Grants:IdeaLab/Make Wikipedia beautiful

From Meta, a Wikimedia project coordination wiki
Jump to navigation Jump to search
Make Wikipedia beautiful
Venus of Arles Louvre Ma439 n01.jpg
Wikipedia skins are old, cluttered and ugly. They are not up to today's web standards. A beautiful skin would be welcome by both readers and editors, and may invite new users to join.
Hex icon with lightning white.svg
idea creator
Hex icon with flask white.svg
Hex icon with circles white.svg
Hex icon with hand white.svg
this project needs...
Hex icon with hand black.svg
Hex icon with circles black.svg
created on21:46, 12 March 2015 (UTC)

Project idea[edit]

What is the problem you're trying to solve?[edit]

The problem is that all the available Wikipedia skins are old, cluttered and ugly. The current default skin, Vector, was developed in 2009 with an emphasis on usability and accessibility, not beauty. That's more than 6 years ago, a long time for the Internet, and it has hardly changed since then. The other available skins are older and uglier still. Furthermore, any developer who has taken a look at the mess that is the Vector skin code (or any other) would agree with the need of a change without even seeing the rendered skin. New developers find it very difficult to understand and contribute to the code, as it doesn't even comply with the latest skin standards at

There is another strong reason for wanting to develop a new and beautiful default skin. When I ask people if they ever edited Wikipedia, several times I was surprised to hear that frequent readers had never even seen the Edit button. According to this research, more than 18% of users never knew they could edit Wikipedia. This is not their fault, it is a design flaw. If we want to rise the number of editors, then not seeing the Edit button should be made nearly impossible. If we change the default skin, then current editors will continue to edit, but those who didn't know about the Edit button will have a new chance to see it, and so start contributing. According to this research, making the history button more prominent almost doubled the clicks to it. Considering that the History and Edit buttons are right next to each other, similar results could be obtained by just making the Edit button more prominent.

This proposal started as part of the Inspire Campaign because I suspect that female participation will rise if we make a better and prettier skin. This may well be true, but it is hard to prove beforehand, even though some other users agreed with this intuition. In any case, even if improving the skin doesn't significantly increase female participation in percentage, it may very well increase it in absolute terms along with general participation, and even if general participation doesn't increase, still the existing community would greatly appreciate a more beautiful skin to embellish the content created by their hard work.

What is your solution?[edit]

The solution is to develop a new default skin, with a strong emphasis on beauty and aesthetics, not only usability and accessibility. Reading and editing Wikipedia should be a joy to the eyes. The VisualEditor moves us a step forward by not requiring us to see that ugly wikitext every time we edit, but still the skins leave tons of room for improvement.

Take a look at to get an idea of how different Wikipedia may look.


The goal is to develop a new and seriously beautiful MediaWiki skin to be deployed as the default skin for Wikipedia. Below are some probable features of the new skin. Please feel free to add new features or comment on existing ones.

  • Full integration with:
  • Generous spacing.
  • Hide elements when not needed. For example only show buttons to edit a section when hovering over a section. This would not only make the interface cleaner, but it would also make the [edit] buttons more conspicuous, by popping in the face of the user, while not being annoying, as we've already learned from other sites that do this, such as Facebook, YouTube and Google.
  • Fully responsive, maybe even mobile-first.
  • Full integration with the math tags, maybe by using the same font. The success of the TeX styling in the scientific community should not be ignored, but harnessed.
  • Clean code with unified conventions for class naming and ids (in the current Vector skin, some classes and ids use camelCase, others-have-dashes, some have mw-prefix, others don't, etc).
  • The Vector skin has an obvious design flaw. The action tabs (Read, Edit, History, Move, etc.) are at the same visual level as the namespace tabs (Article and Talk), which is absurd given that the former depend on the latter, meaning that if we are on the Talk page, then the Read tab refers to the Talk page, whereas if we're on the Article page, it refers to the Article page, and so on with the rest of the actions. Also, because both the namespaces and the actions are presented as tabs at the same level, their various combinations become visually confusing, especially for new users. It would make much more sense to have the actions next to the main title of the article, in the same style as the buttons for editing sections, or even better with icons rather than text, such as a pencil for editing, an eye for reading, etc. (aligned to the right, of course). These suggestions may be used for improving the current Vector skin as well as for designing the new one.

The success of the skin can be quantified in the following way: we activate the new skin for 50% of the new IPs and those IPs that don't have any edits. Then after a month, we compare the number of edits made by the IPs that edited Wikipedia with the new skin, with those that edited Wikipedia with the old skin.

Get Involved[edit]


  • Felipe (talk) 22:06, 12 March 2015 (UTC) - I'm a MediaWiki developer and have done several skins already. I could help implementing this proposal if it gets support.
  • Volunteer I could do some graphic design. Supaiku (talk) 20:40, 31 March 2015 (UTC)
  • Volunteer review. 21:51, 26 October 2015 (UTC)


  • I love that people want to do this, though I'm cringing in anticipation of the epic pushback. I think a successful effort would need some plans to deal with the inevitable resistance, and I'd love to hear your thoughts on that. -- Skud (WMF) (talk) 00:58, 18 March 2015 (UTC)
I think that the first thing would be to gather proof, evidence or at least more opinions saying that this idea may actually work. Else it's just our intuition. How can we do that? A poll? Where? --Felipe (talk) 01:40, 21 March 2015 (UTC)
I also think that we need some professional designers' opinion here. As a commoner I can't imagine any skin that's less disturbing than the current one. After all, Wikipedia is Wikipedia because it's an important content provider not because it looks cool. Fluffs and vivid colors don't really belong to an encyclopedia, where the focus should be the text not the UI. That said, a UI redesign, if well-done, is welcome. Ahyangyi (talk) 04:43, 23 March 2015 (UTC)
  • Making the site feel look more beautiful and less hardcore tech-y will go a long way to feel more accessible to all kinds of people. This should especially benefit women, who are usually more self-critical of their own capabilities. Goldfishbutt (talk) 06:15, 21 March 2015 (UTC)
  • Making the default skin something prettier than "Vector" would make reading easier for new users and this would make them more likely to contribute. Sure, it will cause contributor friction, but that is mainly because Wikipedia has accumulated a group of editors who enjoy (or tolerate) ugly websites. Falkirks (talk) 03:06, 22 March 2015 (UTC)
  • yes it has to be more colorful! add more pink and pics of cats and some more of what women find attractive ...Sicherlich Post 18:25, 23 March 2015 (UTC) stereotypes? .oO
Do I detect a note of cynicism in your post, Sicherlich? Wordreader (talk) 08:41, 25 March 2015 (UTC)
of course. Because; WP more pretty; yes, why not. ... but I doubt, that it raises more women. - and to research what is more beautifull you would need to ask people not only in the US but in a lot of countries ....Sicherlich Post 06:53, 29 March 2015 (UTC)
  • Wholeheartedly endorse! I have long felt that Wikipedia is clunky-looking and dated. Streamlining pages and making them look less visually polluted will help keep the site looking as relevant as the content is. Design must always be with an eye to making the site more user-friendly and instantly understood, though. Well developed wayfinding through an Internet site is even more important than through a physical building. And for people like Sicherlich and men who think aesthetics has to do with lace and satin ribbons, they should look at the work of Phillip Johnson's Glass House, Tadao Ando's Azuma House, and Barragán's Gálvez House. Those are beautiful and not a ribbon in sight. Translate that feeling into electrons. Wordreader (talk) 08:41, 25 March 2015 (UTC)
  • Brilliant! Supaiku (talk) 20:39, 31 March 2015 (UTC)
  • Considering the recent changes made to this Idea, I will provide my tentative endorsement. I still consider there to be some problems with this Idea, in particular with the use of highly subjective language, but I believe the underlying objectives are worthwhile to endorse. Hopefully, this Idea will expand and develop into a more comprehensive project, one which helps initiate some major aesthetic changes to Wikipedia. –Nøkkenbuer (talkcontribs) 08:09, 9 April 2015 (UTC)
  • Remember Flow has gone, and don't forget that Vector was deliberately written to encourage more readers in the pre Vector era when we weren't short of editors. Monobook is more useful for editors. A skin improvement is certainly possible and probably overdue. But you have several circles to square and we may need multiple skins or one that self adapts. Also remember that people get extra buttons as they gain userrights, an uncluttered skin may look very different once tabs for block, delete, protect etc are added. Ideally a skin should give people one click access to the tools they use most, and tempt people to the next step in editing. One of the drawbacks of Vector for editors is that it hides many options behind irritating submenus without even the saving grace of meaningful symbols like a pen for editing, but readers probably consider that an advantage. That said there was some research a while back that showed people blank out the edit box because of familiarity, and simply changing it a little every few weeks should in theory solve our editor recruitment problems. WereSpielChequers (talk) 20:50, 8 March 2016 (UTC)
  • This idea describes a well scoped research project that we discussed on the Growth team (now defunct). I think that this should happen and I'd like to support it with behavioral analysis of the effects of a more prominent edit link. --EpochFail (talk) 16:11, 1 April 2016 (UTC)

Expand your idea[edit]

Do you want to submit your idea for funding from the Wikimedia Foundation?

Expand your idea into a grant proposal