Wikipedia talk:Vector 2022

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

This is an old revision of this page, as edited by Stockmausen (talk | contribs) at 14:09, 21 January 2023 (→‎Width issue: Reply). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

Graphic images in Vector 2022 search results

Apologies in advance if this sounds a little silly given Wikipedia's policies. Just switched to using Vector 2022 and, given that its wide rollout to the default skin is only a few days away, I feel it prudent to put it out there that I have noticed a potential issue with a new feature where images appear when using the search bar. For instance, if a user were to innocently search for 'peninsula', the 'penis' article appears while typing (thanks to the first four letters matching) complete with, for lack of a better phrasing, a photograph of an elephant's schlong. This is still far from the worst such example!

I fully understand that Wikipedia is not censored and such images are often necessary for the encyclopedic value of the articles where they are placed. However, on the old skin these pictures never appeared while using the search bar, so people making innocent searches didn't have to see them. A new Wikipedia user could be discouraged from using the site if they are offended and/or disturbed by the images appearing as they try to search for topics that aren't explicit at all. This seems especially likely if they're not logged in, with no idea how to hide the images.

On that note I am curious to know what (if anything) will be done to prevent these unfortunate incidents while not compromising the quality of the articles. I am wondering if there will be some sort of technical means implemented to hide specific images from search results and previews, like how the bad image list ensures certain pictures (many of which are sexual, violent or otherwise considered repulsive by many) aren't shown on pages other than which they are allowed. There could also be an option for logged-in users to choose whether to allow explicit images in their search results, with the default setting being to hide them.

Thanks for taking the time to read through this and would love to know your thoughts on the matter. Entranced98 (talk) 03:16, 14 January 2023 (UTC)Reply[reply]

This is a decent point. While we've traditionally had WP:NOTCENSORED, some of this is based on the philosophy that someone visiting the penis article should expect to see a penis. It's obviously not the same philosophy if you're given search suggestions when trying to find a different article, and of course we do avoid potentially objectionable material on unrelated articles, if there is no clear encyclopaedic purpose. (Wikipedia:Offensive material is a decent summary.) ProcrastinatingReader (talk) 13:03, 14 January 2023 (UTC)Reply[reply]
These images have appeared like this on mobile for some time, so just imagine what all the mobile users have to deal with. :)
There is a separate list (MediaWiki:Pageimages-denylist) that can be used to eliminate search results without affecting the page directly. You can also add |class=notpageimage to a specific image if it's inappropriate to be the page image. As page images are only ever used in 'secondary' locations, this concern is a good reason to use the class. (I don't know if all infoboxes support the second option today.)
That there is a second list is the subject of phab:T306246. Izno (talk) 18:05, 18 January 2023 (UTC)Reply[reply]
We should have a way to prevent that article from showing up on the list until you type out its full name. While I understand that if you type out the full name of an article it will show up on the list, it's name in this case is not a header of "peninsula". Animal lover |666| 06:55, 19 January 2023 (UTC)Reply[reply]
That's somehow more restrictive and functionally disruptive than what's being proposed. Instead of not showing an image, autocomplete is basically gone for fear of words. Artoria2e5 🌉 15:30, 20 January 2023 (UTC)Reply[reply]
The WMF added images to the global search results years ago. I warned them that there would be disruption from people showing up demanding they be censored. I'm pleasantly surprised we avoided it for a couple of years.
Anyway, WP:NOTCENSORED is policy. One of the reasons we agrees on that policy is that we don't want the massive disruption from people arguing over which images they do and do not want to censor. Before anyone attempts to actually advocate this really really bad idea, let me preview how this goes. First of all, our number one group of objectionable images are the images of Muhammad. There no zero chance our medical and scientific and sociological articles are going to have their images censored because "ohmygodboobies", without first dealing with the images of Muhammad issue. Good luck with that. Secondly, next up on the list will be demands for an equal right to censor any image related to Christianity or any other religion. Third, censorship-advocates all know it's simple and obvious which images to censor, but they wind up battling each other because no two can agree on which ones. NOTCENSORED have been repeatedly and exhaustively debated, and challenges to it always fail. It would be unrealistic to expect otherwise. Alsee (talk) 11:50, 19 January 2023 (UTC)Reply[reply]
One of the main arguments for NOTCENSORED (which I am one of the biggest advocates for) is that potentially offensive images are only shown in an appropriate context, but they are being shown here out of context. e.g. images of a penis are contextually appropriate (and indeed encyclopaedically necessary) on the penis article and nobody here is arguing to change that. However images of penises are not contextually appropriate on the peninsula article (not even at Mull of Kintyre#Mull of Kintyre test) - the argument here is only that people searching for things that are not penises should not see images of penises in their search results. It may also be the case that people would wish to prevent images of the Islamic prophet being shown to people searching for Muhammad Ali but, just as with the male genitalia argument, this is not censorship in any way shape or form. Thryduulf (talk) 12:26, 19 January 2023 (UTC)Reply[reply]
Of course it's censorship. It's just censorship which you happen to agree with, but that's always true when anyone advocates cesorship Alsee (talk) 21:35, 19 January 2023 (UTC)Reply[reply]
I've explained in detail why choosing not to display irrelevant images to people looking for something completely different is not censorship, you cannot successfully refute that by just asserting that it is - especially as the ideal long term solution to the issue would result in irrelevant images not being shown regardless of potential offensiveness (e.g. a person searching for Chessington World of Adventures would not see the lead image of the chess article). Please explain how and why this is censorship, with reference to something other than hand waving and personal assertions. Thryduulf (talk) 23:49, 19 January 2023 (UTC)Reply[reply]
Your asserted rationale is not to display irrelevant images to people looking for something completely different. But that's not really true, now is it? You're not trying to prevent the irrelevant peninsula image from being displayed when I search for penis, are you? No. You're proposing to selectively censor images that you consider offensive.
the ideal long term solution to the issue would result in irrelevant images not being shown - if you want to improve the search engine to avoid irrelevant search results, of course, everyone supports that. But you're not advocating eliminating irrelevant results. No. You know the search engine is already making a best-effort to show the most likely and most relevant search results. What you're proposing is to censor images from search results even when they are exactly the desired result. You either overlooked, or didn't care, that the proposal is selectively discriminatory and detrimental against the people who were searching for exactly those articles.
Setting aside all that, your most fatal point is in trying to define an image list. We've debated this issue repetitively and exhaustively and it always ends the same way. You don't get to be dictator deciding which images go on the list, and the community doesn't want endless disruption arguing over individual images. The debate always goes the same way. The pro-censorship side walks in assuming it's obvious which images should be censored, and by the end the debate they've pretty well killed their proposal themselves warring with each other over which content should_be/cannot_be be censored. Alsee (talk) 09:24, 20 January 2023 (UTC)Reply[reply]
Nice theory, shame about the facts. The image list (which currently exists in the form of the MediaWiki:Bad image list, which doesn't see endless arguments) is being proposed as a workaround for the issue. If there is a way to show images of penises when people are searching for penises but not at other times (and the same for other images, of anything), then please go and propose it at the phabricator task where devs are (or at least should) be listening. This is what I and at least most other people want long term, however we realise that this is a very hard problem (not just for MediaWiki, its at best a not-fully solved problem in computer science) so in the short term we are looking for a solution to an actual problem that has minimal side effects. You going to have to work a lot harder to demonstrate that not showing people some images in search results (but showing them those exact images when they look at the article) is either censorship or discrimination, because they simply are not. Thryduulf (talk) 11:48, 20 January 2023 (UTC)Reply[reply]
To paraphrase someone on the phab, the BIL page is an administrative tool to prevent people from spamming pics, not to really define what should be surpressed (since we are talking about "not censorship"). We don't have a spamming problem here, do we? Artoria2e5 🌉 15:57, 20 January 2023 (UTC)Reply[reply]

I support the point forwarded by [Entranced98] and supported by [Animal lover |666] and [Thryduulf].


First: Respect for more than 15 years.

On the issue:
If one has finished typing something and hits [Enter], one, of course, should get this. This is "Non-censor".

But if one is still typing and has NOT yet hit [Enter], there is no need to bother the more sensitive users with images which infringe on their morale. They should feel comfortable too to use the WP, if possible.

This is the same respect which we are supposed to show regarding living persons or non-binaries or why we do not tolerate personal attacks. To them it IS an offence, but it is avoidable by some technical efforts, without having to censor what others do want to read.

Therefore we (our programmers) should try to find ways to satisfy both.

Ping welcome, Steue (talk) 14:28, 20 January 2023 (UTC)Reply[reply]

Keep the old skin

I am against changing the Wikipedia design. I hate Vector 2022. SonicIn2022 (talk) 15:44, 14 January 2023 (UTC)Reply[reply]

the main issue you cannot hide on mobile the contents on the left, it is not saved. Which makes reading stuff impossible. So much for dynamic rendering. Valery Zapolodov (talk) 21:29, 14 January 2023 (UTC)Reply[reply]
@Valery Zapolodov You can??? Just press hide for the table of contents and the《 for the actual sidebar Aaron Liu (talk) 02:18, 15 January 2023 (UTC)Reply[reply]
It is not preserved. You need to press it every time. Valery Zapolodov (talk) 12:02, 15 January 2023 (UTC)Reply[reply]
Ah. Fair point.
I guess they expected mobile users to still use Minerva Neuve. Aaron Liu (talk) 15:20, 15 January 2023 (UTC)Reply[reply]
@Valery Zapolodov, if you're referring to Vector 2022 on a mobile browser, I've tried this and for me, it is persistent - when I make it uncollapsed and go to another page, it's still uncollapsed. When I make it collapsed, it stays collapsed. It doesn't work like that for logged-out users, because it's basically a preference, and there are no preferences for logged out users (that part isn't related to the skin at all). Could you confirm if you're logged in when browsing on mobile? SGrabarczuk (WMF) (talk) 21:09, 18 January 2023 (UTC)Reply[reply]
I do not use a PC for wikipedia. And of course I was logged in, I use Vector 2022 for 1.5 years already. Valery Zapolodov (talk) 00:57, 19 January 2023 (UTC)Reply[reply]
@SGrabarczuk (WMF) I can reproduce this issue. It doesn't recollapse automatically even if I just refresh after clicking hide. Aaron Liu (talk) 01:52, 19 January 2023 (UTC)Reply[reply]
@SonicIn2022 @Valery Zapolodov and others, thanks for your feedback. We are working on making the collapsed/uncollapsed state of the table of contents persistent across pages. You can follow that work here:
We made the table of contents collapsible upon request from the community. We also decided to automatically collapse it at 1000px. However some people said that when they collapsed it, or when it automatically collapsed, they didn't know where it had gone. So we then created a task for adding an indicator to make it super obvious where it has gone:
As a fall-back, until we add the indicator, we thought we should make the table of contents reappear if you switch pages or reload, in case people can't figure out how to get it back. However perhaps we should just skip to making the table of contents state persistent, and then add the indicator afterwards. AHollender (WMF) (talk) 16:56, 19 January 2023 (UTC)Reply[reply]
Yeah it should be persistent Aaron Liu (talk) 17:47, 19 January 2023 (UTC)Reply[reply]
@SonicIn2022 Have you tried for at least a week? Any overhaul would require a week to get used to. What specific grievances do you have against the new skin? Also, we’re not going to remove the old skin as your title “keep the old skin” suggests, you can always change it in settings. Aaron Liu (talk) 02:20, 15 January 2023 (UTC)Reply[reply]
I already tried it on Miraheze. The new skin sucks. SonicIn2022 (talk) 13:10, 15 January 2023 (UTC)Reply[reply]
The Miraheze version hasn't had a lot of new fixes introduced yet, Miraheze is on 1.38 while the latest version is 1.39. Try it out on THIS wiki. Plus, what specific grievances do you have towards the new skin? The possibility that it hasn't been bought up in the mentioned rfc is very low. Aaron Liu (talk) 15:19, 15 January 2023 (UTC)Reply[reply]
I can't even log in on the redesigned site in NetSurf (no javascript; can't use the "hamburger icon" even if I wanted to). And it renders poorly. Using it for a week would be torture. Wyatt8740 (talk) 19:45, 19 January 2023 (UTC)Reply[reply]
Agreed. I absolutely hate the redesign. I think it's important to remember that the web consists of more than mobile devices. The new layout on desktop browsers is terrible. I literally thought I had accidently loaded the mobile version and then thought there must be a bug that keeps serving me the mobile version. I was shocked to learn this was the new layout. I really hope this will be fixed. Wikipedia is far too important to be this degraded on desktop. In my view all of the changes should be reversed on desktop. At the very least give us an alternate URL that can serve the old design. Something like Reddit did with when they moved to their mobile-first redesign. (talk) 21:21, 19 January 2023 (UTC)Reply[reply]
The official oh-so-visible faq says this:

Anonymous preferences would make the pages load too slowly.
Most traffic comes from logged-out users. To handle that, we have a few "caching servers" which only save and send "snapshots" of web pages. These "snapshots" are up to 7 days old, are a replacement for generating web pages, and are the same for all logged-out users. This allows us to serve pages quickly.
Preferences require generating different versions of web pages. Making this possible for logged-out users would overload our servers. We don't want to do that also because we need to reduce cache fragmentation.
The only possible way of providing preferences for logged-out users now is making the settings load always after the page. This takes much more time to load and looks odd. For example, if a logged-out user was to see the dark mode in action, then immediately after loading each page, they would first see the light interface for a short moment, and then the interface would become dark.
— mw:Reading/Web/Desktop Improvements/Frequently asked questions

The same rationale could be used to say why they don't do the old reddit thing. It is possible to redirect every link to with browser extensions such as fastforward though. Aaron Liu (talk) 22:34, 19 January 2023 (UTC)Reply[reply]
Desktop should not be forced to use a mobile design. It's as wrong as forcing a mobile device to use a desktop design. The web is intended to allow any device to connect and view content. It's a great idea that works well when web designers accept that their content will look different on different devices. This current design is complete failure to consider the needs of desktop users. It should be reverted entirely on desktop. (talk) 22:20, 19 January 2023 (UTC)Reply[reply]
The same FAQ also talks about how the thing isn't mobile influenced. I do agree that the white(not the limited width grey) is too excess though. Aaron Liu (talk) 22:38, 19 January 2023 (UTC)Reply[reply]
Literally just made an account to say I find it distasteful it as well. So much wasted screen real estate. The fact I had to make an account to revert it makes for a very negative user experience. Scerttle (talk) 08:30, 20 January 2023 (UTC)Reply[reply]
Agreed. Forcing me to create an account in order to disable a skin when you can literally EDIT the pages without one is absolute madness. TheDelmain (talk) 03:53, 21 January 2023 (UTC)Reply[reply]

I hate it too, and I've hated it on French Wikipedia since I first saw it (some years ago, so don't preach about "getting used to it" -- it's been terrible for years!). It's a massive waste of screen space that now appears bloated more than practical. It's been terrible since it started and even using the "expand width" button still leaves less than half the content per screen of the old design. And don't tell me that account preferences cover this problem, because half or more of my browsing is not logged in. Please immediately improve this design to match the old design in terms of content-per-screen area (or of course just reintroduce the old design). Damn I feel like I need to zoom out my browser just to read it properly. Oh and the Table of Contents is much harder to find and use, it may as well just be deleted entirely. Spartan S58 (talk) 23:56, 18 January 2023 (UTC)Reply[reply]

So true about ToC! Valery Zapolodov (talk) 00:59, 19 January 2023 (UTC)Reply[reply]
I completely agree. The ToC should be an essential part of the article's body and should stay below the lede. The new ToC is completely useless and confusing. Also cfr. my RfC comment about it. Æo (talk) 19:23, 19 January 2023 (UTC)Reply[reply]
oh yeah, I totally forgot it was the skin on fr.wikipedia before being here as well Vincent-vst🚀 (talk) 11:40, 19 January 2023 (UTC)Reply[reply]

I hate when people try to reinvent the wheel. The new interface is just annoying for us not invested. The skin should be turned off by default.--Blockhaj (talk) 00:54, 19 January 2023 (UTC)Reply[reply]

The new skin is an abomination. "A fresh new look!" has been the bane of Internet users since the beginning. If something works and looks good, DON'T CHANGE IT! I see absolutely no benefit to the new interface being the default. Ambarenya13 (talk) 03:17, 20 January 2023 (UTC)Reply[reply]
I see some benefits In Vector 22, the sections are displayed in the sidebar which before were within the article. If one chooses to hide it, it will optionally be accessible over the bullet menu left to the article name. Then the dropdownmenu at the person in the upper right, which provides much more space in the toolbar. Then also a direct link to the uploaded media in commons is good. Paradise Chronicle (talk) 09:27, 20 January 2023 (UTC)Reply[reply]
Such big spaces around centered column of text is ridiculous. Turn it off by default!
New theme is a waste of space and nothing more. You want to provide the better experience via new navigation - well, make it hide-able. Without any unnecessary spaces around article. (talk) 22:39, 20 January 2023 (UTC)Reply[reply]

Keep the old buttons

If I hide the menu, it turns into one of those Google-influenced three-line menu icons. This is dumbed-down mobile phone iconography. When I started using mobile phones (grudgingly) its meaning was mysterious to me. On the English Wikipedia, why can't it just be a button that says "menu", you know, in English? Isn't this a site for people who can read? Similarly, if I want to log in now, I have to press a button that looks like three dots. In what way does three dots mean "log in"? Is this another icon with a familiar meaning to the young and stupid? It doesn't mean anything to me, and why could it not be a button written in English which says "log in"? And then there's the contents on talk pages: when hidden (which I will be doing all the time because the changes take up too much screen space) it becomes an icon consisting of three lines with bullet points, which you're supposed to infer is some kind of menu different from the first kind of menu, and thus means "contents" ... right? But what would be wrong with a button that says "contents"? This site is not for an international audience. The articles are not written in pictograms. Why do the buttons have to be like that?  Card Zero  (talk) 20:19, 18 January 2023 (UTC)Reply[reply]

I concur. I don't think it's going to matter, but this should be changed. YouCanDoBetter (talk) 20:57, 18 January 2023 (UTC)Reply[reply]
While I concur, the icon isn’t mobile or Google influenced, it’s the Hamburger button and it has been around since the first commercial PC with a GUI in 1981 Aaron Liu (talk) 22:44, 18 January 2023 (UTC)Reply[reply]
I wondered whether somebody would provide a history! But it's been popularized by mobile apps, I think. The article cites "Loving & Hating the Hamburger Icon", written in 2018, which says "this icon has exploded in popularity".  Card Zero  (talk) 10:54, 19 January 2023 (UTC)Reply[reply]
Doesn't matter how long it has been in existence, I hate the 'Hamburger button'. It is very inconspicious and usually looks like some graphic element or decoration. It is not at all intuitive to find if you don't know where to look.
I general, I support the new look as the old one produced unreadable walls of text in large displays: however the new one is over-minimalistic. "Clean" web design is a mean, not an end. Here is a hint to new UI designers who want to hide stuff behind drop-down or hover-the-cursor menues: DON'T. They suck. Put the damn links where I can see them.-- (talk) 11:41, 19 January 2023 (UTC)Reply[reply]
Hey @Card Zero and others — thanks for your feedback here. A few things:
  • I agree it's best practice to use an icon + a text label for buttons. We try to sparingly use only icons.
  • In the cases above — main menu, personal tools menu, etc. — we believe the icons are sufficient. Whether we like it or not these icons have become ubiquitous, and are used on all major websites. This doesn't necessarily make it a good design pattern in some fundamental sense, but it means that in fact most people are familiar with those icons and their meanings. We have tested this with people on They are easily able to locate the menus and find items within them.
  • Because Wikipedia is served in over 300 different languages, using labels creates a lot of variation in terms of the size of those elements. For example, we currently have a button with a text label that allows you to pin the collapsed table of contents back to the sidebar. In English the button label is "move to sidebar", whereas in German the button label is "In die Seitenleiste verschieben", which makes for a pretty wide button.
  • Another thing to consider is the visual effect the buttons have on the overall interface. For example, in the sticky header we tried out using buttons with text labels, icon buttons, and buttons with icons + text labels. Having buttons with text labels in the sticky header, sitting at the top of the screen, to us felt heavy and crowded. In the past we've been able to get away with using smaller elements (styled as links) for buttons, which can help with this. But these days we have a design system and accessibility requirements we have to adhere to, which ultimately controls the size of buttons.
I hope some of this context is helpful. Please feel free to respond and we can continue to discuss this. AHollender (WMF) (talk) 17:07, 19 January 2023 (UTC)Reply[reply]
With respect, the Wikimedia ethos is not to serve "most people", but to serve "all people". We strive for maximum accessibility, even when it sacrifices aesthetic conciseness. Ibadibam (talk) 18:24, 19 January 2023 (UTC)Reply[reply]
  • There is currently hover text for "main menu" (the hamburger button), but the three dots have hover text "more options", which is completely uninformative. (It tells us only that the button won't do anything immediately.) It could say "user", but that's also vague, and it would be best if it said "log in/out, user talk page, contributions", maybe on three lines. It's not like there's any shortage of space for mouseover text. Don't imagine Wikipedia users are unwilling to read. The third button should have hover text saying "contents", but it doesn't have any. Also, forget about "best practice" (see criticisms), consider what's best for this site.
  • Like the IP editor above said, "clean" shouldn't be a goal. If there's one website (note: not an app) where content matters over form, it's Wikipedia. We're not selling anything, and we're not trying to impress anyone. If "clean" helps you find information, that's good. If it hides information it's awful. Who cares if icon buttons are "sufficient"? We can do better than sufficient. You had no reason to get rid of the text buttons at all, other than trendy aesthetics. If people complain about being confronted by too much text, they didn't mean those little bits of text. If people complain about Wikipedia not looking "modern", ignore them.
  • Your example of "move to sidebar" invalidates itself, because that's a text button you've added. You haven't replaced that one with an icon, despite your reasoning that it would be long in German.
  • "Having buttons with text labels in the sticky header, sitting at the top of the screen, to us felt heavy and crowded" – I don't mean to be brutal, but nobody cares how you feel. We want Wikipedia to display information. Some people will complain how it "feels", sure, but why give them credibility? Designers are the wrong people to redesign it.
  • "We have a design system and accessibility requirements we have to adhere to, which ultimately controls the size of buttons" – what's this "design system"? A way of passing the buck? I assume this is about unifying skins in order to enable further tinkering across different languages of Wikipedia. Do you imagine we want that to be easier for you? We don't. You can have 20 different skins and deal with it, and if that slows down your tinkering, so much the better.
    • Accessibility requirements, I would have thought, require you to keep text, for screen readers.
How about backing down, dropping your pet idea, and putting the text back? That would be a remarkable thing if you could manage to do it. You seem to have an implicit mission of making Wikipedia look cool: I don't think anybody was consulted about that. Wikipedia should look comfortably unremarkable, like a good font.  Card Zero  (talk) 19:08, 19 January 2023 (UTC)Reply[reply]
Agreed - This log in problem (the 'hamburger menu') is especially a stupid and problematic change because it prevents users on browsers like NetSurf that don't use Javascript from logging in and returning wikipedia to a state of usability. Wyatt8740 (talk) 19:44, 19 January 2023 (UTC)Reply[reply]
I don't think it's just JavaScript. When I disable javascript on waterfox the entire page functions fine though many dropdowns don't open and take me down the link. Aaron Liu (talk) 22:31, 19 January 2023 (UTC)Reply[reply]

Width issue

The Vector 2022 update looks alright, however the default appearance with "limited width mode" is a major waste of monitor space. I recommend that limited width mode is disabled by default when the update goes through. Cards84664 16:18, 15 January 2023 (UTC)Reply[reply]

Hi @Cards84664. Thanks for your general support. Of course, you're free to disable the limited width permanently in your preferences. (Un-check the option "Enable limited width mode" or click the toggle in the bottom right corner of the screen.)
However, regarding the default view for everyone, I'd like to recommend reading the section of the FAQ: Why is the width of the content limited?. Also note the UX Myth #28: White space is wasted space. You'll find even more information on this feature page on
In the future, we, the Web team, would like to talk about the possibility of increasing the default font size (of course, having addressed a number of issues with tables, navboxes, infoboxes, etc.). That would be implemented together with an increase of the (still limited) default width. I'm not encouraging (or trying) to start this conversation now, though.
Thanks! SGrabarczuk (WMF) (talk) 17:24, 15 January 2023 (UTC)Reply[reply]
For times when limited width was needed there was an exceedingly easy solution in the old UI: resize the window. In the old system snapping the window to one side of the screen had the same effect as this, without forcing it on you in the many cases where it makes for a worse experience.
In the new UI, the large amount of white space and the narrow article leaves the reading experience feeling very squeezed, claustrophobic, and even oppressive. If you're talking about increasing the font size that's going to make it even worse. 2607:FEA8:2D24:8900:0:0:0:151C (talk) 18:44, 18 January 2023 (UTC)Reply[reply]
Per - the examples given as "Some beautiful websites with generous white space" now use minimal whitespace compared to or the new mediawiki skin - it may have been true a few years ago, but the new thing seems to be using the entire screen, but limiting the size of each "section" of the page such that the entire page can be filled with these "sections", rather than just using a minority of the screen for the content and keeping the rest blank, which is what uxmyths and vector 2022 both seem to do. 1rre (talk) 19:03, 18 January 2023 (UTC)Reply[reply]
In your link to UX Myth #28, there are no viable sources given, save for one mention of a "lab research PDF" from Wichita State University, and that link is broken. It is plausible the blog post in question was skewed in favor of advertisers to better accommodate ad space, something WMF does not need to concern itself with. Please endeavor to remove this link from the FAQ to prevent the spread of confusion or misinformation. (talk) 13:16, 21 January 2023 (UTC)Reply[reply]
Somewhat Agree. I understand whitespace is a critical, yet subtle part of web design, however there are times where I would like more text on screen at once, such as if I'm multitasking and cannot easily scroll the page. Perhaps a toggle button on the page between limited and unlimited width mode that dynamically resizes content (while keeping you where you were on the page) would be useful. Of course, I also understand that it is likely too late to even consider including design elements like that, but that is something I would appreciate. Deathstar3548 (talk) 22:14, 15 January 2023 (UTC)Reply[reply]
@Deathstar3548 - thanks for your question! Such a toggle button is currently available. The button is available on the bottom right side of the screen and it is persistent as you scroll down the page. You can also set the width to default to full with from the appearance section of your preferences by unchecking the enable limited width mode checkbox. OVasileva (WMF) (talk) 15:28, 16 January 2023 (UTC)Reply[reply]
Yes please do disable limited width mode. Having this enabled was putting me off Vector 2022. Aaron106 (talk) 16:11, 18 January 2023 (UTC)Reply[reply]
Wider tables benefitted from the wider screen. The sideways compression has produced an amateurish appearance for such tables. It would be more useful to restore the former version as the default and make the new version an option for those who may find it useful. DMBanks1 (talk) 17:36, 18 January 2023 (UTC)Reply[reply]
The button is bugged. Opening a new page resets it back to limited width mode every time. 2607:FEA8:2D24:8900:0:0:0:151C (talk) 18:35, 18 January 2023 (UTC)Reply[reply]
Yes, Please fix this.
Having to reset view from garbage mode where 40% of screen is wasted, to normal usable mode where entire screen is used, for every single page opened is just... I don't have words. I don't want to insult anybody, but that entire article that is being linked as reason for whitespace, is one massive [citation needed]. I can write an article that says exact opposite, and without actual peer-reviewed scientific research done on the subject, both will jsut be unsubstantiated tosh.
I have witnessed many pages change their UI, and every single time when the UI was changed into "modern" tablet-and-phone-friendly version, it just got objectively worse to use. Same operations would require more input actions (like clicking the hamburger to get the menu - why? why not just have it there all the time? That would be a PERFECT use for the space the new design is wasting with limited mode. But no, all the apps have to be Apple copycats and not have any UI elements at all apparently. Because got forbid there is any UI in the UI. All we care about is looking modern, and functionality be damned.
Can someone explain to me, why is the current trend to ignore existing technology stack that makes splitting the page into normal version and mobile version trivial, and instead PC users are forced to use garbage mobile interfaces? Wikipedia was one of last web pages that had true and functional UI. UI that was subject to the functionality it was to facilitate. This new UI is the other way around. Everything is upside-down - the functionality is subject to the design and "modernness" of the UI.
Yes, the old UI was not great for mobile - that's why pages are supposed to have mobile versions, not have the UI gutted for PC because many people prefer to caress their phone screens.
Objective downgrade. (talk) 11:47, 20 January 2023 (UTC)Reply[reply]
I completely agree, the limited content width mode is a waste of space. It makes the content too condensed and difficult to read, as it requires more scrolling and doesn't allow to fit as much content on the same screen. On some pages the content is unreadable in this mode (e.g. see the table in
Also, requiring user registration in order to be able to fix the now-broken page design is unacceptable. Wikipedia should remain freely readable by unregistered users. And by "readable" I mean the user should not have to disable the limited content width mode on every page he visits or install browser extensions to revert to the old skin that actually works.
Please disable the limited content width mode by default or make it remember the user's choice for unregistered users. 2A02:2168:84D9:2200:D6CB:69C2:6A20:23C2 (talk) 21:13, 18 January 2023 (UTC)Reply[reply]
Completely agree. My desktop browser is using like 50% of available space for no good reason. Developers please stop and think about how your web layout will look on more than just phones.
There is simply no justification for this fixed width design. (talk) 21:33, 19 January 2023 (UTC)Reply[reply]

Even with limited width mode disabled the Vector skin wastes a lot of space. I have made some minor CSS tweaks in my own CSS file to fix these problems, and I offer them to the community or to the developers to consider, either as a permanent change or to incorporate into your own personal CSS. See User:Anachronist/vector-2022.css. ~Anachronist (talk) 00:01, 19 January 2023 (UTC)Reply[reply]

It is possible to minimize eye movements AND use the available space. Use columns, like newspapers do for a long time already. Ideally a column would be so narrow that a reader could see all words in one eye fixation, of multiple lines. This would allow the brain to process text in parallel. Yes, not does much matter of words the sequence. The current solution does not allow fast parallel reading. Words of one sentence on one longer line require horizontal eye movements. With long lines the eye sees words of previous and next line, but the brain has to ignore those. What a waste. Use narrow columns for fast reading and maximum usage of screen space!

Uwappa (talk) 10:21, 19 January 2023 (UTC)Reply[reply]

@Uwappa — columns are an interesting idea that other people have suggested. However columns don't work unless you have pagination (like newspapers, or PDF journal articles do). If you don't have pagination it becomes very difficult to decide how tall to make the columns. Try testing your idea out on an article with a long section. AHollender (WMF) (talk) 17:10, 19 January 2023 (UTC)Reply[reply]
Thank you. On screens the alternative seems easy: max column height could be viewport height, 100vh.
Unfortunately there seems no CSS tag available to specify max column height. So, I see your point, long sections will create columns that are higher than the viewport.
seemed promising, with css that can go both ways, with and without columns depending on viewport height, but :( that example uses a height hardcoded in px (pixels). So I see your point, the column idea is appealing, but the implementation is tricky.
An idea for an article with a long section:
  • Apply a set of columns per paragraph, the html p tag.
  • A new paragraph will start new columns.
  • blank line after a paragraph for vertical separation.
The idea put to the test, with a real life example, the long intro of Elizabeth_II, four paragraphs:

Elizabeth II (Elizabeth Alexandra Mary; 21 April 1926 – 8 September 2022) was Queen of the United Kingdom and other Commonwealth realms from 6 February 1952 until her death in 2022. She was queen regnant of 32 sovereign states during her lifetime, and was head of state of 15 realms at the time of her death. Her reign of 70 years and 214 days was the longest of any British monarch and the longest verified reign of any female monarch in history.

Elizabeth was born in Mayfair, London, as the first child of the Duke and Duchess of York (later King George VI and Queen Elizabeth The Queen Mother). Her father acceded to the throne in 1936 upon the abdication of his brother Edward VIII, making the ten-year-old Princess Elizabeth the heir presumptive. She was educated privately at home and began to undertake public duties during the Second World War, serving in the Auxiliary Territorial Service. In November 1947, she married Philip Mountbatten, a former prince of Greece and Denmark, and their marriage lasted 73 years until his death in 2021. They had four children: Charles, Anne, Andrew, and Edward.

When her father died in February 1952, Elizabeth—then 25 years old—became queen of seven independent Commonwealth countries: the United Kingdom, Canada, Australia, New Zealand, South Africa, Pakistan, and Ceylon (known today as Sri Lanka), as well as head of the Commonwealth. Elizabeth reigned as a constitutional monarch through major political changes such as the Troubles in Northern Ireland, devolution in the United Kingdom, the decolonisation of Africa, and the United Kingdom's accession to the European Communities and withdrawal from the European Union. The number of her realms varied over time as territories gained independence and some realms became republics. As queen, Elizabeth was served by more than 170 prime ministers across her realms. Her many historic visits and meetings included state visits to China in 1986, to Russia in 1994, and to the Republic of Ireland in 2011, and meetings with five popes.

Significant events included Elizabeth's coronation in 1953 and the celebrations of her Silver, Golden, Diamond, and Platinum jubilees in 1977, 2002, 2012, and 2022, respectively. Although she faced occasional republican sentiment and media criticism of her family—particularly after the breakdowns of her children's marriages, her annus horribilis in 1992, and the death in 1997 of her former daughter-in-law Diana—support for the monarchy in the United Kingdom remained consistently high throughout her lifetime, as did her personal popularity. Elizabeth died at Balmoral Castle, Aberdeenshire, in September 2022, at the age of 96, and was succeeded by her eldest son, Charles III.

Please test the example above: make your window smaller and wider. Test it on a mobile and computer monitor. The columns will adapt. Uwappa (talk) 05:59, 21 January 2023 (UTC)Reply[reply]
The example is extremely unintuitive. I have never seen such a design anywhere, in print or on the web. I cannot imagine anyone would prefer this to Vector 2022. Stockmausen (talk) 14:09, 21 January 2023 (UTC)Reply[reply]

how to KEEP the old skin?

Here we go again - there is a FAQ entry for turning off the new skin, but complete silence as to what I should do today to ensure I'm not switched away from my preferred skin: Vector legacy.

I'm not interested in Monobook, Timeless etc. I'm not interested in having a look at the new skin, even for a second. How do I decline the switch? CapnZapp (talk) 18:57, 15 January 2023 (UTC)Reply[reply]

I believe you'll be switched, but you can immediately go to your preferences and click Vector legacy (2010) to go back. It's right above MinervaNeue and Monobook. —Femke 🐦 (talk) 19:56, 16 January 2023 (UTC)Reply[reply]
CapnZapp - Femke is correct. We cannot decline. Declining this feature is not an option. Vector 2022 will be the default, all editors/readers will be switched over to Vector 2022 and then folks who don't want Vector 2022 will have to manually opt-out. Shearonink (talk) 00:21, 18 January 2023 (UTC)Reply[reply]
Just another example of the atrocious attitude WMF has towards its users then. With fiasco after fiasco where the engineers foist new functionality upon users (structured discussions, flow, banners you can only get rid of by overly technical changes, and a general insistence to dumb down the interface) with zero thought of asking them, allowing them to choose, and/or properly documenting "what if I don't want to?" it is clear the value put upon customer relations is next to zero. But thanks to you two for clearly stating and owning up to what the dev team clearly couldn't be bothered to! CapnZapp (talk) 07:00, 18 January 2023 (UTC)Reply[reply]
Honestly I don't see a very easy way to do this. The dev team have said that adding a new preferences option requires a lot of work so I don't think they'd be willing to add something new just for a week. Aaron Liu (talk) 13:08, 18 January 2023 (UTC)Reply[reply]
Hey @CapnZapp I understand where you're coming from, and I'm sorry you're frustrated. I agree that we should have made it easier for people to avoid getting their skin switched automatically.
I don't think your characterization of our team's process and efforts here is quite fair. Over the past 3 years we've engaged deeply with over 1,900 individual community members (I'm talking pages upon pages of phab task discussions, talk page threads, etc.), and hundreds of logged-out users. The feedback we've gotten has been overwhelmingly positive, and of course there are people who don't want change. We have been mindful of the second group of people all along, and have tried to make it as easy as possible to opt out at all points in time. We're far from perfect, but the characterization that we don't care, have an "atrocious attitude", and aren't constantly trying to do better is inaccurate in this case. AHollender (WMF) (talk) 17:20, 19 January 2023 (UTC)Reply[reply]
"and have tried to make it as easy as possible to opt out at all points in time"
This is not true.
If this was true you would have made this change an "opt-in for logged in users" only change.
You did not.
You chose to enforce the change on everyone with the feedback of less than 3000 people.
On a website that has millions of logged out users each day.
I use Wikipedia daily, multiple times.
I never log in, and i never once saw anything mention about how you were plotting to ruin the wiki experience of millions of desktop users to make wiki look a bit flashier on mobile.
I you had done a proper consultation you would have been told to not do it.
Which, of course, is why you all carefully limited who would find out about this change until it was too late to object or do anything.
Duplicitous is the correct word for this series of choice made by the WMF.
Shame on all of you for being dishonest.
Give us Old Wiki back and set this silly "modernist" design as an opt in for the people who want tons of white space on the screen pointlessly. (talk) 10:56, 20 January 2023 (UTC)Reply[reply]
forgot to (talk) 10:57, 20 January 2023 (UTC) cos i was annoyed still at the stupidity of the new "design" (talk) 10:57, 20 January 2023 (UTC)Reply[reply]
They acted the decision on a misguided community closing of WP:V22RFC. In that closing two very esteemed community members misguidedly decided that if certain things were fixed there would be no need for another consultation and the thing could just roll out. They did do a proper consultation. Plus in the week before this change there were banners everywhere telling people there was going to be a skin change. Aaron Liu (talk) 12:27, 20 January 2023 (UTC)Reply[reply]
To add: obviously I realize that I could have switched temporarily to Minerva or Monobook just through the switch and then switch back. But why isn't even a single engineer asking the question "why do we force users that want no change to do change?" :-( CapnZapp (talk) 07:03, 18 January 2023 (UTC)Reply[reply]
Sigh, another site getting a new UI that normal users didn't even heard of, I guess the "don't fix it if it ain't broken" philosophy does not apply to the IT field. I am switching back to the legacy version. WZNGT (talk) 07:20, 19 January 2023 (UTC)Reply[reply]

Did I miss something? I thought this was supposed to happen when - today, January 18th in this timezone - is Vector 2022 going to come online as the default? Shearonink (talk) 15:33, 18 January 2023 (UTC)Reply[reply]

@Shearonink See the § What to expect on January 18th, 2023 section Aaron Liu (talk) 15:36, 18 January 2023 (UTC)Reply[reply] seems to have just switched over at least on some pages...yuck...too much whitespace. Enabling my Vector Legacy preference in 3...2...1 Shearonink (talk) 15:55, 18 January 2023 (UTC)Reply[reply]
Oh CapnZapp, you might have to search around to find your Preferences. In the previous Vector iteration Preferences were laid-out in a single line across the top of the page. With Vector 2022 they've been moved to the very top right-hand corner, in sort of a drop-down menu arrangement or button. Shearonink (talk) 16:02, 18 January 2023 (UTC)Reply[reply]
I set my skin to 2010 globally a couple of days ago and wasn't switched over today, so something was possible. -- LCU ActivelyDisinterested transmissions °co-ords° 19:08, 18 January 2023 (UTC)Reply[reply]
When grey space appears there’s a bottom right button to go full width, unless you’re talking about the lack of dividers which I agree is a problem Aaron Liu (talk) 22:26, 18 January 2023 (UTC)Reply[reply]

HELP! I am a user with accessibility needs. On my desktop (my main way to view Wikipedia), the new skin is awful! I cannot easily click the silly, tiny hamburger menu button in the top left corner.

I shouldn't have to click on that tiny thing and THEN wait for a drop-down and THEN read through the drop-down for the link that I want (e.g. the "Permalink" link). With the old skin, I would just click the link that I wanted, in one step!

I don't want to have to have an account just to be able to read Wikipedia in a usable way. Please provide a way to use the old skin without having to sign in! (talk) 00:44, 19 January 2023 (UTC)Reply[reply]

If you click the hamburger one time without closing it afterhand the site should remember your choice and have it stay until you click the hide guillemet. Regarding choice for logged out users, it has been declined due to reasons listed at mw:Reading/Web/Desktop Improvements/Frequently asked questions § Why is the opt-out link not available for logged-out users? Aaron Liu (talk) 01:45, 19 January 2023 (UTC)Reply[reply]
"This is because of the limited capability of our servers. The logged-out users can use browser extensions allowing them to personalize their interface, or they can create an account." This is an utterly BS excuse and I think you know it. With the insane amount of money the wikimedia foundation brings in through its yearly begging campaigns, saying that they can't afford to direct some of that to having the server capability to let anyone switch to a different CSS style (perhaps by redirecting some of the funds away from being passed on to third party organizations that the original donators didn't even know about) is simply insulting.
2601:405:4400:9420:DC22:E380:D2EE:15A9 (talk) 04:23, 19 January 2023 (UTC)Reply[reply]
No it isn’t. Click on the link to the Why are there no preferences for anonymous users section provided in my link Aaron Liu (talk) 12:45, 19 January 2023 (UTC)Reply[reply]
@Aaron Liu: you said, If you click the hamburger one time without closing it afterhand the site should remember your choice and have it stay until you click the hide guillemet. Unfortunately, this does not work. If I click the hamburger once to open the drop-down, and then go to a different article (either by clicking a link in the article, or by clicking "Random article"), then the drop-down disappears.
Regarding choice for logged out users, it has been declined due to reasons listed at mw:Reading/Web/Desktop Improvements/Frequently asked questions § Why is the opt-out link not available for logged-out users? As 2601:405:4400:9420:DC22:E380:D2EE:15A9 pointed out above, the reason given there doesn't make any sense at all. Frankly, if the new skin (Vector 2022) is such a strain on the servers, then the WMF should have just stuck to Monobook to keep Wikipedia usable for everyone.
At Wikimania 2006, the WMF and the community reached consensus: Many Wikimedians live in situations (or parts of the world) where having a Wikipedia account is a liability. No-one should require an account in order for Wikipedia to be usable for them.
Rolling out Vector 2022 as the default is a thoughtless, ableist change to the interface and the WMF should be ashamed. (talk) 09:35, 19 January 2023 (UTC)Reply[reply]
The strain on servers isn’t from the new skin, it’s from saving IP preferences. You should also see the other section I gave to the IP above. Aaron Liu (talk) 12:47, 19 January 2023 (UTC)Reply[reply]
@Aaron Liu: I have now read that section. Either the WMF should find a fix for that issue, or should revert to Monobook instead of adding needless UI obstacles (as Vector 2022 does) to the experience of disabled users. (talk) 18:59, 19 January 2023 (UTC)Reply[reply]
The fact you guys store this preference server side is hilarious. Incompetence at its finest. (talk) 20:13, 20 January 2023 (UTC)Reply[reply]

Was this a beta feature?

Congratulations on Vector 2022, I like it! I think I was getting much of the appearance already because I had checked "Automatically enable most beta features" in Special:Preferences#mw-prefsection-betafeatures; I assume that preferences section had toggle(s) for the new appearance that are gone now that it's about to become the default. I don't know if this is worth mentioning on the page. -- Skierpage (talk) 06:32, 16 January 2023 (UTC)Reply[reply]

@Skierpage - thanks for your question! Yes, the preferences page allows you to switch the new skin on and off (as well as to choose any other available skin). It's located in the appearance section of the preferences page, under skins. OVasileva (WMF) (talk) 19:38, 16 January 2023 (UTC)Reply[reply]


I have not tried this new skin , but if it is anything like the change on Simple English Wikipedia I will find it difficult for a while. That appearance, makes editing and moving around on Wikipedia tedious. I could be wrong and the new Vector is brilliant. But my general feeling is that ”if it’s not broken why fix it”. We have a great design of the Wikipedia at present. --BabbaQ (talk) 08:40, 16 January 2023 (UTC)Reply[reply]

I tried out new Vector 2022 and didn't like it, nor any of the other skins. It looks and navigates too much like a mobile website. I don't use mobile devices except to maybe look something up when I'm on the go. I would never edit using the mobile interfaces. So I switched it back to Vector legacy. At least I feel confident now on how to access Preferences and switch it back after the auto-change (which hasn't happened yet, even though it's January 18 today). Grorp (talk) 01:32, 18 January 2023 (UTC)Reply[reply]
I don't get what your point is here, we still use the desktop interfaces to edit. Aaron Liu (talk) 13:11, 18 January 2023 (UTC)Reply[reply]

”if it’s not broken why fix it” Since Wikipedia isn't fueled by adverts, this drive towards new and more users, and subsequently the constant desire to dumb down the interface that is excellent for experienced Wikipedians is incomprehensible to me. Unfortunately it appears WMF is hiring engineers that constantly must dabble to justify their existence :-( CapnZapp (talk) 07:07, 18 January 2023 (UTC)Reply[reply]
I don't think the interface was dumbed down. They just moved a ton of things. The closest to dumbing down I can think of is moving a bunch of buttons under the user which does look better to me. Aaron Liu (talk) 13:19, 18 January 2023 (UTC)Reply[reply]
Yes it is great but I think there are a couple of grievances with the previous vector. Aaron Liu (talk) 13:12, 18 January 2023 (UTC)Reply[reply]
I don't like it at all, either. The notion that they are limiting width of content when most desktops use widescreen monitors is absolutely wild. This skin is a step back for sure. Melmann 16:33, 18 January 2023 (UTC)Reply[reply]
I agree. In what way is this better? The proportions of the various elements aren't well judged. And a bit more white space, fine, but why this much? There just isn't enough on the left in the navigation menu in 9 cases out of 10 to justify it. (talk) 18:11, 18 January 2023 (UTC)Reply[reply]
I'm glad people are saying this. I seriously doubt that after the obligatory week of adjustment that most people are going to suddenly prefer the version to the old. Bad call from Wikipedia. YouCanDoBetter (talk) 20:59, 18 January 2023 (UTC)Reply[reply]
It's not like the RFC about deployment went majority against the change and WMF ignored it- oh wait, that's precisely what happened. Taking part in Wikipedia community continues to feel more and more futile. Melmann 21:07, 18 January 2023 (UTC)Reply[reply]
No it isn’t. The opinion was split in half and summarized by two notable editors not involved with the foundation. I’ll quote it:

The most substantial concern, and the only clear blocker, was the issue of fixed-width. The idea of using a community-maintained gadget is deemed insufficient. It should be possible to achieve a full-width experience using a WMF-maintained toggle, which is clearly visible and available to both logged-out and logged-in users. There were also notable concerns about non-intuitive icons in the sticky header and the behaviour of the language selector, which we believe need to be addressed to achieve a firm consensus.
If all the concerns outlined above are satisfactorily addressed then we see community support to roll out the change, and in our view no further RfC would be required, although the Web team is free to hold one if they wish. Since the discussion was dominated by fixed-width we can’t say for sure if the other concerns are blockers. As such, our close aims to be conservative, given also the scale of the change and that each opposing editor likely did not list every concern which they agreed with if it was already raised by others.
— Wikipedia:Requests for comment/Deployment of Vector (2022) § Discussion, authored by ProcrastinatingReader and ScottishFinnishRadish

Aaron Liu (talk) 22:34, 18 January 2023 (UTC)Reply[reply]
They also did provide a preference and a button in the bottom right to disable limited width Aaron Liu (talk) 22:38, 18 January 2023 (UTC)Reply[reply]
So my first impressions on the new skin is that the wikipedia header image, and the three bar menu should move down the page with the sidebar to make it easy to access regardless of how far down on the page you are. Also, I think the light grey box used in that menu should be extended to the navigation menu as well, so that the white space is broken up rather than just feeling misaligned. A little bit of subtle cues goes a long way to making things readable. Lines are good. (talk) 21:07, 18 January 2023 (UTC)Reply[reply]
@ you might want to say that in the link at the header of this page Aaron Liu (talk) 22:35, 18 January 2023 (UTC)Reply[reply]

I agree with Melmann here. It feels like opposition to the change was uniformly written off as 'non-substantive' or whatever, and then this was pushed through. "Consensus-based"? Given that passionate opposition was ignore just 'cuz, I don't think so. (And this says nothing about the way it was rolled out...)--Gen. Quon[Talk] 16:44, 19 January 2023 (UTC)Reply[reply]

@BabbaQ @Grorp @Melmann and others: as a member of the WMF firstly I can assure that we never make changes for the sake of change, to follow trends, or to justify our existence. Making changes to the Wikipedia interface is extremely difficult technically and culturally. We are very far from a world where changes are made carelessly.
The fact of the matter is that we've done our homework here (years of it), and the large majority of editors we spoke to (over 1,900 in total), and the large majority of readers (hundreds), have given us overwhelmingly positive feedback. If you take the time to read through the various rounds of feedback we collected (1, 2, 3, 4, 5) you will see this for yourself. This includes feedback on line-length, and other seemingly controversial topics. On top of that the data is very clear: people are searching more, engaging with the table of contents more, and generally exploring knowledge more deeply. On top of that we are now meeting more WCAG requirements than ever before, including most importantly their requirement around line-length. There are of course people who don't like change, don't like the foundation, and don't like the new interface. And these people inevitably scream the loudest in all possible places. This can be misleading, because it makes it seem like everyone hates the changes, and then you conclude in your mind that we must be evil because we're ignoring all of these people. This is far from the truth. We make decisions based on research, data, accessibility standards, and feedback. If you take the time to see the full picture I am confident you will come to a different understanding here.
Also, regarding the RfC, to set the record straight:
  • A bunch of people who had not been engaged with the project showed up and expressed very harsh first impressions, mainly regarding the fixed-width (which is now configurable)
  • The RfC had a favorable closing, which is done by the community and is the way that the outcomes of RfCs are determined (not by counting votes)
  • Our team followed the instructions of the closing of the RfC
I hope this helps you all make a little bit more sense of this frustrating change and time. I know from past experience that it takes hundreds of exchanges with volunteers to start to build any kind of trust, so I know you won't trust me for a while. But I promise you that our team wants to make the website better for everyone, and wants to help the Wikipedia movement continue to grow. If we take the time to zoom out a little bit I think we'd all realize that we're on the same team here, working to make free knowledge accessible to as many people as possible.
Cheers, AHollender (WMF) (talk) 17:49, 19 January 2023 (UTC)Reply[reply]
I've been around to know that no conversation on Wikipedia goes anywhere without assuming good faith, so I will give you the benefit of trusting that you earnestly believe what you have written here, but those words do not communicate to me and other detractors in the community what you think they do.
This all feels to me like you saying "We've got our carefully selected data and carefully selected feedback that works for us, and supports what we want to do, and community's feedback is not relevant unless it serves to further the plan that we have laid out." You have a majority negative RFC and instead of going away, making changing and coming back to do another RFC to bring the community on board, you press on because you think that the community's opinion is wrong, and the community simply lacks the grand vision that being on WMF payroll bestows upon you. This was clear when the community had to pull you back from displaying false and misleading claims during the last donation drive, and it is evident here. Did you maybe consider that people who just showed up 'out of nowhere' to take part in the RFC may have first encountered the theme during the RFC or that RFC may be a venue where they felt they can engage with the WFM as RFC is one of the primary consensus building tools on Wikipedia? Or is any discussion that goes against WMF automatically subject to canvassing?
The way I see it, as WMF your core challenge, which will define the long-term success or failure of this project, is whether you can halt and reverse the constantly shrinking editor and admin pool. Editors are fleeing the project, and the last thing you need is a divisive change that ultimately achieves nothing, like this one. The fact that you (as in WMF) don't see that the pathway towards that goal starts with listening to community more, not less, is evidence of how far your reality is removed from editors' reality. Melmann 18:57, 19 January 2023 (UTC)Reply[reply]
you press on because you think that the community's opinion is wrong isn’t right, what happened was the people from this community that closed the rfc represented it as “if you change this this and this then you can change it to the new skin without an rfc”. The first bullet point does sound like "The opposers just weren’t accustomed" though. Aaron Liu (talk) 19:12, 19 January 2023 (UTC)Reply[reply]
@Aaron Liu ah, thanks for calling that out. Good point about my phrasing. I guess what I was trying to get at is that deep, constructive conversation doesn't happen when people shout at you about their first impressions of something. The unfortunate reality is that interfaces are complex, as is the whole wiki-ecosystem, so it takes a long time to build shared context about a project and the possible tradeoffs. I'm not claiming to know more than volunteers, just that I have spent the past 4 years learning about their workflows, and it seemed like people showed up to the RfC (without previously engaging in the project) and made snap judgements. So it didn't feel mutually respectful, or that we were similarly informed about each other's perspectives.
As you mentioned, we clearly listened to the people in the RfC. I guess my bigger point was: arguing about whitespace is in the end relatively trivial, compared to something like editor retention, pageviews, etc. AHollender (WMF) (talk) 19:43, 19 January 2023 (UTC)Reply[reply]
@Melmann "the constantly shrinking editor and admin pool" — I completely agree that this should be a top priority. I think we (WMF + Community) should have shared goals around top level metrics like editorship, pageviews, and other straightforward but important factors. Without that kind of common ground we end up spending a ton of energy arguing about white space and icons. I think it's unfortunate that the conversations here, and on the RfC haven't been focused on high-level goals like that.
To your point about new editors — our current theory is that by simplifying the interface we've created the opportunity to emphasize key entry points like Create account, Edit, History, or others. Our plan now is to run a bunch of experiments to figure out which entry points work best for newcomers to the site. AHollender (WMF) (talk) 19:39, 19 January 2023 (UTC)Reply[reply]
You didn't hear the opponent, again. It was you (WMF) who introduced the divisive change (despite vocal opposition) and now you reproach users of "spending a ton of energy arguing about white space and icons". Sad. 2dk (talk) 23:01, 19 January 2023 (UTC)Reply[reply]

Desktop/mobile/app mode

I get the impression that this skin is desktop browser only; is that right? But my understanding is that the majority of users use the mobile browser view, right? So what about that mobile interface – is there any change? Myself, I am increasingly aware of such mode changes as I switch from phone to Chromebook to Android/iPhone apps on both. And note that my Chromebooks have multiple monitor sizes including a swivel screen which will go portrait (HP Chromebase). Many people use multiple devices and so there ought to be some common approach to avoid confusion but my impression is that each interface is designed and developed separately. Anyway, this documentation should do more to make it clear which mode is being changed. Andrew🐉(talk) 20:56, 16 January 2023 (UTC)Reply[reply]

Hi @Andrew Davidson - yes, you are correct, the change is only to desktop browsers. Thanks for calling that out! I made it a bit clearer on the page. In terms of mobile - yes, the majority of the users are on the mobile site. While we didn't focus on mobile for this project, there are many other projects across the team as well as WMF product that work on mobile specifically. Over the last couple of years, the main focus there has been on editing on mobile, but we are planning on making some improvements to reading on mobile as well in the upcoming year. While building the skin, we also considered bringing it closer in visual design to the mobile site, so that people reading on mobile can still recognize Wikipedia in its desktop form as well. We also aimed to reduce code for skins overall so that it's easier in the future to build features and adapt them across both desktop and mobile skins. OVasileva (WMF) (talk) 23:40, 16 January 2023 (UTC)Reply[reply]
Hello. I sometimes use Wiki on mobile, but nearly always for reading only. The editing experience on mobile is atrocious. But I think the better approach is to just accept this - I don't see a complex system like this ever being able to be edited on mobile (unless we get reliable AI guided voice editing or some other science fiction concept). Focus instead on serving your core experienced desktop users. That the mobile experience is simplified and catering to the new influx of users is fine, since it is next to useless for us veterans anyway. Making the desktop experience more like mobile is most emphatically not fine. Please forget any ideas to make desktop grow your user base, and instead consider treating desktop as your "power user" interface. Thank you. CapnZapp (talk) 07:13, 18 January 2023 (UTC)Reply[reply]

Deployment notification?

I've just learned that Vector 2022 is being deployed as a default to all users in just a few hours. Apologies if I somehow missed it, but was there ever a sitewide notification to inform users of this major change? Not everyone watches the Village pumps, or subscribes to Tech News, or participated in the RfC, and I am certain there are those who are not even aware of this skin's existence. I feel like there should have been a push notification, or a talk page message, or maybe even an email sent out to all active users as a heads-up earlier this month or this week. Seems very abrupt in my opinion, many editors are likely unprepared by this change and will probably be taken aback when they log on to Wikipedia tomorrow. The Web team should brace themselves for pandemonium during the next 24 hours. InfiniteNexus (talk) 07:40, 18 January 2023 (UTC)Reply[reply]

Indeed. Strange that the RFCs establishing this change were not advertised on Watchlists, only the deployment. And the original November RFC seems to have more opposition than support, yet was nodded through anyway. I find the new UI dreadful, all that wasted white space, it's made French Wikipedia much less pleasant to visit.  — Amakuru (talk) 08:22, 18 January 2023 (UTC)Reply[reply]
You can change the width in the right lower corner, you can also change it permanently in your preferences. Coldbolt (talk) 10:05, 18 January 2023 (UTC)Reply[reply]
Hey @Amakuru, the RfC was advertised as a WLN, too. Then, we posted three updates in the Village Pump (in November, December, and January). SGrabarczuk (WMF) (talk) 15:02, 18 January 2023 (UTC)Reply[reply]
@SGrabarczuk (WMF): Oh, that's my bad then, I somehow completely missed it! Thanks for the response. Cheers  — Amakuru (talk) 16:07, 18 January 2023 (UTC)Reply[reply]
Yes. I noticed 'a banner' at the top of the page but completely ignored it since I have banners turned off. For the umpteenth time I logged in and saw this banner reappear, I finally read it... 24 hours before the scheduled changeover. I receive dozens of emails a day telling me about this or that notice on my watchlist, but not a single email about an interface change. I receive little red marker when someone uses my account name on a talk page... but not a single red indicator appeared at the top of my screen. The only notice was what looked like an advertising banner; colorful and easily 'tuned out'. Grorp (talk) 08:24, 18 January 2023 (UTC)Reply[reply]
Knew about it from Tech News that today was the day. I too have banners turned OFF. I don't remember seeing a maintenance or special banner (which cannot be turned OFF) though. I already had Vector legacy (2010) selected, and was surprised that I was affected. Not sure if I had to... but after resetting it a couple of times in Preferences/Appearance, as far as I can tell, all seems good once again. :) — WILDSTARTALK 16:31, 18 January 2023 (UTC)Reply[reply]
I had to reset my preferences to legacy three times (it was starting to get annoying) but it seems to have stabilized now. DB1729talk 16:42, 18 January 2023 (UTC)Reply[reply]

@SGrabarczuk (WMF): It's not too late, I strongly suggest the Web team send out talk page messages to all active/regular users acknowledging the change, providing guidance on how to switch back to the old skin, and including links to more info and to give feedback. InfiniteNexus (talk) 18:04, 18 January 2023 (UTC)Reply[reply]

Thanks @InfiniteNexus for your suggstion. We will discuss this. SGrabarczuk (WMF) (talk) 19:17, 18 January 2023 (UTC)Reply[reply]
@SGrabarczuk (WMF): Have you? And what's been decided? 2dk (talk) 23:53, 19 January 2023 (UTC)Reply[reply]

Main page

I assume the big change has now taken place, and was expecting to see it immediately when I fired up Wikipedia as a logged-out user. However, I noticed that the Main page, and indeed talk pages etc, still have the old skin, while it's just the articles themselves are switched. Is that intentional? Cheers  — Amakuru (talk) 16:19, 18 January 2023 (UTC)Reply[reply]

Heh I always thought admins would already know about any WMF/skin changes moreso than the rest of us. Nice to see your question here... yeah I noticed that the Vector 2022 skin seemed to be "live" on some pages and not on others...hope my Preferences choice holds throughout the complete changeover. Shearonink (talk) 16:25, 18 January 2023 (UTC)Reply[reply]
He he... being an admin gives me a few extra shiny buttons, but it doesn't come with a magic connection to the inner thoughts of the WMF Smiley.png. No doubt some admins and others do have their ears glued to the ground and also live their lives through Discord and IRC and wherever else, but personally I just find things out (or not) the same way everyone else does. Cheers  — Amakuru (talk) 16:28, 18 January 2023 (UTC)Reply[reply]
@Amakuru, @Shearonink - Thanks for your question. It will take around 24 hours for the change to propagate across the majority of pages. This means that certain pages will show the new skin, while other pages will show the old skin. This gradual release process protects our servers and ensures there is no risk for site performance. In terms of preference choices, the staged rollout we performed affected local preferences for about 1 hour. Now that the rollout is complete, preferences are now stable. Apologies for any disruptions! OVasileva (WMF) (talk) 17:49, 18 January 2023 (UTC)Reply[reply]

Will The Old Skin Ever Be Removed

I knew the original skin got removed, later replaced by vector, so will the Vector Legacy ever be removed? Happy Editing! -I Followed The Username Policy (talk) 16:37, 18 January 2023 (UTC)Reply[reply]

oh gawd I sure hope not... Shearonink (talk) 16:40, 18 January 2023 (UTC)Reply[reply]
Right? Are they planning on improving the new skin further because dear lord it sucks... why did they roll it out half-baked? VQuakr (talk) 17:16, 18 January 2023 (UTC)Reply[reply]
Hi @I followed The Username Policy, @Shearonink, @VQuakr - we will not be removing the Vector Legacy or any other existing skins and will continue maintenance and bug-fixing of existing skins into the future. OVasileva (WMF) (talk) 17:45, 18 January 2023 (UTC)Reply[reply]
@OVasileva (WMF): I see a lot of repetitive questions coming up on this. Is there a tersely-written FAQ somewhere? VQuakr (talk) 17:47, 18 January 2023 (UTC)Reply[reply]
Hi @VQuakr - yes, an FAQ is available on the project page on mediawiki here and there's some info relevant to English Wikipedia specifically on this project page. OVasileva (WMF) (talk) 17:58, 18 January 2023 (UTC)Reply[reply]
Thank you for this assurance. I was very surprised that the new skin was given a name which duplicates one already in use, and concerned that this presentation might herald an attempt to deprecate the existing Vector skin. I'm pleased to read that the skin which many prefer will remain supported. Certes (talk) 21:29, 18 January 2023 (UTC)Reply[reply]

New Skin - poor implementation and communication

I was quite surprised this morning to see this new skin appear. I don't know how this was supposed to be communicated, but I saw nothing until after the change. Which seems odd to me, given that it's hard to miss the donation requests.

Also, there's been no communication to undo this. It showed some kind of return to old skin (can't remember the phrasing), but when you press it, it takes you too a dialogue, rather than switching to the old skin. In the dialogue, there's no indication which of the multiple skins available WAS the old skin! Finally, when you realize that the old one was probably Vector 2012, and you select it, it again doesn't stick, and you have to realize that you need to page way, way down to a save button.

Why isn't the save button beside the change? Why isn't there an indication of which WAS the old skin? Why not just have the button to change back without going to the dialogue? And why not have some kind of banner alerting people to the change, and to a clear simple FAQ?

I'd suggest rolling back this poorly implemented change until communication and implementation are improved! Nfitz (talk) 16:46, 18 January 2023 (UTC)Reply[reply]

  • I'm not at all happy with these changes. Until they have been fully tested, I also think we should return to the old presentation.--Ipigott (talk) 16:52, 18 January 2023 (UTC)Reply[reply]
Yeah. What's the point of enforcing this new skin? I can't recall anyone ever complaining about the old skin. The new one feels half-baked and claustrophobic. Almost as bad as using the mobile site on desktop. Elli (talk | contribs) 17:02, 18 January 2023 (UTC)Reply[reply]
There is a banner and there have been many updates up to this point (and an RFC).
Is it possible you have opted out for communication via the gadget "Suppress display of all CentralNotices (To suppress only certain classes of notices use the Banners option in preferences)" on Special:Preferences#mw-prefsection-gadgets ? Jdlrobson (talk) 19:55, 18 January 2023 (UTC)Reply[reply]

How to turn off the new skin*(Found at WP:Vector 2022 in the How to turn off the new skin subsection)
You can turn off the Vector 2022 skin in two ways:

  1. From the left menu (sidebar), select the link “Switch to old look”, then select your preferred alternate skin (note: the previous default skin was Vector legacy)
  2. Open the user menu from the button at the top right corner of the page, then select preferences. Go to the appearance tab on the preferences page and scroll to the Skin section on the preferences page

Annnnnd... -->>>HIT the SAVE BUTTON AT THE BOTTOM OF THE PAGE. Shearonink (talk) 17:12, 18 January 2023 (UTC)Reply[reply]

PS - I added that^^^ extra line. Shearonink (talk) 17:12, 18 January 2023 (UTC)Reply[reply]
Sure, if one can find the FAQ. With no communication to ordinary users, there's no indication of a FAQ. There isn't even a link to it at preferences! The management of this transition (which I'm not convinced is even necessary) is beyond dreadful. Where did we approve this transition, BTW. I know there were certain improvements and preconditions that had been set before this happens ... and it appears that they haven't been met yet! Nfitz (talk) 19:10, 18 January 2023 (UTC)Reply[reply]
Hi @Nfitz - thanks for your feedback. If you're interested in learning more about the process and communication here on English Wikipedia, we've written up a summary on the project page here. On this page, you'll also find some instructions on the different ways you can opt-out of the skin. We are currently running banners to all logged-in users that link to this page so that everyone has a chance to see the opt-out instructure. Thanks @Shearonink for adding the clarification about the save button! OVasileva (WMF) (talk) 19:19, 18 January 2023 (UTC)Reply[reply]
We need to put this crucial information somewhere that everyone will see it, preferably in flashing bold 96-point capitals. I would even go as far as a banner on the Main Page, though that's mainly aimed at readers who can't easily opt out (and must be banging their heads against their screens in millions trying to do so). A watchlist notice explaining concisely how to continue enjoying the last 13 years' experience is the least we should do. Certes (talk) 21:35, 18 January 2023 (UTC)Reply[reply]
Somewhere at least. User:Shearonink says there are banners running - and yet I've not seen any. If only I had an extra account to test if my change of skin nullified the banners before I had a chance to see them. Nfitz (talk) 23:34, 18 January 2023 (UTC)Reply[reply]
Ohhh, I don't think I mentioned banners did I? Perhaps you're referring to Jdlrobson or OVasileva (WMF)... Shearonink (talk) 23:42, 18 January 2023 (UTC)Reply[reply]
No you didn't - looking carefully, following the mention of the words were "Thanks @[[User:Shearonink ...] and in my haste reading wiki-markup I assume that was your comment!
Hey @Nfitz - you're correct. The current banners are only appearing for people using the new skin, both logged-in and logged-out. The banners that announced the timing of the change and linked to the project page here with more information were showed to all logged-in users, regardless of skin preference up until we started deploying the change earlier today. OVasileva (WMF) (talk) 00:09, 19 January 2023 (UTC)Reply[reply]

Use 2010 Vector as an IP

Is there a way to restore the 2010 vector while logged out? I do not log in on my work computer, but do visit the site for information during the day here, and this is not something I want to use. (talk) 16:59, 18 January 2023 (UTC)Reply[reply]

Short of adding ?useskin=vector to the end of the url for every page you visit or logging in to an account (Alt or main) there is no way for logged out users to see the old skin. Terasail[✉️] 17:02, 18 January 2023 (UTC)Reply[reply]
You could use an extension like this for Chrome/Brave/Edge to redirect to include the ?useskin=vector in the URL. These settings shown here work for me (two rules, one for regular /wiki/ URLs and one for editing /w/ URLS). Enjoy. I'm sure Firefox will an equivalent extension. -- (talk) 17:13, 18 January 2023 (UTC)Reply[reply]
Thank you so much for this! There is one for Firefox that works for me with the settings you gave: [1] (talk) 21:01, 18 January 2023 (UTC)Reply[reply]
And here's a Firefox version of the Redirector addon. Just for ease of copy-pasting, here are my 3 rules, in this order:
1. Keep the useskin parameter, if one is present:
* Include pattern:[^#?]*)(\\?[^#]*)useskin=([^#]*)(#.*)?$
* Redirect to:$1/$2$3useskin=$4$5
* Pattern type: Regular Expression
2. Add the useskin parameter, if there are other parameters:
* Include pattern:[^#?]*)(\\?[^#]*)(#.*)?$
* Redirect to:$1/$2$3&useskin=vector$4
* Pattern type: Regular Expression
3. Add the useskin parameter, if there are no other parameters:
* Include pattern:[^#?]*)(#[^?]*)?$
* Redirect to:$1/$2?useskin=vector$3
* Pattern type: Regular Expression
You can type various URLs in the Example URL field to test these patterns, it doesn't have any functional effect of the addon behavior. 2A02:2168:84D9:2200:D6CB:69C2:6A20:23C2 (talk) 22:46, 18 January 2023 (UTC)Reply[reply]
This does work. The only annoying thing is that if I want to copy a URL and say use it to create a link in a document, I generally would have to hand-edit the URL to now *remove* the extra `&useskin=vector` parameter, unless I now want all such links to include it for all time. Eblon2 (talk) 22:40, 20 January 2023 (UTC)Reply[reply]
You may also use a bookmarklet! :) Copy that code and just replace "monobook" with "vector". SGrabarczuk (WMF) (talk) 19:10, 18 January 2023 (UTC)Reply[reply]
We're working on that here. I expect add-ons for the popular web browsers will appear within days once the world wakes up to this change. Certes (talk) 21:37, 18 January 2023 (UTC)Reply[reply]
Although designed for Japanese Wikipedia, this script works for me in Tampermonkey with @match modified to* Greasy Fork is open source (GPLv3 licensed). Certes (talk) 21:53, 18 January 2023 (UTC)Reply[reply]
@Certes Thank you for adding this. Patafisik (WMF) (talk) 12:21, 19 January 2023 (UTC)Reply[reply]
+1. Everybody's efforts and suggestions on this issue are appreciated. (talk) 14:18, 19 January 2023 (UTC)Reply[reply]

Mystery Meat

I gave feedback some months ago about the use of unlabelled icons in the top toolbar - as usual, I was totally ignored. Now you've rolled out the new skin with its usability issues still in place. "Mystery meat" - using icons without labels - is considered poor interface design - there's even a Wikipedia article on this, so you can't claim ignorance. It has the potential to confuse the user if they don't happen to understand your chosen hieroglyphs. Either add labels or allow the user the option to activate labels in their preferences. Cnbrb (talk) 17:03, 18 January 2023 (UTC)Reply[reply]

I looked all over for my contributions until I found they were hidden behind an icon. And I am fully sighted.--Ipigott (talk) 17:22, 18 January 2023 (UTC)Reply[reply]
Oh that was it!! I couldnt find the preferences button easily, the button for the page you need to access to remove this mode can't be easily found on this mode. The concept of using logos over text isn't that great. Franfran2424 (talk) 17:34, 18 January 2023 (UTC)Reply[reply]
Hey @Cnbrb I don't believe your comments were ignored (I remember discussing this topic with you). I've just written another response to this topic above, so I'll link there instead or re-posting: Wikipedia talk:Vector 2022#Keep the old buttons. AHollender (WMF) (talk) 17:53, 19 January 2023 (UTC)Reply[reply]
"We believe the icons are sufficient"... " Having buttons with text labels [...] to us felt heavy and crowded". Says it all really. Feedback ignored. Bad job, baffling design, meaningless icons. I've switched back to the old design - not using this usabliity disaster. Cnbrb (talk) 18:40, 19 January 2023 (UTC)Reply[reply]
Well I think it's more like: we considered the feedback, ran some tests, and made an informed decision. But at the end of the day you and I aren't really having a conversation, are we? You're feeling upset and ignored, and I fear that no reasoning, data, or explanation will suffice. I am honestly sorry to have caused you frustration, it's not a good feeling. If I could make everyone happy all at once I would be a very incredible person : ) But for now all I can do is follow the data, the feedback of the majority of people, and the intuition of the design professionals. AHollender (WMF) (talk) 19:54, 19 January 2023 (UTC)Reply[reply]
@AHollender (WMF) I agree that you can't please everyone, nor ever will. I'm beginning to get used to the new layout and don't have a problem with it. But I do see an inconsistency in that virtually all the main links on a page have Alt-text that appears when you mouseover them. Except, that is, for the key user icon, plus the link to the Mentor dashboard when it's clicked and the dropdown menu appears. All the rest of the sub-menu links have alt-text, so that seemed odd to me. Oh, and the languages link in article pages doesn't show any alt-text either on hover.
None of this bothers me personally, but it does seem inconsistent. And from a UX perspective, I do genuinely feel the need to question how these omissions will impact users with visual impairments? Some will be reliant on alt-text and screen readers to announce what an image or a button/icon is for. Surely the user icon is now an absolutely key navigational route to other pages, and should it not have alt-text on mouse hover like the rest of them?
Huge thanks to everyone for their hard work on this. Nick Moyes (talk) 22:12, 19 January 2023 (UTC)Reply[reply]
Hey @Nick Moyes, thanks for pointing out those shortcomings. We've been trying to update all of the buttons in the interface to make sure they have the appropriate alt-text and titles. I've added those to our list. Cheers, AHollender (WMF) (talk) 22:22, 19 January 2023 (UTC)Reply[reply]
@AHollender (WMF) You're welcome. Is there a published list of those, or shall we just ping you when we spot them? I could add there's also alt-text needed for the 'Preview' button when editing with Source Editor, as well as its associated 'Okay, got it' buttons for first time use of this and for the Cite and Links buttons in WP:VE.
BTW: Can you point me to a Commons Category containing all the new images and arrangement of icons and menu layouts? We're going to need these to update all the 'How to Guides'. It took ages to update everything here when all we had to worry about was the switch from 'Save changes' to 'Publish changes'. This is going to be a whole lot more complicated than that, I suspect. Thanks, Nick Moyes (talk) 23:40, 19 January 2023 (UTC)Reply[reply]

I actually like the new skin

As someone who isn't used to change and doesn't like it, I like this change. I am not going back. I'm already used to this format via the French Wikipedia. Electos242 (talk) 17:16, 18 January 2023 (UTC)Reply[reply]

  • Chacun à son goût. I've coped with this interface for some time on the French wiki and have seen it slowly creeping in on the Scandinavian languages too but for the type of editing I do every day, I find the 2010 interface much more efficient. That's not to say the 2022 interface could not be improved on the basis of editor reactions.--Ipigott (talk) 17:28, 18 January 2023 (UTC)Reply[reply]
Same here, I would have preferred a bit more comunication, but the new skin makes editing so much easier Vincent-vst🚀 (talk) 11:46, 19 January 2023 (UTC)Reply[reply]
Specify what's easier.  Card Zero  (talk) 02:21, 21 January 2023 (UTC)Reply[reply]

Same here. I, of course, saw the notices about the RFC and implementation decision, but mine actually changed early (a week or two or three ago?), and I've been quite pleased with it. Based on the research-backed rationales, it seems like a no-brainer. — Fourthords | =Λ= | 17:25, 18 January 2023 (UTC)Reply[reply]

@Electos242, Ipigott, and Fourthords: Thank you for your feedback!--Patafisik (WMF) (talk) 19:22, 18 January 2023 (UTC)Reply[reply]
I like it too (once I realised where the watchlist link was), I have edited here for 16 years, with more than 200,000 edits and spend about 7 hours a day here, the new look is bright and spacious, well done. Theroadislong (talk) 12:34, 20 January 2023 (UTC)Reply[reply]
But why did you need to search? "Watchlist" is pretty intuitive, the upside down flag of Togo means absolutely nothing (apart from to Togo of course). That was one of the key reasons I reverted. Trying to memorise obscure icons isn't something that comes easily to me, I far prefer English words. Martin of Sheffield (talk) 12:41, 20 January 2023 (UTC)Reply[reply]


Putting it simply, the previous version had a greyish background whereas the new-and-unexpected one is VERY BRIGHT. I've reverted my interface for the moment until I can track which are the right css defs to make it darker. --AlisonW (talk) 18:02, 18 January 2023 (UTC)Reply[reply]

Hi @AlisonW, thank you for your feedback, we tested the prototype for visual design with users, for further information please read this page. Actually we are not working on the dark mode but a lot of users asked it, see this section of our FAQ. Patafisik (WMF) (talk) 19:07, 18 January 2023 (UTC)Reply[reply]
Agreed. I found the greys pleasing, and the "BLASTING WHITE AT YOUR EYEBALLS ALL THE TIME" look of Vector 2022 actively unpleasant. --Kizor 22:43, 18 January 2023 (UTC)Reply[reply]
Hey @AlisonW — the easiest way to do this would be adding whatever background color you want to .mw-page-container, and then adding a white background as well as some padding to main#content. That will at least get you most of the way there. Let me know if you'd like more help in getting that setup.
.mw-page-container { background-color: #f8f9fa; }
main#content { background-color: #fff; padding: 20px; } AHollender (WMF) (talk) 17:57, 19 January 2023 (UTC)Reply[reply]

More whitespace than content

I tried the main page while logged out on my laptop. I have a 1080p screen, and I just measured with a ruler: the whitespace portion of the page was 18cm wide, while the actual main page content was only 16cm wide. Is it intentional that this layout shows more whitespace than content on the screen? Clay (talk) 18:05, 18 January 2023 (UTC)Reply[reply]

Hi @Clay, thank you for your feedback, you can personalize your experience and use the full width. Please read our FAQ to know why the width of the content is limited. Patafisik (WMF) (talk) 18:31, 18 January 2023 (UTC)Reply[reply]
Hi @Patafisik (WMF): 1. None of the reference websites listed in that FAQ have a ratio of less than 1 when it comes to width of content vs. width of whitespace, as is the case with this new layout. 2. You can't personalize very easily when you're logged out. 3. This is the first impression that people get of Wikipedia now, just a tiny sliver of text with massive walls of whitespace surrounding it.
I don't want to beat a dead horse, since I have seen the discussions elsewhere about this issue, but it just plain looks bad in this aspect. Clay (talk) 18:55, 18 January 2023 (UTC)Reply[reply]
Hi @Patafisik (WMF): I tried the new skin after realizing that the "whiteout" look was not due to my browser acting up (I use Wikipedia daily and didn't know about this rollout). I was dumbfounded by the choice to make the site look like I was browsing from a mobile so I started personalizing my experience since I'm using a desktop PC. After a while I realized that I was removing all the new features and simply switched back to the old skin. None of the explanations in the FAQ look convincing ("reducing eye strain", really?) Rizzardi (talk) 08:32, 19 January 2023 (UTC)Reply[reply]
Article tools will move soon at the right
Hi @Rizzardi: thank you for your feedback. I respect your choice and I don't want to persuade you to use Vector 2022, however I would like to give you some links: about the limited content width you can find useful references in the page dedicated to the feature and in our Repository (e.g. Computer text line lengths affect reading and learning by Peter Orton, Ph.D. IBM Center for Advanced Learning).--Patafisik (WMF) (talk) 12:52, 19 January 2023 (UTC)Reply[reply]
Hi @Patafisik (WMF): The article by Peter Orton has been written in 2007; it contains the line "Isn’t reading text on a low resolution computer monitor difficult enough?" that shows that the research is somewhat dated. The articles listed in the Research section of the MediaWiki have been published in the following years: 2005, 2002, 2001, 2013, 2002, 2003, 2004 (the most recent, written ten years ago, addresses the problem of line length in the context of dyslexia). Modern monitors have exceptional resolutions, it seems strange to base one's design choices on decades old research. Rizzardi (talk) 14:15, 19 January 2023 (UTC)Reply[reply]
Is there more recent research saying something different? – Joe (talk) 14:36, 19 January 2023 (UTC)Reply[reply]
@Clay: 1. Whitespace at the right will be soon occupied by the Article tools (image). 2. look at this discussion below, there are some useful suggestions for IPs. For the "reason why" see here.--Patafisik (WMF) (talk) 12:52, 19 January 2023 (UTC)Reply[reply]

Yeah, this looks god awful. I have been using and editing Wikipedia for a very long time but this makes it repulsive. --Tumbledee (talk) 17:38, 20 January 2023 (UTC)Reply[reply]

Im positive

I think the skin is as good as it is as logged in users can go back to the old skin. For readers it's much better and I read lots of people that were actually positive. People just need to get used to it. Coldbolt (talk) 18:05, 18 January 2023 (UTC)Reply[reply]

Thank you @Coldbolt for your feedback! Patafisik (WMF) (talk) 18:43, 18 January 2023 (UTC)Reply[reply]
Are there a lot of non-readers using Wikipedia? That doesn't make sense to me intuitively. Nfitz (talk) 19:12, 18 January 2023 (UTC)Reply[reply]
@Nfitz: I don't know precise data about non-readers using Wkipedia, but you can find stats of Wikipedia here, or asking in the Village pump. Hope this should help.--Patafisik (WMF) (talk) 19:33, 18 January 2023 (UTC)Reply[reply]
That seems to not have any indication of user literacy. Nfitz (talk) 23:45, 18 January 2023 (UTC)Reply[reply]

This skin is horrific

If it ain't broke, don't fix it. The new skin is simply poor work compared to the old default. The lack of a table of contents causes infoboxes to push images down (and sandwiching issues also arise), and the TOC on the left squeezes everything together. It's horrible. Who came up with this crap? ~ HAL333 18:35, 18 January 2023 (UTC)Reply[reply]

At the very least, you really should give IPs the choice to use the old skin. ~ HAL333 18:36, 18 January 2023 (UTC)Reply[reply]
Hi @HAL333:, thank you for your feedback. Please discover the story behind Wikipedia’s updated interface and read this section of our FAQ.--Patafisik (WMF) (talk) 18:49, 18 January 2023 (UTC)Reply[reply]
There's nothing in that story, User:Patafisik (WMF) to indicate when we approved the transition on the English Wikipedia. Can you point to that please? Also there's no explanation on why regular users weren't notified, nor why there's no clear link to the FAQ, nor why you can't simply switch back with one click. The information all seems to be about the skin - not the very unacceptable way it's been communicated and implemented! Nfitz (talk) 19:15, 18 January 2023 (UTC)Reply[reply]
Hey @Nfitz, the link to the request for comment is used twice; it appears in the third paragraph, in the lead section. You'll find more links to the relevant discussions in the section "A very brief timeline". We were consulting a number of experienced and trusted Wikipedians who are experts both in the policies and technical things, incl. ArbCom members, and they helped us announce the RfC by the book. So for example, it was announced in the Watchlist notice. But before that, for example when we were asking random Wikipedians for feedback (that happened five times), we were running CentralNotice banners. SGrabarczuk (WMF) (talk) 19:37, 18 January 2023 (UTC)Reply[reply]
So glad 'trusted' Wikipedians were consulted because the community at large clearly cannot be trusted to make the right decision. Remember folks, WP:TINC. Melmann 21:13, 18 January 2023 (UTC)Reply[reply]
Hi, I'm not sure I expressed myself clearly - I was merely referring to the way how the RfC was announced. We were trying to do everything about the RfC in accordance with the policies and good practices. SGrabarczuk (WMF) (talk) 21:30, 18 January 2023 (UTC)Reply[reply]
If you're going to link the reasons supporting the changes, maybe we should also link Wikipedia:Requests for comment/Deployment of Vector (2022) where most of the editors responding, opposed this change for various reasons. DB1729talk 19:19, 18 January 2023 (UTC)Reply[reply]
@Nfitz and DB1729: Yes, thank you for the link to the RfC. Please read this answer of Alex too and to the rest of that discussion.--Patafisik (WMF) (talk) 19:27, 18 January 2023 (UTC)Reply[reply]
@Patafisik (WMF): Yours links to a blank page.[2] DB1729talk 19:34, 18 January 2023 (UTC)Reply[reply]
Fixed. This answer should also help.--Patafisik (WMF) (talk) 19:36, 18 January 2023 (UTC)Reply[reply]
WTAF -User:Patafisik (WMF) - that's the RFC that authorized this? - WP:Requests for comment/Deployment of Vector (2022)? There's more opposing than supporting, and some very clear discussion on improvements that are necessary before implementing. If there's not more somewhere, can I propose that this be reverted and those that made the change be blocked from Wikipedia? Nfitz (talk) 23:50, 18 January 2023 (UTC)Reply[reply]
@Nfitz: That was it. There's even boilerplate at the top that they wouldn't implement this nonsense if it didn't get community support. — LlywelynII 08:36, 19 January 2023 (UTC)Reply[reply]
@Patafisik (WMF): I would love working with you guys so of course nothing but appreciation for your intentions and attempted work. If that story in any way reflects the team's thought processes and rationale, though, no, you guys were inside a bubble and not thinking about how any of this works in practice. It's also silly that you went out of your way to document "only" one-fifth changed in the pilot projects when it's abundantly clear that this system hides any way to change. One-fifth shows absolutely overwhelming opposition if it was this hard for them to find how to make the switch in the first place. Where's the honest story about how this improves the backend? since the UX is only screwed over here... — LlywelynII 08:36, 19 January 2023 (UTC)Reply[reply]
I completely agree with HAL. This design defies all modern web standards, and goes back to the really bad "best viewed in 800x600" times.
Responsive design means to adapt to the user's screen size, and not enforce a certain size on the user's device. Because you simply don't know what device the user is using, and whether or not the chosen margins look good. It might look nice on the designer's screen in a specific resolution, but now imagine an ultra-wide 4k monitor that now consist of 90% blank white space and suddenly the whole thing looks horrific. TwoThe (talk) 12:57, 20 January 2023 (UTC)Reply[reply]
Hi, Olga has posted an update to the technical Village Pump with some information about the deployment, responses to feedback we've received so far, and some upcoming changes we will be making to the skin. We encourage you to check it out and leave any comments or questions. Thank you! Patafisik (WMF) (talk) 15:02, 20 January 2023 (UTC)Reply[reply]

Any developer involved in this redesign should be absolutely ashamed of themselves

How do you go to your job every day and do it wrong? Absolutely criminal. You should start refunding donations out of your outrageously inflated salaries. And I know your salaries are inflated because the value of the product you've produced is zero ($0) meaning any money you earned for it is tantamount to theft.

I highly encourage any of you to attend confession and seek absolution for the crime you have committed. Reservationsatdorsia (talk) 19:07, 18 January 2023 (UTC)Reply[reply]

Hi @Reservationsatdorsia - thanks for your feedback. We understand that at a first glance it might be difficult to see the impact of this change. If you're curious, we have compiled some of our findings on the benefits of the skin for readers and editors on this page. We also encourage you to check out our project documentation for more information on individual features, the research and conversations the team has had with the community, and our overall goals for the project. OVasileva (WMF) (talk) 19:22, 18 January 2023 (UTC)Reply[reply]
Reservationsatdorsia You are certainly entitled to disagree with what has been done here, but you owe these people an apology for your offensive comments. Unless you get hired by the Wikimedia Foundation, you don't have any right to insult them as to how they do their jobs. Disagree, even be angry, sure. But they are just people like you. 331dot (talk) 19:36, 18 January 2023 (UTC)Reply[reply]
The idea that you can't criticize someone unless you're their cohort is frankly uneducated. What you're suggesting is akin to defending food staff not washing their hands because you don't work at a restaurant. It's an objectively bad decision to force this change on the users, no one has asked for this. They can't even deliver dark mode without requiring a log in, it's pure incompetence. BadChange (talk) 20:12, 18 January 2023 (UTC)Reply[reply]
BadChange That isn't what I said at all. People can disagree and even be angry- but these "developers" are people too. It's their website and they can design it as they wish. This has been worked on and tested for years with input. I'm sure it's not perfect, but there are 7 billion humans on this planet all with differing ways of using the site. There are technical limitations to what you propose. 331dot (talk) 20:15, 18 January 2023 (UTC)Reply[reply]
"Re "no one asked for this"- have you surveyed all 7 billion humans on this planet to know that's true? What you mean is you didn't ask for it. That's fine- use the old skin. But you have no way of knowing if that is true or not. I don't think the site needs to stay the same for all time to satisfy you. 331dot (talk) 20:17, 18 January 2023 (UTC)Reply[reply]
I'm willing to bet most people are going to be upset by this redesign, regardless if its "objectivly good" or not, most people don't like change, especially not on stuff they're used to being the same for so long. (Just look at how angry people get at each YouTube redesign).★Trekker (talk) 20:21, 18 January 2023 (UTC)Reply[reply]
> have you surveyed all 7 billion humans on this planet to know that's true?
Have you? As someone who actually imposes a change on their users, it is the developers' job to ensure the change is welcome by the users, not the other way around. And I don't remember any polls or announcements or even beta tests on Wikipedia prior to this change. (And no, I don't follow social media, in case if there was something there.)
Look, nothing is perfect for everyone, this is obvious. But the new design has glaring objective functional flaws, such as content being not viewable, and serious inconveniences for unregistered users, which are the absolute majority of users, I'm sure. If you can't provide a decent experience to most of your users then just do not apply this change, plain and simple. But frankly, I don't see the problem in having the preferences saved in a cookie or something. 2A02:2168:84D9:2200:D6CB:69C2:6A20:23C2 (talk) 22:23, 18 January 2023 (UTC)Reply[reply]
You can certainly critizise, but you shouldn't say they ought to feel ashamed or are criminals just because you don't like their design. I'm not a fan of these changes either but your insults are uncalled for.★Trekker (talk) 20:21, 18 January 2023 (UTC)Reply[reply]
I owe them nothing. It's not my problem that they have chosen to be terrible at their craft. If the wikimedia foundation hired me tomorrow, I would revert all changes, fire anyone associated with this debacle, and then resign to give ted talks about how now to ruin your core product. Reservationsatdorsia (talk) 20:22, 18 January 2023 (UTC)Reply[reply]
Well thankfully they're not going to lose their jobs over you disliking this.★Trekker (talk) 20:25, 18 January 2023 (UTC)Reply[reply]
They should do the honorable thing and resign. Web development isn't for everyone. Perhaps they could try their hand at digging ditches since they obviously love huge swathes of open space. Reservationsatdorsia (talk) 20:32, 18 January 2023 (UTC)Reply[reply]
I don't agree that would be an "honorable thing". Maybe you should consider giving constructive criticism instead to actually improve the situation.★Trekker (talk) 20:34, 18 January 2023 (UTC)Reply[reply]
Reservationsatdorsia Instead of saying how the developers are terrible human beings for trying to make improvements with community input as they see them, please offer specific criticisms in a civil manner since you seem to know more than they do. If you would care to read more about it, they have reasons for the open space. 331dot (talk) 20:41, 18 January 2023 (UTC)Reply[reply]
  1. revert to 2010 vector
  2. stop trying to make desktop sites into mobile apps
  3. keep drawing your enormous salaries in obscurity.
Reservationsatdorsia (talk) 21:00, 18 January 2023 (UTC)Reply[reply]
Your projection documentation is a joke. You threw out nearly 65% of the responses to the sentiment survey because they used "foul language". Get this: if your re-design was bad enough to make the average wikipedia start sending you curses, that means your re-design wasn't very good. Reservationsatdorsia (talk) 21:15, 18 January 2023 (UTC)Reply[reply]
Nicely put. ~~ AirshipJungleman29 (talk) 15:06, 19 January 2023 (UTC)Reply[reply]
@OVasileva (WMF): I know I'd love to have a job working for you guys so of course thanks for your dedication and attempted work. As noted on that page's talk page, though, the "data" it employs is apparently entirely specious, given that one-fifth of users changing when you completely hide the ability to change speaks to overwhelming negative response, not any kind of general support. — LlywelynII 08:28, 19 January 2023 (UTC)Reply[reply]
At the end of the day I worry that these conversations are not only unproductive, but also end up leaving all parties involved feeling badly. Without any kind of relationship and trust, it seems difficult (impossible?) to have constructive discussions about product & design. You all are suspicious of us, which is fine, so any information we give you, or any attempt we make to explain our process, is inadequate. Or worse, somehow further incriminates us in your eyes.
I know from past exchanges with volunteers it takes hundreds of messages back and forth to build any amount of trust, so I know that won't come soon here since I haven't spoken to any of you before. But that's okay. If anyone here is genuinely interested in digging into the details, learning why we made certain decisions, and getting to constructive conversation, I am very much available for that. If you just needed to vent your frustration, that's totally fine to.
Cheers, AHollender (WMF) (talk) 18:49, 19 January 2023 (UTC)Reply[reply]
I feel like all of my comments have been highly constructive. Frankly, I should send you a venmo request so you can compensate me for my consulting. If you prefer to pay in cash or check, DM me for payment instructions. Reservationsatdorsia (talk) 19:40, 19 January 2023 (UTC)Reply[reply]
To be honest, I found them antagonistic and quite inappropriate. If you can't communicate without resorting to silly hyperbole and insults, then you're likely just to be ignored. I have left a polite reminder of this on your talk page. Nick Moyes (talk) 22:32, 19 January 2023 (UTC)Reply[reply]
If you can't communicate without being a high-handed, self-righteous jackass, you're likely to just be ignored. Reservationsatdorsia (talk) 23:33, 19 January 2023 (UTC)Reply[reply]
It's one thing to express feedback about the design itself – I've had some myself – but verbal abuse directed at people is unwarranted, unacceptable and undignified. And the fact you're using a burner account to do it suggests you know that. This section should be removed as a breach of WP:TPG. Ibadibam (talk) 18:41, 19 January 2023 (UTC)Reply[reply]
My dream was to never have an account at all. I had to make one in order to get back to the good design. The people who made this godawful design are highly paid and highly educated professionals. They don't need you to defend them. They need to own up to the fact that they are a bunch of slaves to the trend of mobile designing everything. They need to admit they made a huge mistake and revert it. They won't do it because they're a bunch of cowards, though. Reservationsatdorsia (talk) 21:43, 19 January 2023 (UTC)Reply[reply]
I think there are a number of things we could have done to make the transition easier on people. I'm sorry that it's causing you so much stress. As far as the design itself goes, again, if you become interested in some point about having a conversation about it I'd be happy to. Until then, I am also happy to be a punching bag for your frustration : ) It's more or less part of my job as the lead designer on this project. AHollender (WMF) (talk) 22:28, 19 January 2023 (UTC)Reply[reply]
Nobody wants a punching bag. They want answers and solutions. Rollback to vector 2010. An easy, clickable button without needing a login to switch back to 2010. etc. Just admit that your design was bad and move on. Not everyone is cut out for design, but the post office is always hiring. Reservationsatdorsia (talk) 23:38, 19 January 2023 (UTC)Reply[reply]
AHollender (WMF). Sir, you are obviously a galant person. And Members of the WMF staff. Noone wants to insult you. We understand your difficult position: you must follow your orders. But we seek your courage! Carry the message of this subjective matter (as this comment here) of courtesy towards unlogged, or when-unlogged readers, so much overlooked in the years of preparation of this switch. Stand up for us! At least, give us a little visible button on top of pages for switching to Classic Style. Is it so impossible? Thank you, Sir. 00:19, 20 January 2023 (UTC)

Infobox width

The Platypus article, showing the relative width of the infobox in relation to other features of the page.

I understand that the new theme has been in development for a number of years, but did anyone on the development team ever think about the interaction between the new restricted width and infoboxes? The already limited width is even narrower and harder to parse when an infobox is alongside. And yes, before I get a response, I do know that I can change my own preference on width or switch to the old theme, but this looks ridiculous for a logged out user. Eilidhmax (talk) 19:35, 18 January 2023 (UTC)Reply[reply]

Hey @Eilidhmax — by any chance might you have some userscript or gadget affecting the width of the infobox? I have not yet been able to replicate what you are seeing. For me in looks like this:
Platypus article on Vector 2022
Can you try viewing the article in a private/incognito window?
In general: the max-width takes into account the most common infobox sizes. Unfortunately, since we don't yet have global templates, there is a lot of variation in the widths. One way we might overcome this in the future is to make the max-width based on the width of the infobox, or just make some exceptions if the page is using a specific infobox template that is extra wide. AHollender (WMF) (talk) 19:45, 18 January 2023 (UTC)Reply[reply]
Why why why is there all that wasted white space, especially on the right side of the article window...what purpose is there to push the information/the text aside in favor of...white space? That wasted space is present regardless if editors/readers toggle the relative width. Or not. Don't readers, Wikipedia customers, come to the site to, oh I!? Shearonink (talk) 20:22, 18 January 2023 (UTC)Reply[reply]
The right side is for the page tools, which will launch next week. —TheDJ (talkcontribs) 21:00, 18 January 2023 (UTC)Reply[reply]
Oh GOODY. Now at least I know what New! Improved! feature is headed my way. Lulz I think I'm going to cling to Vector 2010/Classic Vector for a good long while, maybe until it is pried from my cold dead editing fingers. Shearonink (talk) 21:06, 18 January 2023 (UTC)Reply[reply]
I've tested and it's related to my chosen thumbnail size - a) why does that break this? and b) why is there no indication of what the default option is? Eilidhmax (talk) 20:38, 18 January 2023 (UTC)Reply[reply]

Do not force the creation of a user account

I, amongst doubtless thousands of others, was surprised to find Vector2022 as the appearance of wikipedia this morning. I doubt most viewers of wikipedia were aware of this project, and in particular that this redesign was going to arrive. I do not particularly like Vector2022, finding it annoying and a solution in pursuit of a problem. However, new UIs being generated for no apparent reason is a common trend, and I begrudge that fact. What was particularly galling was the lack of ready options to revert from Vector2022 to legacy Vector 2010. The only options available at present time are to either append "?useskin=vector" to every single wikipedia URL, a loathsome process (though doubtlessly one that will soon be automated by browser extensions), or to create an account and revert the appearance in preferences. I chose the latter, and am not particularly pleased by that. I did not want a wikipedia account. I am not of the invaluable sort who edits and contributes to this resource. I do not care to log-in to a website that I do not really interact with outside of base research and answering questions.

Please implement an option to revert to Vector 2010 without the need for a user account. A great many of those who use this website will be grateful. DoubleSashed (talk) 19:39, 18 January 2023 (UTC)Reply[reply]

DoubleSashed I think the only way to avoid using the new skin is to create an account, sorry. 331dot (talk) 19:40, 18 January 2023 (UTC)Reply[reply]
Thank you @DoubleSashed for your opinion. The reason why logged-out users can not save their preferences on Wikipedia is strictly related to the privacy policy (we just don't collect that data) and the architecture of our servers (what you see as a logged-out user is mostly a cached version, and not a version generated directly from the database). So as a logged-out user, you may use a bookmarklet. Copy that code and just replace "monobook" with "vector". Here you will find more information on that. SGrabarczuk (WMF) (talk) 20:08, 18 January 2023 (UTC)Reply[reply]
@SGrabarczuk (WMF): Can you clarify the privacy issue, or point me to the discussion where it was raised? I just started a brand-new, private browsing session, visited, and found six cookies, called "enwikimwuser-sessionId", "enwikiwmE-sessionTickLastTickTime", "enwikiwmE-sessionTickTickCount", "GeoIP", "WMF-Last-Access-Global:" and "WMF-Last-Access". How would adding one more cookie called "enwiki-skin" or whatever, be a problem strictly related to the privacy policy when all these other cookies aren't? Suffusion of Yellow (talk) 20:47, 18 January 2023 (UTC)Reply[reply]
@SGrabarczuk (WMF): ^^^^ Not sure if you noticed the question above. Suffusion of Yellow (talk) 20:20, 19 January 2023 (UTC)Reply[reply]
Hi @Suffusion of Yellow. Unfortunately, I'm not an expert in this matter. I only know what I've been able to collect and write or link to in the FAQ. I'll make a note to ping you if I learn more anything new soon. SGrabarczuk (WMF) (talk) 23:57, 19 January 2023 (UTC)Reply[reply]
Nonsense. You can save one cookie with the preference. There is no privacy involved. This is such a cockup, seriously. As someone who has been a donor for over a decade I'm frankly annoyed that I have to now manage accounts on ALL of my devices just so I can read a page that is laid out properly. G4zeroj (talk) 00:21, 19 January 2023 (UTC)Reply[reply]
While this is a solution, it is far from accessible. Please consider adding an option for laypeople.
I do not believe that forcing users to create an account furthers privacy. DoubleSashed (talk) 15:06, 19 January 2023 (UTC)Reply[reply]
As it was put in the RfC: "We can't pretend the settings are accessible to everyone when the user would have to go through all the steps of creating an account and logging in to use them. That would be a dark pattern." --Kizor 20:59, 18 January 2023 (UTC)Reply[reply]

Mousewheel support?

Mousewheel scrolling doesn't seem to work in this skin. Is that a bug or a "feature"? VQuakr (talk) 19:39, 18 January 2023 (UTC)Reply[reply]

It works for me. 331dot (talk) 19:40, 18 January 2023 (UTC)Reply[reply]
Hi VQuakr could you share a little more information? Does the issue occur with your gadgets / user scripts disabled (e.g. on this URL )?
Which browser and operating system are you using? Jdlrobson (talk) 19:52, 18 January 2023 (UTC)Reply[reply]
@Jdlrobson: Win 10 Pro/Chrome 109. Does the issue occur with your gadgets / user scripts disabled? sort of. It appears to be related to the TOC. If I expand the TOC enough to get a scrollbar, mousewheel will briefly scroll the TOC instead of the entire webpage, then switch to the entire page regardless of where I am hovering, then not scroll at all from mousewheel input (or, I'm seeing now, from left clicks on the light gray portion of the scrollbar or the direction arrows). But it's not totally frozen; keyboard input and click-drag on the scrollbar still have effect. The behavior is the same in normal mode as with safemode. VQuakr (talk) 20:05, 18 January 2023 (UTC)Reply[reply]
It does NOT work for me properly on my phone. Wow. Everytime you scroll it tries to scroll up. Valery Zapolodov (talk) 01:03, 19 January 2023 (UTC)Reply[reply]

Help Desk/Teahouse questions

The Help Desk and Teahouse are getting flooded with questions about this; is there some sort of blanket announcement message that could be sent out, perhaps directing people here? 331dot (talk) 19:41, 18 January 2023 (UTC)Reply[reply]

@331dot - thanks, that's a good idea! We're currently running banners to all users. For logged-out users, it's directing them to a landing page with general information about the skin and links to give feedback on the main Mediawiki page for the project. For logged-in users, it's linking them here. Potentially we can add a link to this page on the landing page of the foundation site too so logged-out users can see both conversations. OVasileva (WMF) (talk) 19:51, 18 January 2023 (UTC)Reply[reply]
No banner here. But I do get one for the next Wikimania Conference, if I run an incognito tab and then log in. Nfitz (talk) 23:55, 18 January 2023 (UTC)Reply[reply]

I'm on the fence

This feels like it was meant for mobile users. Some thoughts

1) So much space on both sides can be used/employed for viewing. 2) For logged in users, a contributions link and should be visible with one-click, not as part of a drop down 3) for me, I rather see all the left hand column stuff as is, without having it "hidden" behind one button.

Ebbedlila (talk) 19:42, 18 January 2023 (UTC)Reply[reply]

Ebbedlila They are trying to balance the needs/wants of new users with those of longtime users. That's probably the reason for the drop down menu and left hand column stuff. 331dot (talk) 19:44, 18 January 2023 (UTC)Reply[reply]
Hi @Ebbedlila, thank you for your feedback, please look at this section of our FAQ. Patafisik (WMF) (talk) 19:45, 18 January 2023 (UTC)Reply[reply]
Well, the FAQ's answers are disappointing. I'll have to look for external options to bring back the previous layout because I'm not interesting in logging in just to re-enable the old skin. Despite what the FAQ claims this new layout clearly has the trappings of mobile-friendly web design. There's simply too much wasted empty space and less text displayed on screen. - (talk) 21:37, 18 January 2023 (UTC)Reply[reply]
Hey @Ebbedlila,
Regarding the space on the sides: the starting point was introducing a limited line-length (more on the reasoning behind that here). With that change implemented, anyone on a big screen has a lot of blank space. We don't see that as an issue in and of itself. However it does present an opportunity. Next week we'll be releasing an update, which uses some of that space for page tools (see prototype here). Until we think of super high value uses for that space, we don't plan on filling it just to avoid having blank space.
Regarding the personal tools being in a menu: we're looking into options for showing more of those tools outside the menu at larger screen widths. The data regarding the usage of those tools indicates that people don't use them more than once or twice during a session, so I don't think there is much of downside to having them in a menu. On a higher-level this is an effort to clean up and organize the interface, and then reevaluate what the highest utility features are (which might be existing things, or new things we haven't built). For example imagine we develop a dark mode, and want to put a button/toggle in a consistent location for logged in & logged out people — now we would be able to do that. AHollender (WMF) (talk) 20:14, 19 January 2023 (UTC)Reply[reply]
Limiting line length strictly worse. In HTML body if you resize the window, the text will flow to match the new width. If you prefer a narrower (or wider!) width, simply resize the window. Setting a line limit removes that ability from the user. It seems like there was a huge amount of effort for what is a very limited impact based on "data" (I am skeptical of reasoning from the general to the specific when it comes to usability. The array of devices and individuals is so broad and high dimensional doing any kind of data analysis is going to fail. And judging by the response and feedback it has.) Eblon2 (talk) 22:33, 20 January 2023 (UTC)Reply[reply]

How to go back without logging in?

I don't want to log in on this computer just to restore the site back to being functional. Awful, hideous change. (talk) 19:46, 18 January 2023 (UTC)Reply[reply]

If there are specific concerns you have, please discuss them. Most if not all features are present, just in different places. This has been studied and developed for years to balance the needs/wants of every human on this planet, all of whom use the site differently. There may be browser extensions you can use to restore the old appearance, but without those, you will need to create an account to be able to use the old appearance. 331dot (talk) 19:48, 18 January 2023 (UTC)Reply[reply]
Could you actually address the question instead of trying to convince people they actually do like the thing they obviously don't? Thelaftwardbard (talk) 16:19, 19 January 2023 (UTC)Reply[reply]
Hey! You may use a bookmarklet. Copy that code and just replace "monobook" with "vector". SGrabarczuk (WMF) (talk) 19:49, 18 January 2023 (UTC)Reply[reply]
If the driving factor behind the change to Vector 2022 is usability, then why make the option to change back to the more familiar skin so esoteric? It's not impossible (or frankly difficult) to let the user change the skin (drop-down menu at the bottom of the page?) and save it as a cookie. Keep all traffic with default skin settings (i.e. no cookie) on Vector 2022, but redirect traffic to the ...?useskin=skin version of the page if a cookie is detected. This a.) avoids the caching issue, since most IP users won't care enough to change from the default, and b.) from my understanding, can be done entirely in a couple lines of client-side javascript.
If Wikipedia wants to promote usability, they should recognize that not everyone has the technical ability or freedom to create a Wikipedia account, create bookmarklets or install browser extensions. Those people should have the ability to choose what skin is most usable for them.
Thank you for your consideration. (talk) 00:09, 20 January 2023 (UTC)Reply[reply]
I love how IPs, the main target of this change, are saying "this doesn't look good" and the first things you recommend are things most people probably won't know how to do, or won't bother to do. Nothing can be simple for the targets of the change(especially considering how in the RFC, it was stated it would be difficult to poll them). Ijustlikeproperspelling (talk) 02:05, 20 January 2023 (UTC)Reply[reply]

Misplaced, jumbled tables and images

It was pointed out at Wikipedia:Requests for comment/Deployment of Vector (2022) that the new skin crowds tables and images to such an extent that it misplaces and awkwardly stacks them. This was, and still is, evident on the very article supplied as an example in the RfC, Pluto, See Pluto#Mass and size section. This is happening on countless articles. See 2022 Formula One World Championship for example. It's mess now with the new skin. --DB1729talk 19:54, 18 January 2023 (UTC)Reply[reply]

This is a longstanding issue that severely impacts our mobile readers (forcing them to have horizontal scroll on the entire article). It an be fixed by editors like in this change using the instructions on mw:Recommendations for mobile friendly articles on Wikimedia wikis#Wrap wide images and tables. Jdlrobson (talk) 20:22, 18 January 2023 (UTC)Reply[reply]
Every article that has a table of contents on the left, balanced with a vertical series of images on the right, will now have to be redesigned. Like 2022 Formula One World Championship. There are a lot of pages with this type of design that are now 'broken'. However, redesign or don't-redesign, it will look bad in one skin or the other. Your Pluto#Mass and size example looked equally bad on both skins on my computer... until I zoomed out (on legacy) and the images/tables fell into place; but on the new skin they never fell into place no matter how much I zoomed out. Grorp (talk) 20:28, 18 January 2023 (UTC)Reply[reply]
Pointing out when you stated Your Pluto#Mass and size example looked..., that wasn't my example. That was the example presented to us by the WMF in their own RfC. DB1729talk 16:42, 19 January 2023 (UTC)Reply[reply]

The lack of space between the lead section and rest of the content makes every article look like a stub

Off all the changes this one is the one I'm least a fan off, it really does not jive with what I've become used to. Having the TOC below the lead section was easy on the eyes Imho. ★Trekker (talk) 19:58, 18 January 2023 (UTC)Reply[reply]

Can you give an example of an article URL wihich has this issue? Jdlrobson (talk) 20:17, 18 January 2023 (UTC)Reply[reply]
Its happening when I'm logged in and editing, I don't think its an issue tied to an IP but just an issue I personally have with the new visual.★Trekker (talk) 20:31, 18 January 2023 (UTC)Reply[reply]
Hello! Thanks for your feedback - this is an interesting opinion, a rare one! I guess this is a bit of a tradeoff - sure, the previous look had some advantages the current one doesn't. But its disadvantages are the reason why we believe the change is an improvement nevertheless. You may learn more here about the tests we did, versions we compared, and related stuff. (And just as Jon, I'm curious if there are any specific articles where this looks particularly odd.) SGrabarczuk (WMF) (talk) 20:18, 18 January 2023 (UTC)Reply[reply]
I was reading this article and it took me a longer while that I would like to admit to realize that the TOC was now located by the side of the article. But really I don't think its a problem for one specific article as much as its a major break now compared to how I was used to looking at articles. Generally having a TOC separate the lead from the rest of the article was an indication to me that a page was not a stub but a more extensive article (but that may have been a personal bias of mine in the end) so now with that gone each time I glance at a page and there is no TOC I immediately get a 'stub!' alarm in my brain.★Trekker (talk) 20:31, 18 January 2023 (UTC)Reply[reply]
I can kind of see the reasoning for moving the TOC, but I think there should still be more space between the lead section and the rest of the article. Perhaps the Wikipedias in other languages do it differently, but on en-wiki, the lead section is supposed to be a summary of the body, which I feel means it should be somewhat set apart from the rest of the article. And adding a little space after the lead section would at least reduce the image-crowding problems that other editors have complained about. A. Parrot (talk) 17:44, 19 January 2023 (UTC)Reply[reply]
I fully agree with StarTrekker's point. ToCs are a necessary part of the article's body, separating the lead from the subsections. Also cfr. my RfC comment about it. Æo (talk) 16:32, 19 January 2023 (UTC)Reply[reply]

Poor announcement/publicity of this change

I would have to strongly agree with the comments above... this new change is not being clearly communicated. A (en.wikipedia-)global significant change has been made/published to the wikipedia layout today. There should be a front-and-center page saying, at the least, "To learn more about the global reformatting of wikipedia, check out *page*". It should be EASY to find, not just for active editors, not just for logged-in users, but for EVERYONE browsing Wikipedia, even anonymous Wikipedia viewers. (and ideally, even for people using a no-javascript browser). I had to spend a couple of minutes wandering around Wikipedia to find out what was going on ("Is my browser broken?"), and it would have been a lot longer if I wasn't already familiar with the behind-the-scenes parts of Wikipedia. Bsammon (talk) 20:15, 18 January 2023 (UTC)Reply[reply]

Some pages that don't mention this highly visible change, and I would suggest they should (maybe just a one-line "see here for more information"):
Bsammon (talk) 20:30, 18 January 2023 (UTC)Reply[reply]
  • I can say as an active editor that I as well was not prepared or aware of this change before it happened.★Trekker (talk) 20:32, 18 January 2023 (UTC)Reply[reply]
    As a very active editor and admin I also had no idea this was happening, I thought Wikipedia was having a server issue. But lets not focus on what has happened, lets focus on the future. Canterbury Tail talk 21:12, 18 January 2023 (UTC)Reply[reply]
Hi @Bsammon - thanks for your feedback and sorry to hear you weren't aware of the changes ahead of time. For logged-in users, we have been running banners asking people to try out the new skin for the past couple of months. We also ran banners to all logged-in users announcing the time of the change with a link to opt-out instructions and other information. Throughout the process, we've also sent out two watchlist notices - one during the RfC here on English Wikipedia, and another one last week about the upcoming change. We are also currently running banners with more information to all users, logged-in and logged-out. In the future, we can consider what additional notices might make information about the change even more visible. OVasileva (WMF) (talk) 23:02, 18 January 2023 (UTC)Reply[reply]
Add my name to the list of people who saw no such banners - I use Wikipedia almost daily, logged-in and logged-out. The last banner I saw was the donation banner. Frankly, the fact that an administrator of all people was unaware that this change would be happening should be proof enough that you didn't do nearly enough to ensure people were informed of this change and given the opportunity to comment. WalnutBun (talk) 02:54, 19 January 2023 (UTC)Reply[reply]
I've been using wikipedia for well over a few months and I have most definitely not seen a single one. Ijustlikeproperspelling (talk) 02:07, 20 January 2023 (UTC)Reply[reply]
I read Wikipedia logged out. Yesterday, when I first noticed this change, I used at least 30 minutes trying to fix my browser, because there was NO WARNING about this change. The banner telling about this change was not there until maybe 60 minutes later.
You need to warn people about changes like this. There was no advance warning for logged out users.
You should go back to old system now. After that reate a proper test version which you advertise widely and then ask feedback also from users who don't have wikipedia account. Make changes based on that feedback and try again. 2001:14BA:2BF7:F700:B26:BF56:169:9663 (talk) 08:10, 20 January 2023 (UTC)Reply[reply]
  • I've seen none of these banners User:OVasileva (WMF). There was something a year or two ago ... but nothing post-Covid. Perhaps this change can be reverted until the banners are fixed? Nfitz (talk) 23:57, 18 January 2023 (UTC)Reply[reply]

Odd effect with Wikilink to section

When I click this link here in the new skin - 2012 Africa Cup of Nations knockout stage#Zambia vs. Sudan - on Google Chrome (Windows 10), it loads the page and initially seems to go to the correct anchor, but then for some reason the page moves down, followed by the menu bar sliding on top of it, leaving the reader too far down the page. The top of the section linked should be visible. Cheers  — Amakuru (talk) 20:54, 18 January 2023 (UTC)Reply[reply]

This is the issue described in MediaWiki_talk:CommentsInLocalTime.js#This_discussion_renders_hash_fragment_unusable. Jdlrobson (talk) 22:03, 18 January 2023 (UTC)Reply[reply]
Hey @Amakuru, I believe the issue you're describing will be fixed by these two tasks:
AHollender (WMF) (talk) 20:46, 19 January 2023 (UTC)Reply[reply]

Another odd effect

While I'm here, might as well report this too - I've noticed a few times that when I scroll down a long page using the scrollbar, after a while the text on the page starts to appear in bold rather than the usual font. When I left go of the mouse button to release the scrollbar, the bold persists for a second or two and then disappears.  — Amakuru (talk) 20:59, 18 January 2023 (UTC)Reply[reply]

Hi @Amakuru, thanks for reporting this. If I identify your problem correctly, this is an issue of the browser (Chromium and alike) itself, and they seem to be working on this. More info here phab:T322978. SGrabarczuk (WMF) (talk) 21:21, 18 January 2023 (UTC)Reply[reply]

Vector 2022 in the News

It's already hit the headlines: [3] I don't know yet whether I like the new skin or not -- but it isn't worth the 10 years of effort and the money it cost, as described in the Slate article.Smallchief (talk) 21:08, 18 January 2023 (UTC)Reply[reply]

Who said it was 10 years of effort and money ? We just didn't do anything since 2010. —TheDJ (talkcontribs) 21:22, 18 January 2023 (UTC)Reply[reply]
According to the article I linked, Senior Designer, WMF User:Jorm proposed a new skin and developed a prototype in 2012-- and the proposal has been floating around ever since. The title of the article is "Wikipedia Has Spent Years on a Barely Noticeable Redesign." Any estimate on how much the new skin cost in staff time and salaries? Smallchief (talk) 21:48, 18 January 2023 (UTC)Reply[reply]

Is it possible to change the skin while logged out?

I would like to be able to change the skin, but it appears that it is only possible to change the skin while logged in. I prefer to browse Wikipedia while logged out, and it appears I have no recourse to change the skin now? 2607:2400:110C:FFFE:2:0:0:93 (talk) 21:10, 18 January 2023 (UTC)Reply[reply]

Hello. Thanks for asking! Yes, may use a bookmarklet. Copy that code and just replace "monobook" with "vector". SGrabarczuk (WMF) (talk) 21:28, 18 January 2023 (UTC)Reply[reply]
You must click it again each time you navigate to another page (if you would like to view the page to which you navigated in that special skin). Two clicks to view every article? That's not even a remotely viable solution. Suffusion of Yellow (talk) 21:31, 18 January 2023 (UTC)Reply[reply]
@Suffusion of Yellow, interesting, I've just tried this out in incognito and it's persistent for me. I don't know why there's this info about clicking again. SGrabarczuk (WMF) (talk) 21:44, 18 January 2023 (UTC)Reply[reply]
@SGrabarczuk (WMF): That shouldn't even be possible, unless something has changed about how the servers interpret useskin. Just to clear, you:
  1. Save bookmark to javascript:var url = new URL(location.href); url.searchParams.set('useskin', 'vector'); location.href = url;
  2. Open an incognito window.
  3. Visit
  4. Click on the bookmark, and wait for the page to reload in Vector-2010
  5. Click on the link to Peloneustes.
  6. ...And you're still in vector-2010???
That's not what that code does. The only way that would happen is if the servers were rewriting all the links in the page any time it sees useskin. Suffusion of Yellow (talk) 22:02, 18 January 2023 (UTC)Reply[reply]
Oh I see what's happening. The old Vector-2010 pages haven't completely flushed out of the caches yet. So sometimes you get a vector-2010 page, sometimes you get a Vector-2022 pages. This has nothing to do with whether you previously used the bookmarklet. Suffusion of Yellow (talk) 22:06, 18 January 2023 (UTC)Reply[reply]

Side menu on top

Oh, so many complaints! Well, of course it is to be expected; Wikians tend to call ourselves "progressive" but try tampering with our old and trusted ways, and hear us snarl! And yes, despite using the new skin on Simple English for many months, I still have a complaint.

My old eyes are fading, though not yet to the point that calls for special treatment. I just have to crank up my font size a little, to see easily. Problem is with how the side menu works. Yes, the option to hide it or see it is definitely good. Trouble is, when I click to see it, it pops up above the page. Is an optional top menu a bad thing? you may ask. It's a good thing if it's a proper top menu, and that's what's wrong. It's just a properly formatted side menu, but moved to the top where its format is not proper. Top menus, like bottom menus, ought to spread out across the page, not stick up far above. The format should be redesigned; I'm surprised if I'm the first to be whining about this particular feature as it's got to be a nuisance to some other editors besides us cranky old people.

One possibility might be to pop the menu to spread across the same line as the newly abbreviated top menu, making it take a few lines, along with all the features that used to be there and are now in a submenu. But anyway the side menu, when it must pop to the top, ought to become a proper top menu. Jim.henderson (talk) 21:12, 18 January 2023 (UTC)Reply[reply]

That means your page is less than 720px wide and the layout switches to a mobile compatible safety net.. You are pretty far zoomed in, or your screen is not big to begin with and you are zoomed in. I too hope that this will get some attention soon, as it doesn't really feel like what most people would expect. —TheDJ (talkcontribs) 21:21, 18 January 2023 (UTC)Reply[reply]
If I'm understanding what you are describing correctly, this will be addressed next week by phab:T302073. The current plan is to deploy to Italian and Vietnamese tomorrow and if no issues, to roll out the change to English and others next week. This will make the main menu a dropdown on tablet-like devices. Jdlrobson (talk) 22:08, 18 January 2023 (UTC)Reply[reply]
My physical display is 1920x1080 which would be plentiful except that I am doing Opera zoomed at 200% font so as to see clearly, so your assumption is close enough to being correct. Same happens with other browsers. Anyway I am hopeful. I fail to understand, from the short description, that the linked Phab is relevant to this particular gripe, but that doesn't mean it isn't, and anyway an industrious team is clearly at work and we can reasonably expect it to turn out well. I think I'll take a look at Italian tomorrow, though I don't know the language. Thanks for your attention. Jim.henderson (talk) 00:10, 19 January 2023 (UTC)Reply[reply]

Significant backlash

As I predicted hours before the deployment, there has been absolute chaos since the major UI change, and we're not even 12 hours in. Someone should seriously consider starting a new RfC or petition to restore the old default skin at least until a majority of the community's concerns have been addressed and there is strong consensus to adopt Vector 2022. The closing comment on the previous RfC that said an RfC would not be necessary after the fixed-width thing was addressed was misguided, in my opinion. InfiniteNexus (talk) 21:32, 18 January 2023 (UTC)Reply[reply]

I agree 100%.★Trekker (talk) 21:35, 18 January 2023 (UTC)Reply[reply]
I am looking at the reactions from Twitter (search "Wikipedia layout") and they largely echo the comments here. It seems the WMF is being more dogmatic about the whitespace issue than it is willing to work with the community. Clay (talk) 21:45, 18 January 2023 (UTC)Reply[reply]
Just read the comments to Wikipedia's tweet announcing the change. InfiniteNexus (talk) 21:53, 18 January 2023 (UTC)Reply[reply]
Twitter isn't a good place to gauge anything apart from hate Terasail[✉️] 22:17, 18 January 2023 (UTC)Reply[reply]
I disagree, but to each their own. Clay (talk) 22:22, 18 January 2023 (UTC)Reply[reply]
What regular person is going out of their way to tweet that they like the new wikipedia skin? They will only load twitter up if they have some complaints that they want to be heard... Terasail[✉️] 22:28, 18 January 2023 (UTC)Reply[reply]
People regularly comment on things when they like them, especially on Twitter where everything typed is treated as a throwaway comment. If you actually search the above on Twitter, you will even see some positive comments mixed in there. Clay (talk) 22:31, 18 January 2023 (UTC)Reply[reply]
Yes, Twitter is a good place to gauge hate. There seems to be a lot of hate today. Certes (talk) 22:27, 18 January 2023 (UTC)Reply[reply]
That changed after Elon. Valery Zapolodov (talk) 01:06, 19 January 2023 (UTC)Reply[reply]
It seems to me that the pro Vector 2022 people are dismissing a lot of the concerns as a kneejerk reaction to change. What are the odds that some of these reconfigurations are actually fixed? WikEdits5 (talk) 19:59, 19 January 2023 (UTC)Reply[reply]
I agree on the creation of a new RFC as well, the WMF has been so hush-hush about this rollout and did not bother to mass-notify anyone until the week of deployment. Any opinions expressed before today were of the very active technical users and clearly does not represent the overall consensus of actual editors and readers. Cards84664 22:15, 18 January 2023 (UTC)Reply[reply]
I strongly agree. I saw some post by a defender (designer?) of the new UI somewhere above on this page, claiming that "the majority" was favorable to this change, while linking a page in which about 160 people had expressed their opinion. This doesn't represent the millions of wikipedia readers (most of whom have no account) anyhow. Just give unlogged users a (highly visible) way to revert to the old UI and gather data about how many people reverted. Then you'll know if your new UI is liked or not. (talk) 03:40, 19 January 2023 (UTC)Reply[reply]
I can assure you that I don’t have a COI. 300 people is already a quite large survey sample. You can see mw:Reading/Web/Desktop Improvements/Frequently asked questions §§ Why is the opt-out link not available for logged-out users?​ and Why are there no preferences for anonymous users? for the last part. Aaron Liu (talk) 13:04, 19 January 2023 (UTC)Reply[reply]
Hi @InfiniteNexus - thanks for your feedback. If you're curious, I wanted to point you to to this section which highlights the conversations and collaborations with the English Wikipedia community both before and after the RfC. These conversations helped us in determining which changes to make to ensure we're meeting the needs of both the community and our readers. In addition, we'd like to point out that the skin was very popular across many logged-in users over time, many of whom were not part of the more technical members of the community. Prior to deployment it was the most popular non-default skin, with more active editors using it than any other non-default skin (such as Monobook, Timeless, etc). Hope this is helpful! OVasileva (WMF) (talk) 22:23, 18 January 2023 (UTC)Reply[reply]
The point is a lot of people are hating it despite community consensus, the changes that are made is irrelevant. What matters is a lot of people hate it for some reason so there should be another rfc to let their voice be heard and do “damage control” Aaron Liu (talk) 22:40, 18 January 2023 (UTC)Reply[reply]
How many people made up the community consensus?★Trekker (talk) 23:07, 18 January 2023 (UTC)Reply[reply]
The rfc was at Wikipedia:Requests for comment/Deployment of Vector (2022). THere were about 300 people. About half opposed but according to the closing notes written by two very notable community members unaffiliated with the foundation they agree to pass it if the issues the opposing people brought up were addressed, and they were. Aaron Liu (talk) 23:31, 18 January 2023 (UTC)Reply[reply]
I don't understand -- a majority of participants in that RfC opposed the deployment, yet it went forward anyways? Tkbrett (✉) 00:32, 19 January 2023 (UTC)Reply[reply]
RfCs don't operate on votes. —Tenryuu 🐲 ( 💬 • 📝 ) 00:47, 19 January 2023 (UTC)Reply[reply]
300 people (half of which opposed the change) is what you consider a "community consensus" on a website used by millions? Even if we only considered active users with an account, this wouldn't even be 5% of the community... (talk) 03:45, 19 January 2023 (UTC)Reply[reply]
Hi, RfCs don’t operate on !votes, they operate on consensus. In this case the consensus appears to be that if the issues brought up by the opposed are resolved then it can be deployed. The problem here is that these issues appear to have been resolved controversially.
300 is already quite large for Wikipedia RfC standards, and it’s a large enough survey sample. If 300 random people have achieved consensus why would the millions’ opinion be much different? Aaron Liu (talk) 12:59, 19 January 2023 (UTC)Reply[reply]
1. Selection bias. 2. Consensus has clearly changed. IWantTheOldInterfaceBack (talk) 13:14, 19 January 2023 (UTC)Reply[reply]
  1. There was no selection bias here, save for the fact that all who !voted were pretty active.
  2. That's what I'm saying. What matters is a lot of people hate it for some reason so there should be another rfc
Aaron Liu (talk) 13:16, 19 January 2023 (UTC)Reply[reply]
Of course there was selection bias. As a heavy user of the website but not an editor, I had no idea this was going on. Only those deeply plugged into the interminable bureaucracy had any idea this was happening. Pretending this is not the case does not help your cause. Furthermore, it's very different to talk about potential future changes than to actually react to the impact of them. For example, see the House voting 60 times to repeal Obamacare when Republicans held it but not the Senate, and then failing to get it repealed when they had a full trifecta. IWantTheOldInterfaceBack (talk) 13:20, 19 January 2023 (UTC)Reply[reply]
No. All accounts who were active enough should've seen notices plugged in several places, such as {{centralized discussion}}, watchlist notices, the village pump, arbcom elections, admin newsletters, the signpost newsletter, responses at the help desk and teahouse, etc. Though the average reader that doesn't care about signpost wouldn't know about it, almost all editors with a changed watchlist have seen the publicizing of that rfc. Aaron Liu (talk) 13:25, 19 January 2023 (UTC)Reply[reply]
@Aaron Liu: I'd call myself a very, very active editor and I was totally blind-sided by this sudden switch; many other editors who are fairly active are saying the same thing, too! So stating that "all accounts who were active enough should've seen notices" about it is completely wrong.--Gen. Quon[Talk] 16:48, 19 January 2023 (UTC)Reply[reply]
Agreed. I'm here every day and similarly did not know about it until it was implemented. I would have opposed things in the RfC, but it doesn't seem like it would have mattered since the majority that opposed implementation was ignored anyways ... Tkbrett (✉) 19:36, 19 January 2023 (UTC)Reply[reply]
I found out about this redesign only after the pages I'm viewing started changing appearance. One page could be Vector 2022, another one Vector 2010, so I basically had two themes depending on which article I was viewing (nice deployment). I thought it was a bug, so I went to preferences, noticed Vector 2022 enabled, cursed a bit (I hate when my settings are changed without my consent), and went back to Vector 2010. It was a couple days ago, and only now I've realized this was actually a site-wide change in defaults. Granted, I'm not that active in "meta-pedia" stuff, but I visit here almost every day, and I was completely unaware about all of this. RoadTrain (talk) 23:44, 19 January 2023 (UTC)Reply[reply]
Also I'll repeat what I'm saying, a lot of people hate it now that it's actually implemented so YES THERE SHOULD BE A NEW RFC Aaron Liu (talk) 13:27, 19 January 2023 (UTC)Reply[reply]
  • Support Changing back to the previous skin, until many issues can be addressed, and communications can be improved. Nfitz (talk) 23:59, 18 January 2023 (UTC)Reply[reply]
    @Nfitz This is not the new rfc (it doesn't exist yet), though somebody can start one, preferably in the same format as the one before. Aaron Liu (talk) 00:11, 19 January 2023 (UTC)Reply[reply]
  • Yeah, this is just generally awful. GMGtalk 01:30, 19 January 2023 (UTC)Reply[reply]
    It's a terrible design, and telling people to get used to it is not good enough. (talk) 01:58, 19 January 2023 (UTC)Reply[reply]
    It doesn't make anything "more readable". (talk) 02:04, 19 January 2023 (UTC)Reply[reply]
  • Support Bringing back Vector 2010. People can say "consensus" how many times they want but in the end its clear that incredibly few people were actually involved in forcing this on the majority who don't want it.★Trekker (talk) 13:09, 19 January 2023 (UTC)Reply[reply]
    Did I skip something? Why is everyone treating this section as an rfc? Aaron Liu (talk) 13:11, 19 January 2023 (UTC)Reply[reply]
    They're expressing consensus. You know, the thing you've been telling everyone Wikipedia operates on. ~~ AirshipJungleman29 (talk) 15:11, 19 January 2023 (UTC)Reply[reply]
    I fail to see how that makes this section the formal discussion, in fact this entire talk page is mainly comprised of people seeking consensus but not in a !vote format. Aaron Liu (talk) 15:22, 19 January 2023 (UTC)Reply[reply]
    Then let's launch the new RfC, is everyone okay with the proposed wording at #Draft RfC question? InfiniteNexus (talk) 17:38, 19 January 2023 (UTC)Reply[reply]
  • Yesterday I stared and then dived for my preferences; thank you for making it so easy to revert back to Vector 2010! As others appear to be reacting with similar dismay, I would support an RfC. (I understand that there is some consideration being given to enabling non-logged in users to choose between formats, as well; please follow up with this!) Butterfly or Chuang Tzu? (talk) 23:59, 19 January 2023 (UTC)Reply[reply]

If there is consensus for an RfC, where would be the appropriate venue? I don't think it should be this page, the RfC will likely grow pretty large, and the section will likely get lost in the swarm of new complaints that will keep coming in during the next several days/weeks. InfiniteNexus (talk) 00:55, 19 January 2023 (UTC)Reply[reply]

We can see the way the original RfC (Wikipedia:Requests for comment/Deployment of Vector (2022)) was done. The starter first drafted it in sandbox, moved it to the subpage to continue drafting, finished it up and started it, and at least requested watchlist notices. I'm not sure what other publicizing was involved atm but ither ways of publiciizing it include asking at the village pump technical while drafting on how good the rfc looks, publicizing it at other village pumb sections, including it in {{centralized discussion}}, and adding a sitenotice. Aaron Liu (talk) 02:20, 19 January 2023 (UTC)Reply[reply]

Loving it!

Just a note on the other side of the "backlash" that I'm loving it. I thought maybe I had gone to the mobile site instead, but it looks great to me. It'll take a little getting used to, but all great things do. -- Michael Scott Asato Cuthbert (talk) 22:23, 18 January 2023 (UTC)Reply[reply]

Thanks @Mscuthbert, we appreciate your comment here! SGrabarczuk (WMF) (talk) 00:01, 19 January 2023 (UTC)Reply[reply]
So you'd be just as happy if it were optional, then? (talk) 02:11, 19 January 2023 (UTC)Reply[reply]
Given that they signed their post as Michael Scott, I assume it was entirely ironic. — LlywelynII 08:23, 19 January 2023 (UTC)Reply[reply]
LlywelynII While your idea did make me laugh, it looks like that's their name.--Gen. Quon[Talk] 17:13, 19 January 2023 (UTC)Reply[reply]
Huh. — LlywelynII 21:34, 19 January 2023 (UTC)Reply[reply]
Not ironic, I wager, knowing MSC :) Who is exacting in music + aesthetics, so I'd take this as a positive signal. – SJ + 16:22, 20 January 2023 (UTC)Reply[reply]

Vector 2022, go to hell.

I am pissed off that this format change was changed on accounts that have specifically chosen to use Vector classic over the others. I've looked at and previewed this and the other skins before and I've rejected vector 2022 and all the others for layout spacing and button placement reasons. Only Vector classic is functionally pleasant to use. I reverted today's unwelcomed change in my settings immediately this morning and I will NOT be supporting this Vector 2022 mess. The spacing of various elements is utter crap in this design. Too wide, bloated, and roomy on the sidebar, and a fucked up narrow squish fest in the read/edit pane. Not pleased with the user account navigation buttons all hiding in a drop down menu.

*IF AND ONLY IF* the left sidebar was brought back to a reasonable width (vector classic widths or so), with the edit pane reclaiming that same amount, AND the account navigation buttons were restored to the top right of the site as they appear in Vector classic, THEN I might be supportive and try out the skin, but these two issues are nonnegotiable details that I will not compromise on.

Some website design changes/overhauls I have loved and been supportive of (on other websites), but this Vector 2022 is currently a farce and I will openly reject it until the end, or until the critical issues stated above are corrected. Until then, Vector classic user for life.
Respectfully, Zinnober9 (talk) 22:45, 18 January 2023 (UTC)Reply[reply]

Why is Vector2022 default?

(moving this question from Teahouse +some notes).
Hello from el and en.wiktionary. I have a username, but also keep a 2nd browser alongside, with no login. Table of Contents is so hard to find, plus other issues (cannot hop from one language to another etc). I have congratulated you on resisting Vector2022 here, but sadly, I see today that horrible new skin at my screen as default in the English Wikipedia. Especially, as per Wikipedia:Vector_2022.

[…] "suggest trying it for at least one week prior to deciding whether to switch to one of our older skins." […] "if you are unsatisfied, you can switch to any of our other skins at any time."

If any of you think that making a trial skin default is not a good decision, you could join the mediawiki discussions, like I did, at "Let the public decide" And always: please keep a visible button for switching to classic skin, not just found only at Preferences.
PS The "support-oppose" procedure is endless. Because de gustibus et coloribus non est disputandum. So, at least, Let the readers have choices. Thank you for your hard work, From el.wikt, Sarri.greek (talk) 22:50, 18 January 2023 (UTC)Reply[reply]

Poor use of space

Why are the left and right 1/4th of my screen completely empty with this new look? This is an incredibly inefficient use of screen space, I think it's fine as an optional skin but this should NOT be the default for users. IAMspecials (talk) 23:24, 18 January 2023 (UTC)Reply[reply]

@IAMspecials Please see § How to expand the width of the new skin Aaron Liu (talk) 23:34, 18 January 2023 (UTC)Reply[reply]
Hi @IAMspecials - thanks for your feedback. Generally, we've limited the width of the content, as this makes it easier to read and also more likely to retain or remember the content which was read. If you're curious, we've added more details in our FAQ about the academic studies we looked at as well as our internal work and thinking around this. That said - the width is configurable and you can change it if you wish. We've left some instructions on how you can configure the width of the content in the new skin on the project page. OVasileva (WMF) (talk) 23:36, 18 January 2023 (UTC)Reply[reply]
You are incorrect, User:OVasileva (WMF) - reducing the width makes it harder to read. And then there's the over-use of white; and my gosh, WTF would I want to see the TOC all the time? And why not use the left-hand side to show languages, instead of having them under an extra click underneath a mystery icon that you have to scroll up to to get? I can only conclude that the interface was designed by someone who can only read Wikipedia in one language! Nfitz (talk) 00:48, 19 January 2023 (UTC)Reply[reply]
No, it doesn't. Reducing the amount of information on the screen makes the website harder to use and nearly unreadable. If I want less information on my screen at a time, I will simply narrow the width of my browser window.
People who want less width can narrow their browser window. Why force this on us all? Most people will never bother to make an account like I did.
I demand whoever did this reverse it NOW. This is clunky, unusable and will make people significantly less productive. IWantTheOldInterfaceBack (talk) 04:38, 19 January 2023 (UTC)Reply[reply]
@IWantTheOldInterfaceBack Keep in mind that everyone can just press the button on the bottom right to stop make it narrow, this was also addressed in the previous RfC. A ton of studies have shown that keeping line width between certain numbers does make reading easier by reducing the movement of eyeballs. I agree that there is way too much stuff replaced with white though. Aaron Liu (talk) 13:09, 19 January 2023 (UTC)Reply[reply]
I don't see that button on my 1920X1080 screen unless I zoom out for some reason. Can we make it visible for everyone? –dlthewave 03:29, 20 January 2023 (UTC)Reply[reply]
You can also search for it in preferences, this is addressed in the faq and the corresponding section in the article. I do agree that it should be more visible, @OVasileva (WMF) could you file/search for a phab ticket for us? many thanks! Aaron Liu (talk) 04:06, 20 January 2023 (UTC)Reply[reply]

Thanks! I have Windows scaled to 125% (this is very common) so perhaps it thinks the display is below the threshold to show the button. –dlthewave 04:35, 20 January 2023 (UTC)Reply[reply]

Some Feedback

- Putting the table of contents on the side is great, that is really going to improve the navigability of longer articles.

- On that note, I do really disagree with where the language settings went. Not only are they hidden beneath an additional menu when they were immediately available beforehand, the new menu is in my opinion in a bad location and does not match with the surrounding style at all.

- As has been mentioned in other comments, more 'mystery meat' icons are a really bad idea for a layout that has as one of it's goals to meet the needs of those that have been newly connected to the internet.

- I think that the editor options at the top of the article should go back to being above the title page instead of inbetween the title and the contents, as they are generally not of interested for most users.

- The enlarge width button on the bottom right, aside from being a mystery meat icon, really should remember it's state between pages, but I guess that might be an issue caused by the rollout.

- Personally I would love to see the available whitespace created by decreasing the article width be used for something else instead of being simply empty. I understand text being not full width is better for the reading experience (even if remain doubtful that it actually leads to better information retention), but I think the free space should be used for information that is not necessarily part of the article. In my opinion, in would be worth experimenting with putting pictures and other information boxes to the side, so that the information density can be increased without making the reading experience go back to widescreen mode. Maybe don't experiment on the main branch next time, however:

- This update came maybe not out of nowhere but it does feel fairly underdeveloped yet, see for instance the issues with tables and images being placed like by a poor LaTeX compiler when they start conflicting with the new article width.

- I am unsure of what I think about the new search function. Adding a short sentence to the result that shows slightly more desciptive information that just the name is one hand a nice feature, but on the other hand again either decreases the amount of search results that can be on screen or makes the search bar result much longer. Personally, I also don't think I will need it, but it might be a neat feature for others.

- The whole skin does not feel like a cohesive whole, as the feel of the old skin is preserved inside of the new article column, but outside of that it looks very different, mostly in how the site responds. Personally, I prefer the old skin where there was not a small animation responding to every mouse move and click, but I think having these two parts coexist on one page is much worse. Compare that to something like wikiwand, where the whole UI feels like one design. 2A04:8400:C180:8601:FDA5:78AC:7FEE:8530 (talk) 00:09, 19 January 2023 (UTC)Reply[reply]

I do agree that the table of contents on the left-side is solid; it makes navigation easier. But the default white space and the 'mystery meat' icons are steps way, way backwards.--Gen. Quon[Talk] 17:16, 19 January 2023 (UTC)Reply[reply]

Whitespace flooding, low contrast/visibility toggle, poorer readbility in general, and account creation requirements to be able to have persistent settings?

  • These changes in total make wikipedia an eyesore to me.
  • There is significantly more whitespace and enlarged margins, even when in the expanded view. The table of contents takes up twice the space of the old side bar, which means it is encroaching upon the body of articles, the right side margin is even larger as well, which compacts the body more.
  • The toggle should be an actual toggle, as in a persistent setting. Not something that goes away when you go to a new page when logged out.
  • The toggle should also be put in a much more visible location. Fandom places it bound to the upper left side of the content, in a bold symbol that is eye catching as well. Instead what is here blends into the square nature of a screen or browser window and becomes lost.
  • When logged out the constant switching between layouts is disorienting.

If this was intended for readability and accessibility I'd give it a 2/10. It makes it worse across the board, with no notable benefits beyond the table of contents being bound to the side. Since this was pushed through despite the opposition to it I have a bad feeling for the future of the project. Deadoon (talk) 00:40, 19 January 2023 (UTC)Reply[reply]

Requests for comment/Reverse deployment of Vector (2022)

As said in Significant backlash above, a couple of people have talked about starting an RFC in order to reverse the deployment of Vector (2022) and leave the old design as the default due to backlash.

Should the decision on the default deployment of the Vector 2022 skin be reversed and make the default Vector Legacy? Pyraminxsolver (talk) 00:59, 19 January 2023 (UTC)Reply[reply]

This is not an RfC. RfC's need to be formatted in a specific way, with specific templates. And I have raised the question of where the RfC should be held, see #Significant backlash. InfiniteNexus (talk) 01:01, 19 January 2023 (UTC)Reply[reply]
Can I get some help with starting it? I've participated in a couple of them but I've never started one. Pyraminxsolver (talk) 01:03, 19 January 2023 (UTC)Reply[reply]
(edit conflict) Well, before we start one, we have to figure out (1) is there enough support for a new RfC to be opened, and (2) where should the RfC be held? InfiniteNexus (talk) 01:05, 19 January 2023 (UTC)Reply[reply]
"The bureaucracy is growing to meet the needs of the growing bureaucracy".
I don't edit Wikipedia or participate in its governance, in fact I created an account just to get away from this new godawful redesign. But surely if it's this unpopular someone trusted can just set up a poll or something and then use the results to determine whether to revert back? IWantTheOldInterfaceBack (talk) 04:30, 19 January 2023 (UTC)Reply[reply]
We know what an RFC template is. Wow. The question is to discuss before it is done. Valery Zapolodov (talk) 01:05, 19 January 2023 (UTC)Reply[reply]
  • Support reverse deployment of Vector 2022 and blocking whichever switched it without communication and consensus. Nfitz (talk) 01:11, 19 January 2023 (UTC)Reply[reply]
  • Support - I support this informal RfC, and I support starting a properly formatted one. There needs to be more discussion regarding the default deployment. Perhaps the new Vector 2022 skin is better for our readership, but it's not so good for editors who want their tools easily available. It is hard to find things! Netherzone (talk) 01:18, 19 January 2023 (UTC)Reply[reply]
I don't think this deployment will be reversed, even if we really wanted to. Also we need to give WMF time to fix out all the little kinks that come with deployment before we can say for certain that this skin is terrible. The skin appears to works mostly well for readers, but for editors there are still bugs that need to be ironed out. This might be a case of Wikipedia:Office actions. We might come to consensus to express disapproval for the rollout, but I don't think much will change technically otherwise. If the deployment is reversed, it may just be delaying the inevitable.
One of the things I would suggest is slow, incremental change, so that the same skin can get improvements, improvements, and improvements without having to do a complete rewrite of a skin. Just like Windows or iOS or Android gets incremental updates, the same should be done with this new skin that I don't think should be even called Vector 2022. Aasim - Herrscher of Wikis ❄️ 01:41, 19 January 2023 (UTC)Reply[reply]
A hypothetical RfC wouldn't close the door on Vector 2022 being deployed in the future. It's just that this launch happened way too soon and way too fast, when so many editors didn't even have the chance to weigh in. If the deployment is indeed reversed, the Web team is welcome to start a follow-up discussion and keep tweaking Vector 2022 until the majority of editors are fully satisfied, and then they can send out notices to all active users (they already do this for things like ArbCom elections, I would argue a default skin change is just as important) to alert them of another RfC. If there is strong consensus to deploy it, sure, let's do it; if not, they can repeat the whole process again. InfiniteNexus (talk) 01:54, 19 January 2023 (UTC)Reply[reply]
No, I can say for certain that it's terrible because it launched without an integrated ability for non-logged-in users to change it. Simple as. (talk) 22:08, 19 January 2023 (UTC)Reply[reply]

Anonymous preferences would make the pages load too slowly.
Most traffic comes from logged-out users. To handle that, we have a few "caching servers" which only save and send "snapshots" of web pages. These "snapshots" are up to 7 days old, are a replacement for generating web pages, and are the same for all logged-out users. This allows us to serve pages quickly.
Preferences require generating different versions of web pages. Making this possible for logged-out users would overload our servers. We don't want to do that also because we need to reduce cache fragmentation.
The only possible way of providing preferences for logged-out users now is making the settings load always after the page. This takes much more time to load and looks odd. For example, if a logged-out user was to see the dark mode in action, then immediately after loading each page, they would first see the light interface for a short moment, and then the interface would become dark.
— mw:Reading/Web/Desktop Improvements/Frequently asked questions

The same rationale could be used to say why they don't. Aaron Liu (talk) 22:39, 19 January 2023 (UTC)Reply[reply]
@Awesome Aasim No, this was not office actions. It was from an interpretation of the consensus from two esteemed editors that closed WP:V22RFC. If the rfc to undo this passes we should be able to ask the foundation to revert this. Aaron Liu (talk) 22:42, 19 January 2023 (UTC)Reply[reply]

Hey, you'd like to talk to the Web team about specific issues, which we'd be happy to discuss, feel invited to our office hours, an open meeting on Discord tomorrow (Thursday) at 20:00 UTC, on the Open Meetings voice channel. I'd also like to encourage you to take a look at basic documentation pages like mw:Reading/Web/Desktop Improvements and learn more about the decisions that have been made. Thank you! SGrabarczuk (WMF) (talk) 01:26, 19 January 2023 (UTC)Reply[reply]

The problem isn't just that Vector 2022 is difficult to navigate and not aesthetically pleasing, it's that the Web team failed to properly communicate these changes to all active users prior to today's launch. That's why you're seeing all these complaints rolling in, a lot of folks didn't even know this was happening until this morning, and they may not have been aware of the previous RfC or Village pump discussions. InfiniteNexus (talk) 01:45, 19 January 2023 (UTC)Reply[reply]
InfiniteNexus The change was communicated to me, there were notices beforehand. If they weren't aware of the prior RfC or other discussions what makes you think they will be aware of the new one? No matter what changes were made, some group of people was bound to be upset. It is impossible to please everyone or expect that the website will never be changed ever. 331dot (talk) 16:20, 19 January 2023 (UTC)Reply[reply]
The new RfC should, in my opinion, be advertised to all active users on their user talk pages in a similar fashion to ArbCom election notifications. Because that did not happen last time, only a minority of Wikipedians were aware of the proposal, and many did not get a chance to weigh in. Furthermore, the closers wrote (bolding my own): As such, our close aims to be conservative, given also the scale of the change and that each opposing editor likely did not list every concern which they agreed with if it was already raised by others. We encourage the Web team to be sure they have addressed the community’s concerns on these issues before moving forward. I don't think that happened. The Web team also chose not to start a second RfC after implementing the fixed-width toggle, so they never got the chance to hear the other concerns editors had with the skin (beyond fixed width). InfiniteNexus (talk) 17:37, 19 January 2023 (UTC)Reply[reply]
The change was 100% not communicated at all unless you were logged in to an account. (talk) 22:10, 19 January 2023 (UTC)Reply[reply]
  • Support starting an RfC on this issue. DB1729talk 02:58, 19 January 2023 (UTC)Reply[reply]
  • Support: Vector 2022 is dreadful, I have already switched my account global preference back to Vector 2010. There is a strong demand and need for a new RfC widely advertised to all users, both registered and unregistered.--Æo (talk) 16:16, 19 January 2023 (UTC)Reply[reply]

Draft RfC question

Should Vector legacy be restored as the default skin on the desktop English Wikipedia site, with the possibility of Vector 2022 being deployed at a later date if there is consensus from the community to do so?

Suggestions on the wording? We can add background info below the question (which has to be brief and neutral). InfiniteNexus (talk) 02:02, 19 January 2023 (UTC)Reply[reply]

One of the problems with the rollout is that it did not come out as expected. Those that had Vector 2010 set as their skin were suddenly forced to Vector 2022. Now almost everyone is on Vector 2022, no one asked, and the people remaining on Vector 2010 are those that opted out. If we were to roll back to Vector 2010, then all the people who would prefer Vector 2022 would wonder why their skin was forcibly changed. It is a lose-lose situation unless if WMF somehow backed up the skin preferences for all editors before changing the skin for everyone.
The rollout should have been done such that the default be changed without tampering with other people's preferences. Granted, logged out users would still be complaining, but logged in users would not have any trouble. People who have had their preferences forcibly changed should have their previous preferences restored. That is the least I can ask for. Vector 2022 is fine for those that are logged out, but I am opposed to changing people's settings just to get people to use a specific skin.
Editors that have been here long enough to witness the transition between Monobook and Vector 2010: Was the change done with or without consensus, and were your preferences left intact after the defaults were changed? It might have been a long time so most people probably do not remember, but it is worth asking anyway just in case. Aasim - Herrscher of Wikis ❄️ 05:41, 19 January 2023 (UTC)Reply[reply]
Looks pretty good to me, I do have some changes for the grammar though:
, with the possibility of Vector 2022 being deployed at a later date
with the possibility of deploying Vector 2022 later
I also recommend including the keep skin workaround, maybe something like

You may keep your current skin by changing the skin in global preferences.

Aaron Liu (talk) 17:53, 19 January 2023 (UTC)Reply[reply]
How's this?

Should Vector legacy be restored as the default skin on the desktop English Wikipedia site, with the possibility of deploying Vector 2022 in the future if there is consensus from the community to do so?

I don't think the "keep your current skin" guidance is relevant to the question, but we can add it to the background info below it. InfiniteNexus (talk) 19:48, 19 January 2023 (UTC)Reply[reply]
Yeah I meant that that part should be added to the background info. Looks good to me Aaron Liu (talk) 21:49, 19 January 2023 (UTC)Reply[reply]
I think it should make three clear points as a prerequisite to any reintroduction:
1. The reintroduction must not be done without clear, broad community consensus on the finalized version of the skin.
2. There must be a way to persistently change the skin and toggle page width for logged-out users.
3. The skin must receive a name of its own. Having to disambiguate between Vector (2010) and Vector (2022 2023) is convenient for exactly nobody and serves only to confuse less-active users. (talk) 22:18, 19 January 2023 (UTC)Reply[reply]
1 is unclear, I think it should be changed to "another RfC on the updated version of the skin".
The foundation has denied allocating server load to rendering two versions of every page. mw:Reading/Web/Desktop Improvements/Frequently asked questions §§ Why is the opt-out link not available for logged-out users?​ and Why are there no preferences for anonymous users? Aaron Liu (talk) 22:47, 19 January 2023 (UTC)Reply[reply]
Any concerns about the skin and future deployment can be expressed in the actual !votes, not the question. Again, per WP:RFCBRIEF, the RfC question must be brief and neutral. InfiniteNexus (talk) 00:13, 20 January 2023 (UTC)Reply[reply]
Agree with this, that the new skin should have a different name. Netherzone (talk) 01:08, 20 January 2023 (UTC)Reply[reply]

Seeing as there is consensus for a new RfC, I'll go ahead and move forward with it. Please stand by. InfiniteNexus (talk) 00:19, 20 January 2023 (UTC)Reply[reply]

Aborted, it turns out someone else already started an RfC at Wikipedia:Village pump (proposals)#RfC: Should Wikipedia return to Vector 2010 as the default skin? several hours ago. Frustrating that this talk page was not notified. InfiniteNexus (talk) 00:38, 20 January 2023 (UTC)Reply[reply]
Sigh. I guess I"ll try to publicize it then. Aaron Liu (talk) 00:44, 20 January 2023 (UTC)Reply[reply]

Noticed the skin was not deployed in a good manner

From what I am seeing above, Vector 2022 was enabled even for logged in readers using Vector 2010. I don't think that was a smart move. Sure, those logged out will be able to try out Vector 2022, but there is no need to force switch it for everyone else!

Some other pieces of feedback:

  • The right sidebar is empty. The right sidebar would be a good place to put links to widgets like recent changes and the page history.
  • The toggleable full width is botched a bit. Content should start out with the two fixed-width sidebars, and then the content be expanded into full width upon touching. There should not be lack of symmetry.
  • On the topic of full width, certain pages like the Main Page should be full width to start, to avoid awkward situations like the site notice eating into the sidebars.
  • Please also maintain consistency from page to page with navigation. The navigation buttons should not move around when moving from page to page.

I think this is a good start, but these fundamental problems are kinda dealbreakers. I am sticking with Timeless for now. Thanks. Aasim - Herrscher of Wikis ❄️ 01:28, 19 January 2023 (UTC)Reply[reply]

Also what happened to making Vector 2022 responsive? The current Minerva does not serve everyone well either. Provided that the responsive design is only turned on for readers and new editors, I don't see a reason not to do this. Aasim - Herrscher of Wikis ❄️ 01:33, 19 January 2023 (UTC)Reply[reply]
I also do hope for a full and proper dark mode for Vector 2022. That way I can read Wikipedia without straining my eyes out at night. Aasim - Herrscher of Wikis ❄️ 01:43, 19 January 2023 (UTC)Reply[reply]
Okay so I am going to give Vector 2022 a chance. Maybe in a week or so I find it easier to use. In its current state the skin has a long way to go before it matches or exceeds the functionality of Timeless. Aasim - Herrscher of Wikis ❄️ 13:50, 19 January 2023 (UTC)Reply[reply]
Aaand... I switched back to Timeless because of the lack of responsive design, despite me having responsive mode enabled in preferences. This might be a bug @SGrabarczuk (WMF)? Aasim - Herrscher of Wikis ❄️ 00:46, 21 January 2023 (UTC)Reply[reply]

Who greenlit this?

Who decided to roll out this hideous design without even adding a way to turn it off without logging in? Who was responsible for this idiotic decision? And no, I don't want your bookmarklet or your Tampermonkey script! I want competent web design! (talk) 01:39, 19 January 2023 (UTC)Reply[reply]

Made an account just so that I could go back to the old version. Did they even test this with people before they rolled it out? It objectively looks much much worse. Cgidwani99 (talk) 02:05, 19 January 2023 (UTC)Reply[reply]
Idk man, but this is unbearable. This design looks like it was made for tablets and smartphones. Both are evil. Please don't use your tablets and smartphones much you must be as uncomfortable as possible using a PC (also not Mac because macs are ridiculous, just PC not linux Chinese stuff etc.) and get up and get some exercise. It's the only way to stop the tiktok idiocy incoming when shcoolkids just have their gadgets access and think this is cool. This is not cool. We need to move. Please revert to classic or at least let me choose, this is terrible! -- (talk) 02:49, 19 January 2023 (UTC)Reply[reply]
what (talk) (talk) 03:23, 19 January 2023 (UTC)Reply[reply]
The fact they hid the possibility of reverting behind an account is mind-blowing. Such a disrespect for the millions of unlogged users... (talk) 03:50, 19 January 2023 (UTC)Reply[reply]
  • If you think you are a better web designer, apply to the Foundation. Calling them idiots is probably not a good way to get the job. 331dot (talk) 16:22, 19 January 2023 (UTC)Reply[reply]
    Who said I was? If you read to comprehend, you might notice that I am in fact specifically criticizing the utter refusal to care about the vast majority of users. (talk) 22:05, 19 January 2023 (UTC)Reply[reply]

Couple of questions

Okay, so this surprised me as the most recent RfC I saw seemed to not draw too much support, but I'm having a look at it. A few immediate things jumped out to me. I've read through above and haven't seen them:

  • Why does the default left side bar width differ for pages with no ToC? It's really weird for it to jump left and right. It doesn't jump too far, but it's disconcerting as I always look to the same place to start reading (well, I look to the 2010 place, but I'm trying to find a new place to look by default now and the jumping left and right is not helping with this).
  • What determines if the floating top bar appears or not as you scroll down the page? Sometimes it does, sometimes it doesn't. I think being in a diff view stops it, but I can't figure out the other reasons. It's not appearing on this page for example.
  • Why is the talkpage yellow bar notification now weirdly off to the right and on a separate line completely? There is a huge black space where it used to be, so it looked like a bug at first (before I realised the skin had changed), so I'd just like to check that this is the intended design.
  • If I click "hide" on the table of contents, the left screen goes totally blank. Where is the "show" button? (And for that matter, since there is now functionality for floating elements to move down the page with you, why don't the previous left-bar links also float down?)

Thanks for any assistance, pings are fine. CMD (talk) 01:46, 19 January 2023 (UTC)Reply[reply]

@Chipmunkdavis I just did some checking and yeah, the ToC moves to a collapsible menu at the top of the screen upon clicking "Hide". Awkward. Aasim - Herrscher of Wikis ❄️ 01:51, 19 January 2023 (UTC)Reply[reply]
Where is the menu at the top of the screen? I can't find it. CMD (talk) 02:54, 19 January 2023 (UTC)Reply[reply]
  • Also, a much more serious problem: the watchlist icon (took me a few seconds to find it) is laggy? When I click on it nothing happens for ages. I thought it was broken and was typing this up saying that until it suddenly worked (deleting the early version of this comment). When I ctrl+click it the new tab opens immediately. (Also the watchlist icon is quite small and feels mobile-optimised, does anyone have a css fix to make it a wider button that I can click on with less precise mouse movement? Much obliged if so.) CMD (talk) 02:00, 19 January 2023 (UTC)Reply[reply]
  • Screen width is different in diff view? I'm not sure what purpose this serves. CMD (talk) 02:54, 19 January 2023 (UTC)Reply[reply]
  • My talkpage and contributions are in a submenu now! The talkpage at least should definitely be there by default, it's one of the most important buttons a new user can have! (I'd like contributions too, but hopefully that can be fixed with a personal css. Talkpage is more important than userpage!) CMD (talk) 04:32, 19 January 2023 (UTC)Reply[reply]
    Hey @Chipmunkdavis, going to try to answer your questions:
    • The layout shifts because when there is no ToC we center the content. As you may have already seen below, once we do this should no longer (or extremely rarely) be a problem.
    • The sticky header appears on a per-namespace basis, which thus far as been determined case-by-case by asking ourselves: would people benefit from having the various tools (talk, history, etc.) available at all times on this page? Open to feedback about that!
    • Think I need a screenshot to clarify what you're seeing with the talk page / yellow bar notification.
    • That will be addressed somewhat by A few people have asked why we don't just collapse the ToC in place. We tried several approaches to that. First using a text label, which then ends up taking up a bunch of horizontal space and kind of defeats the purpose of collapsing the TOC at smaller screen widths (because you want that space back for the article). Then using an icon which felt like it was just kinda randomly floating on the side of the article. So we ended up putting it by the article heading.
    • Hm, I've never experienced that issue with the Watchlist and can't reproduce it. Are you still having that issue? I can't think of any obvious reason why it would be happening (aside from a slow internet connection or some kind of temporary slowness with our servers?).
    • Several community members asked us to remove the 960px max-width for certain special pages, particularly log-type pages (History, Recent changes, Related changes, etc.), in order to reduce line-wrapping. This was well before we added a full-width toggle. Maybe that's what you're experiencing?
    • Right so we've heard a lot of different opinions about which links should be outside vs. inside the menu. I don't have a very satisfying answer other than the most requests we've heard so far have been for the watchlist.
    I hope that is helpful : ) AHollender (WMF) (talk) 22:16, 19 January 2023 (UTC)Reply[reply]
    @AHollender (WMF): Thank you for the kind replies. If you don't mind I have some further feedback/questions
  • The layout shifts because when there is no ToC we center the content: I don't understand this. The most obvious effect of the new skin is that the width is fixed, so it not? There are quite a few internal pages that lack ToCs completely as they lack sections, I don't understand why them jumping to the left is considered useful. The left column should be the same width with or without a ToC in the article.
  • The sticky header appears on a per-namespace basis Not all tools will be useful all the time, but one of the tools is for example the edit this page button, which seems almost always useful to have around? Add/remove from watchlist is useful across all spaces, and page history is nice to have around.
  • talk page / yellow bar notification I will leave a talkback notice on your talkpage.
  • taking up a bunch of horizontal space and kind of defeats the purpose of collapsing the TOC Okay, the heading icon. I had zero clue that was there at all, it is not intuitive. I don't feel the indicator will do too much, given it appears to only be visible if you are for some reason at the very top of the page. I don't understand the horizontal space comment at all, the amount of horizontal space does not change at all when collapsing the ToC (you just get a black void of the same width). There is an entirely separate button (near the Wikipedia logo) to collapse the left space, which is separate from the ToC hide button. What am I missing here? A randomly floating icon seems a very good solution, as the ToC is randomly floating so this would be a clear thematic and visual connection.
  • On the watchlist, if it's a slow connection to servers it's unique to the watchlist, as every other button/link works fine. Still happening, but if you say it's not a skin bug then not for this page.
  • remove the 960px max-width for certain special pages This is great for viewing the diffs themselves as they are in two columns, I do not understand why the article preview below the diffs is not constrained. One point of diffs is to see how the edit affected the article, this change breaks that partially by ensuring the diff view will not show how the article looks for most people.
  • which links should be outside vs. inside the menu Watchlist good! Definitely outside. Losing the talkpage link though, that's a really odd one. They are a crucial part of the community (talkpage links are the one thing that is mandated in all editor signatures for example), and should not be hidden away for new users.
Hopefully this is a good location for such feedback. Thanks again for the replies. Best, CMD (talk) 01:38, 20 January 2023 (UTC)Reply[reply]

Any way to hide contents by default?