wikimedia chapters and organizations which push for more banners. Blue Rasberry (talk) 18:39, 3 February 2020 (UTC)

  "The Wikipedia community needs more banners": Speak for yourself. I genuinely had to re-read that to be sure you were not being ironic. I am part of the Wikipedia community and one thing I absolutely do not need is more frikkin' banners rammed into my frikkin' eyes! That community "unaffiliated with banner campaigns," from which such despicable commentards as myself are drawn, is in fact the very user community which your banners are targeting! Diss us, ignore us, trample us at your peril. The advertising platform that anyone can edit is not what we are here to read. — Cheers, Steelpillow (Talk) 19:45, 3 February 2020 (UTC)

@Steelpillow: You have a strong reaction! Can you say something about how much advertising you experience on Wikipedia? I also fail to understand you. Here is what I would ask -

1. How many days a year do you visit Wikipedia? (I guess 200 for you, based on your edit history)
2. On how many of those days do remember seeing a banner ad? (I guess 30 for you, which would become ~3 if you have banners turned off in settings)
3. On days when you see a banner ad, how many ads did you see? (I guess 1 for you)

Are these numbers right for you? When you say "no more banners", is zero banners the right number for you, or do you have a non-zero acceptable amount per year? From my view, Wikipedia already has so few as compared to any other major website, so I wonder what the tolerable number of banners is for you. What are you experiencing that you feel so strongly about this?

Also, are you speaking for yourself as a logged-in user who can turn off banners in settings, or about what you wish for users without an account, who cannot turn off banners?

I feel that we should (1) negotiate a certain number of banners per year then (2) put some of those in Wikimedia Fundraising and (3) put the rest in control of the Wikimedia community for outreach campaigns. Is there something about any of this which seems strange to you? Thanks for any feedback. Blue Rasberry (talk) 04:24, 4 February 2020 (UTC)

@Bluerasberry: I am very surprised to hear that your guess would be 15% of days on Wikipedia involve seeing a banner. My own impression of how often they're up is somewhere between "most of the time" and "always". (Showing banners 15% of days sounds quite reasonable to me.) I think we need some data, if there's such divergence in their perceived frequency. Re your three points, they make a lot of sense to me (although I'd expand "outreach" to also include some other necessary notices). --Yair rand (talk) 04:40, 4 February 2020 (UTC)

@Yair rand: I agree, a good outcome of this conversation would be the WMF fulfilling a request to provide metrics of when banners run and who sees them. The banner is a valuable resource and it is worth measuring and planned allocation. If you advance this conversation, then along with requesting metrics, I would appreciate you always imagining that some entity will control the allocation of the banners to various causes. When there is a power which the Wikipedia community does not organize to claim, then by default that entity which controls the resource will be the WMF, who will give decision making power to staff. The alternative, which I hope you will support in future conversations, is some stewardship by the Wikimedia community probably led by Wikimedia chapters since banners are for mostly outreach and chapters are the major representatives for outreach. If you have another vision for this then call it out, because banner power is among the most valuable resources the movement has to allocate and I want to keep the precedent of Wikimedia community control of it. Blue Rasberry (talk) 12:19, 4 February 2020 (UTC)

@Bluerasberry: First, I think it quite offensive to anonymous readers that you should be thinking in terms of spamming them and making life easier for their logged-in colleagues such as your self and chums. Yes IP editors have to accept limitations but most every visitor who just comes to read the encyclopedia - which is after all what it is for - will not log in even if they have an account. This needs hammering out for everybody, before we give a dam' about the account holders' convenience. I visit pretty much every day, sometimes for extensive periods of research in IP mode. I log in only when editing. I have NoScript, with javascript disabled by default, and AdblockPlus with a miscellaneous collection of local blocks accumulated over the years. But some useful features require javascript so I often have it enabled here. Sites where unsolicited content nevertheless intrude create a negative "oh, they're that kind of self-focused soul, thinking not in terms of what the reader wants to see but what they want the reader to see." The absolute key thing is to understand banners as informative for the reader and not as a self-promotional tool per se. The last thing you want to do is to fatigue the user so they just go "not another feckin' ad banner" and never stop to read the message. I probably get more banners, things that slide in from the side and that kind of nonsense when I am in IP mode, but quantifying the difference is like counting how many dog turds I step in annually with my left or right foot, it's pointless. More personally, the most important thing for me is that there should be no animations of any kind - sliding, fading, scrolling, anything - in my peripheral vision. That is an accessibility issue, not as dangerous as flashing/strobes are to an epileptic, but equally unpleasant. My view is that a restrained use of banners (no big bright colourings either, maybe half a dozen short universal or UK-only campaigns a year) is OK, but only provided that there is adequate community governance in place to make sure that the look-at-my-dancing-monkey brigade are locked firmly out. That last is singularly what has not been happening, and we at en.wikipedia can lay it in place faster than WMF central can. — Cheers, Steelpillow (Talk) 09:54, 4 February 2020 (UTC)

@Steelpillow: Accessibility standards are great. Of course I agree that when there are banners, they should conform to some standard. The Wikipedia community should set that standard and model good behavior to the world.

I would agree with you to (1) have a negotiated number of banner days or campaigns and (2) negotiate the method of allocating them to project proposals. I think that currently we have no data on how often banners appear and how we give those banners to campaigns, and that you and I have different understandings of the basic facts. So far as I know, a typical viewer coming to Wikipedia most days only sees 6 campaigns a year, which is what you say you want and are not getting. Blue Rasberry (talk) 12:21, 4 February 2020 (UTC)

As I said, I don't count the campaigns or distinguish between one kind of distraction or another. I do think we understand each other better now. — Cheers, Steelpillow (Talk) 14:42, 4 February 2020 (UTC)

• Strong support, thank you Thjarkur for making a very valid point. Wikipedia should have an independent sitenotice process, where we can review Meta's controversial fundrasing techniques to align with our goals. >>BEANS X2t 18:50, 24 January 2020 (UTC)
• I would like to get more information from Seddon (WMF) with regards to it would severely impact some of our biggest most successful programmes. By what metric are they successful, and how much do the banners contribute to that, in concrete terms? (For example, number of edits made or number of sources added, but not simply number of people.) Also, are the extra banners known to attract new editors to Wikipedia, in these or other cases? I'll note that if there are good arguments for the value of these specific examples, the community is unlikely to object to continuing them.

As a separate point, I will generally agree with any attempt to add an extra layer of review to restrict the promotion of surveys with vague, meaningless, or misleading questions. I would also emphasize that while the proposed exceptions are fine, this does not make them exceptions to consensus; rather they are more like "pre-approvals" that can always be revoked (individually or entirely) in the future if necessary. (Similar to the current situation in which everything is pre-approved. Of course, that's not to say that I necessarily expect such a revocation to happen.) Sunrise (talk) 20:32, 24 January 2020 (UTC)

• Strong support the community needing to opt in to the fundraiser banners. It should require at least a village pump RfC (advertised by
  WP:CENT) to include a fundraising banner, with the wording and linking also up for debate. I imagine if we change the banner then the WMF (particularly the legal team) will have to give final approval, but if we can't get agreement then the banner shouldn't go up. I understand the WMF will very much not want us to do this and it would be a tough fight to win, but their current fundraising strategy is intrusive, unhelpful to readers and biggest of all flat-out misleading—they already have more than enough money to stay afloat. If a person is donating then they need to be aware of where the money is going, and where it is not going (e.g. to run our servers). I am fairly sure that at least 90% of (one-off) donors do not understand where most donation money is spent.

  I understand fundraisers were excluded from the original RfC question, and I support the original proposal. My main concern is banners shown to non-logged-in readers. Spamming our editors is annoying but not a huge deal; what the (English-speaking) public see is a hugely important thing that en.wiki should be the deciders of. — Bilorv (talk
  ) 00:18, 27 January 2020 (UTC)
• I wouldn't want a process where volunteer editors reviewed and approved individual site notices. I would certainly approve of restrictions that manage how emphatic, how distracting, and how shrill, these messages can be. We want the WMF to be able to raise money and attract attention to good causes. However I believe that the site notices recently have become a little too effective at diverting visitors' attention to the WMF's priorities. We do have a mission to enable our visitors to inform and educate themselves in a self-directed way, and we're not achieving that while they're looking at ads.—S Marshall T/C 10:40, 30 January 2020 (UTC)
• Wondering why so many of the above volunteers are not participating at meta making the world a better place by changing the process there instead of micromanaging it from here, and wondering if they think they will become active here in actually managing a changed process, for longer than a 3 month attention span until they jump to the next 'crisis'. —TheDJ (talk • contribs) 11:11, 30 January 2020 (UTC)

  @TheDJ: We did that. They didn't listen. Opposition to CentralNotice proposals at Meta doesn't do anything, and CN admins have said they're completely willing to put up CNs without consensus, contrary to the last global RfC on the topic, which concluded in favor of requiring advance consensus for banners. --Yair rand (talk) 01:13, 31 January 2020 (UTC)

• Wholeheartedly support. If someone wants a policy rationale beyond
  WP:NOT#SOCIAL. PS: 'there are seldom "English Wikipedia"-only CN's' – Yep. So we should end up with a lot fewer annoying CNs irrelevant to 95% of editors and 99.9% of readers. That's kinda the whole point, innit?  — SMcCandlish ☏ ¢
   😼  22:06, 2 February 2020 (UTC)
• Support for now. I wholeheartedly agree that the central process is currently not fit for purpose. Too much spam, too much thoughtless ignoring of accessibility rules such as animations off by default. Central need to build a stronger and more democratic community before we can trust them to do it properly. Meantime, this is our baby and it is our responsibility to vet what they put forward before it goes live. Ultimately there is a case for central broadcasts to go over our heads, but that must be subject to proper control. — Cheers, Steelpillow (Talk) 12:26, 3 February 2020 (UTC)
• Support per proposal. ~ ToBeFree (talk) 16:37, 3 February 2020 (UTC)
• Oppose. This proposal does not make sense, it conflates sitenotice with centralnotice, and completely ignores the fact that most centralnotices are geographically-based community efforts, not global campaigns. We already have a process and a !vote on meta, because this affects multiple projects, and we already have the rule of posting pointers to the meta page on the Village Pump. The proposal would effectively be to give a carte blanche to WMF banners and a veto to community-based banners, which is not what we want.--Pharos (talk) 23:36, 3 February 2020 (UTC)

  @Pharos: How would this give carte blanche to WMF banners? All WMF banners (outside particular exceptions listed) would require consensus as much as community-initiated proposals. --Yair rand (talk) 00:28, 4 February 2020 (UTC)

  Also, the !vote process on Meta is, unfortunately, usually ignored by the CN admins these days, in contravention of the last global RfC on the topic. The behaviour isn't exactly legitimate, but it's what we have to work with atm. While your own CN activities have appropriately followed the community notification guidelines, many others, particularly the WMF-sourced proposals, have ignored the rules entirely. I hope we can get to a process that works for everyone, but to get there, we need to have some community control over the process to start with. I know this will make things much less convenient for anyone trying to organize a notice across multiple projects, but I think it's necessary. --Yair rand (talk) 00:28, 4 February 2020 (UTC)

• Oppose - certainly in regards to fundraising. The WMF needs some flexibility in fundraising that shouldn't be taken away by a few ideologues. I've never really been bothered by any banners on Wikipedia. They only take up a small amount of space on the page, and you can scroll down to get them out of sight. You can dismiss them individually. You can opt-out. And they don't seem to appear very often. Can we get some real data? e.g.:
• How many different banners appear each year on all projects?
  • If they can be limited geographically, how many appear in London, Chicago, Sydney, and New Delhi each year?

This just doesn't look like a problem to me. Smallbones_(smalltalk) 01:14, 4 February 2020 (UTC)

@Smallbones: The proposal specifically exempts fundraising. Re data: Better than "how many different banners per year", I think more important would be, "At any given time, what are the odds there's a banner running?" --Yair rand (talk) 01:36, 4 February 2020 (UTC)

@Smallbones: Okay, I spent some time building a script to scrape the on-wiki logs (the CN-log API is useless here and omits most of the important data) to figure out how often banners are running. WIP, but it looks like on 220 of the 365 days in 2019, at least some people in the US saw a CentralNotice banner on the English Wikipedia. (At least 18 of those days, it was only a particular geographic region of the US.) I'll try to work on it a bit more to refine the data, and then generate some tables for other geographic areas and times. --Yair rand (talk) 22:33, 13 February 2020 (UTC)

• Oppose per Pharos. Gamaliel (talk) 02:33, 4 February 2020 (UTC)
• Oppose per Pharos and Smallbones. The ability to have a wide-ranging system for displaying banners is a critical tool for movement-wide awareness and communication. Since Wikipedia's heyday of peak participation in 2007, we have had major breakdowns in almost all our communications channels. The mailing list Wikimedia-L is a fraction of what it used to be, with folks new to the movement not even knowing it exists. The new Wikimedia Space bulletin boards are not picking up a lot of traction. IRC is a ghost town. We now have a smattering of Facebook groups, Telegram chats, Discord channels, Slack groups, et al. that try to piece together some overlapping set of community. CN is the only tool we have left that can effectively communicate across cultures, geographies and communities. Besides: logged in users have a way to hide these types of banners. Personally they have never bothered me either, and in fact have alerted me to projects, drives, conferences and things that keep me in touch with the movement zeitgeist. Optimize, refine, adjust and reform the current system, but don't splinter it off. -- Fuzheado | Talk 17:19, 4 February 2020 (UTC)
• Oppose per Pharos, Smallbones, and Fuzheado. This is even more critical to maintain considering there is no real alternative. While Wikimedia Space attempted to be a hub for events, the WMF needs to figure out how to better support Space and communities who could use it. Jackiekoerner (talk) 20:26, 4 February 2020 (UTC)
• Oppose per Pharos, Smallbones, Fuzheado,Jackiekoerner. There should be ways for individual users to opt out. The communications of community projects depends on these tools that are part of movement resources similar to funds and volunteers. There should be a way around to make compromise or meeting half way around rather than taking it away entirely. Wikilover90 (talk) 23:48, 4 February 2020 (UTC)
• Idea one thing I'd really like to is a) opt out of specific types of banners, and b) when I opt out, that my account opts out on all devices, rather than have a cookie-based 'opting out'. Like for instance, Wiki Loves... whatever. Or photography-related stuff. Or GLAM stuff. I'm not really sure how easily the 'types' can be made, but I know annoyance would go significantly down if I didn't have to bypass those banners every time they popped up, and if I didn't have to bypass them on 3-4 devices every time one popped up.
  b
  } 23:58, 4 February 2020 (UTC)

  @Headbomb: banners do support banner categories, in practice only "campaign" and "fundraising" are really used - we could propose some very broad ones at meta-wiki, it shouldn't be hard to adopt - but if there are many expect lots of errors. — xaosflux ^Talk 00:17, 5 February 2020 (UTC)

• Strong Oppose per Pharos, Blue Rasberry, and Fuzheado. It is pretty obvious that improvements to centralnotice and to its governing processes would be great, but they should be in the direction of more banners, more often, and more engaging. The centralnotice banners are one of the most powerful tools we have to reach communities that have been excluded as contributors to the Wikimedia Movement. It makes no sense in my mind to focus on spending millions of dollars and thousands of volunteer hours to get these communities engaged with the Wikimedia Movement and at the same time tighten the use of an asset that is the cheapest and most effective at actually getting these communities to contribute and engage. Chico Venancio (talk) 12:56, 5 February 2020 (UTC)

  We are spamming billions of people. The readers don't care about the the WMF CEO's latest speech, or some "human interest" blog post, or a hackathon three hundred miles away. --Yair rand (talk) 19:09, 5 February 2020 (UTC)

  @Yair rand: "billions"? Did you know that not all banners are shown to "readers"? For example a recent banner, "WikiForHumanRights_Declaration" is set to only show to "logged in users" and the default banner directions for logged out users is "Anonymous users (remember to uncheck unless necessary)". — xaosflux ^Talk 19:14, 5 February 2020 (UTC)

  @Xaosflux: Very many banners are shown to readers. In my experience, people are much less careful about banners that are shown only to readers and not at all to logged-in users (such as the blog post spam I mentioned above). As I mentioned earlier, I'm much more concerned about the endless spam sent out to the IPs. (IIRC, the other two banners I was referring to were also shown to IPs.) --Yair rand (talk) 19:24, 5 February 2020 (UTC)

  @Yair rand: would you please cite one of the specific banners (from meta:Special:CentralNoticeBanners) you are referring to? — xaosflux ^Talk 19:42, 5 February 2020 (UTC)

  @Xaosflux: Blog1819_Punjabi_Wikisource_1 I already linked above. WikiTechstorm2019 was shown to everyone (including IPs) across most of Europe, with the aim of recruiting ~17 people to participate in a hackathon. (No idea whether it succeeded in that.) I haven't found a link to the banner which advertised the Executive Director's speech at Wikimania, sorry. --Yair rand (talk) 20:00, 5 February 2020 (UTC)

  @Yair rand: those 2 seem OK - in looking at them, WikiTechstorm2019 appears to have only been shown to logged in users, however most unusually Blog1819_Punjabi_Wikisource_1 appears to have been shown to ONLY anonymous users and was configured by Ed Erhart (WMF), seems to have been the only banner they ever made - wonder if it was set up in error? — xaosflux ^Talk 20:11, 5 February 2020 (UTC)

  @Xaosflux: WikiTechstorm2019 was shown to IPs. It only looks like it wasn't because the settings were changed just before the end of the campaign (at 31 August 2019 20:17, according to the logs). Blog1819_Punjabi_Wikisource_1 probably wasn't a mistake, given that it was listed that way in the original proposal as well. --Yair rand (talk) 20:26, 5 February 2020 (UTC)

  @Yair rand: Beyond the obvious lack of awareness of how CentralNotice campaigns work, I find your description of a blog post about the exact kind of work we should be prioritizing as an <<"human interest" blog post>> quite telling, and a little offensive. What I am reading from your proposal and your responses is that you do not care the Wikimedia Movement has not reached several areas of the globe and that a lot of knowledge is not included in our projects as long as your screen is free of these pesky banners. Chico Venancio (talk) 19:31, 5 February 2020 (UTC)

  @Chicocvenancio: Um. I'm quite aware of how CentralNotice campaigns work. I've probably been the single most active volunteer in the process over the past few years, despite not being a CN-admin. And my screen can be permanently free from all banners at the click of a button, which most people can't do. I'm going to politely refrain from commenting on the rest of your post. --Yair rand (talk) 19:39, 5 February 2020 (UTC)

  Yair rand Than why are you citing a campaign aimed at Indian readers, made by WMF comms, for a project everyone should be in complete agreement as an example of SPAM being sent to "billions"? Chico Venancio (talk) 20:00, 5 February 2020 (UTC)

  @Chicocvenancio: Because readers actually come to Wikipedia to read the articles. These notices are unsolicited messages sent out to very large numbers of people. That's literally the definition of spam. (I am aware that the India-targeted campaign did not, alone, reach billions of people. I was referring to the larger pattern of CentralNotice misuse.) --Yair rand (talk) 05:36, 10 February 2020 (UTC)

  @Yair rand: it is advertisement for our own movement, I don't see how it would be SPAM. Why would it be ok for us to advertise for fundraising and not for getting more volunteers? How is WMF funds more important than volunteer time and energy? Chico Venancio (talk) 14:12, 10 February 2020 (UTC)

• I agree to open to more campaigns but I would invite to consider that there is the possibility to have an impression diet and other limitations and to consider that long campaigns or worldwide campaigns are not efficient. SO let's consider to limit massive and invasive campaigns to give more space to well targeted, more geotargeted campaigns. --Ilario (talk) 20:46, 5 February 2020 (UTC)
• Oppose Per what the others have said, I think it's a bad precedence to allow notice discussions on local wikis.. It would lead to a big waste of time and a ridiculous level of bureaucracy, when other wikis also make the same demands. It just doesn't make sense. But I do agree that the process on meta should be more transparent. But then again, the bulk of that responsibility rests on the communities, to increase their engagement on the notice request page.--Jamie Tubers (talk) 13:46, 7 February 2020 (UTC)
• Oppose. Agree with Jamie Tubers, the community should engage more on meta which is an open wiki. Also I dont agree with geotargeting : we do need more engagement on diversity topics and these campaigns are useful for these. Nattes à chat (talk) 08:32, 8 February 2020 (UTC)

  @Jamie Tubers and Nattes à chat: CentralNotice campaigns don't always even have a proposal page to engage on, and when there is, engagement is often ignored. There isn't anywhere to go. --Yair rand (talk) 05:33, 10 February 2020 (UTC)

• Support This seems obvious to me. If there is not consensus to show a banner, it should not be shown. Most of the opposing users are arguing, in essence, that CentralNotice is an important tool, which is entirely irrelevant given that no one is proposing completely disabling it. * Pppery * _{it has begun...} 22:47, 13 February 2020 (UTC)

• Question: At least half of the oppose !votes above seem to have been made during a single ~24-hour period (4 February), before which the discussion was nearly unanimous in support. Was it advertised at a new location? At the time, it had already been on
  WP:CENT for two weeks, so I would have thought that there weren't any additional venues to be notified. Sunrise (talk
  ) 20:44, 17 February 2020 (UTC)

Comments below to Support or Oppose

• Pointless you are already required to give a notice. That's what the big red box before the "Click here to start a new discussion" is all about. We are also not a bureaucracy, dismissing any report because of someone overlooking paperwork is nonsense, and the prohibition of refilling something for 3 days because someone didn't dot all the i's and cross all the t's is equally nonsensical.
  b
  } 08:30, 14 February 2020 (UTC)

There is a point. ANI is sorta like a strong reporting user feature for user to use. Once they get blocked. Its going be very difficult to appeal.Regice2020 (talk) 09:09, 14 February 2020 (UTC)

By pointless, I'm referring to this RFC. You are proposing to require something which is already required.

b

} 09:13, 14 February 2020 (UTC)

Regice2020, the editors mentioned are supposed to be notified at the time of posting. Failing that, the editor is notified as soon as someone notices that that hasn't been done. 24 hours makes no sense in this regard. What, are we gonna keep discussing a user without notifying them just coz the 24 hour deadline isn't crossed? The rest of your proposal has got problems as well. The only merit I see is in making the notice contain a direct link to the section of the discussion cos I think that's not always done and ANI is a pain of a page to skim through. That said, that can't become a rule, just a polite recommendation. Usedtobecool ☎️ 09:28, 14 February 2020 (UTC)

• What problem are you trying to solve here? Is the current system failing in some way? Can you provide an example of a real situation when the current system failed where this new system would have succeeded? – Teratix ₵ 09:22, 14 February 2020 (UTC)

  ^^^^ This is really all that needs to be said. We don't create rules and procedures to fix problems that are merely possible or hypothetical. Even three or four cases out of hundreds wouldn't be enough, and we have yet to see one. ―Mandruss ☎ 09:38, 14 February 2020 (UTC)

• Oppose. Currently, if the reporting editor doesn't notify, someone else does while pointing out to the reporter that they should have done so. Minor overhead. Under this requirement, someone has to watch the clock, close the report with prejudice after the 24-hours-without-notification point, and keep an eye out to ensure that the report isn't reopened until 3 days have passed. Major overhead. And it's highly unlikely that the change would ensure that fewer editors would fail to notify. It doesn't happen that often as it is, and it's those editors who seldom read AN/I who fail to notify and since they don't hang out there, they won't "learn" from others' mistakes. AN/I discussions just don't happen without someone making sure the reported editor has been notified. Schazjmd (talk) 15:41, 14 February 2020 (UTC)
• Oppose. If it isn't broken, don't fix it. --Dthomsen8 (talk) 18:31, 14 February 2020 (UTC)
• Can someone please close this as an obviously ill-thought out proposal? The current instructions mean that any editor who is the subject of an ANI report must be informed (and that would be in far less than 24 hours: more like 24 minutes or even 24 seconds) and the 3-day limitation on bringing up an issue that fails this procedural hoop is simply absurd: if the report is about an urgent incident then we shouldn't wait before dealing with it just because the originator didn't follow the instructions properly.
  Phil Bridger (talk
  ) 18:42, 14 February 2020 (UTC)

• Comment I don't think this is necessary. but I thought of an interesting idea. Similar to how GA process notifies nominees and reviewers. Should a similar template be created that once filled out it notifies the parties in question?Blue Pumpkin Pie ^{Chat Contribs} 19:12, 14 February 2020 (UTC)

• Oppose The effort one goes through to enact this proposal each time someone doesn't leave a talkpage notice would be better spent leaving talkpage notices. If the person who brought someone to ANI forgot to notify them, and you notice this, just do it yourself. Problem solved. --Jayron32 15:30, 17 February 2020 (UTC)

Reply Yes, that can happen, but it better for the person reporting to notify. If you want get someone blocked or topic ban then YOU have do something correctly too. I mean when a user uses ANI to report someone ...they can get blocked or banned. ANI is very powerful reporting system. I do not recall ANI process expansion is never a option? Regice2020 (talk) 01:36, 18 February 2020 (UTC)

Option B

The ANI starting process need improvements. Users who start a ANI about an editor is now required to leave a ANI notice with a direct link to that ANI discussion on editor`s talk page right away. The Admin must check ANI starter before starting a review. Otherwise, the report maybe on hold and achieved at some point by the bot, Regice2020 (talk) 08:09, 15 February 2020 (UTC)

Once again you are making the assumption that the process needs improvement without providing evidence. Can you point to any ANI report that would have been handled better if this change was made? It seems to me that, as stated above, either the person making a report follows instructions and informs the target editor or, if not, someone else does this for them. Have there been any recent ANI reports where the target has not been informed?

Phil Bridger (talk

) 09:01, 15 February 2020 (UTC)