There are updates to this page that haven't been applied because you've entered text. Refresh this page to see updates.
Hide this message.

Why doesn't Wikipedia innovate with regards to their user interface?

For all intents and purposes Wikipedia's interface hasn't changed in any noticeable since its creation. Why doesn't the Wikipedia team try to see if they can make it better?
Brandon HarrisBrandon Harris, Senior Designer, Wikimedia Fou... (more)
261 upvotes by Bradley Voytek, Rob McQueen, Michael Wolfe, (more)
This question is surprisingly more complex than you'd think.  The short answer is "we are working on it" but that's kind of a cop-out.  The longer answer follows:

First and foremost, it is non-trivial to make any kind of changes to a top-five site that has such a consensus-driven community.  And I mean every change, no matter how minor.

Wikipedia is not a privately held company.  Larger sites (e.g., Facebook) can and do make radical interface changes without consulting the user-base but the reason that this is possible for them is precisely because they are privately held.

The Wikimedian community has a strong sense of ownership for the product, and rightly so: these are the people who write the articles.  Thus, the community believes (correctly, in my opinion) that they should be consulted on changes to the interface.

This process takes a great deal of time. It is rather glacial in its nature.

That said, the Foundation has taken a position that allows it to make what we think are minor or important software changes as needed (such as deploying WikiLove, Feedback Dashboard, or the upcoming New Page Patrol interface).  Mostly, these changes are subtle or invisible to "readers" (and most "editors").

Second, writing software is hard. You may think it isn't - especially if you have a passing history doing so - but you must understand that every Wikipedia feature must:

  • Be able to be localized into over 370 languages.  This is an enormous task.
  • Be able to run on pretty much every browser ever made.  Until the usage of IE 6 drops below .03% globally, we have to support it - which nearly always doubles development time. (We were only recently able to remove IE 5 for Macintosh from the support matrix)
  • Be highly scalable. The Foundation has a shoestring budget.  We do not have thousands and thousands of servers with which to generate every page; instead we make very smart use of caching systems.
  • Be extremely private.  Our privacy policy prevents us from doing any kind of tracking of our users.  This makes developing features difficult at times.
  • Be able to be used anonymously. The number of people who use the sites who do not have user accounts is astronomical.  This limits our focus.

Third, we have fewer programmers.  This is because of several reasons:

  • We are a non-profit. As such we pay bottom-of-scale.  It is difficult to attract programming talent when they can make 30% to 50% more money anywhere else.
  • We prefer to employ the Free Culture-aligned.  We prefer to hire people who are mission aligned or dedicated.  A love of the work and what it stands for is extremely important to us.
  • We prefer to think globally.  You'd be surprised how difficult it is to develop things meant for a global audience.  As such, we have determined that hiring from all cultures helps us to understand our weaknesses.  We're not the best at this, mind you, but we're working on it.

That said, we are innovating, we're just doing it very slowly.  Here are some projects you may be interested in:

Neil KandalgaonkarNeil Kandalgaonkar, ex-Wikimedia Foundation softwa... (more)
10 upvotes by Quora User, Andrew de Andrade, Brandon Harris, (more)
New concepts for mobile / tablet interfaces

High level planning on the future of MediaWiki - total parser rewrite, GUI editor

Some notes on the upcoming visual editor:

New features the WMF has looked into or is actually doing, including UI.
Steven WallingSteven Walling, Wikipedian since 2006, 40k+ edits.
10 upvotes by Tracy Chou, Quora User, Coleman Foley, (more)
Wikipedia is innovating. The Wikimedia Foundation (the non-profit behind the site) now has a permanent User Experience team and recently released the first major redesign of the software in years. There's more, including prototypes, at
Paul LumsdainePaul Lumsdaine, UX/UI Designer for Web and Mobile
8 upvotes by Tony Trupp, Ben May, Quora User, (more)
I think people are confused by the difference between simple UI design and pooly done UI design.  I feel that Wikipedia is a prime example of simple UI design. Just because the design is simple does not mean it is poorly done.  I feel that there are so many challenges with reshaping the UI to a more sophisticated or less antiquated design, and that wikipedia has wisely chosen to go with the slow and steady route. 

I dont feel that the interface is poor at all.  It achieves its prime directive of providing the user with quick access to a massive amount of content.  Yes, they are lacking in the bells and whistles, but that does not constitute poor UI.
Vipul NaikVipul Naik, Edited occasionally, now mostl... (more)
1 upvote by Quora User.
I think the previous answers, particularly Brandon Harris's answer cover all the important points but I'll add another one that I don't think has been addressed in sufficient detail.

A Wikipedia page is 90%+ the content of the actual article, and the interface components -- the search box, toolbar, navigation elements -- form a very small part of the page. The interface components are the only things that are controlled by the software (the way it looks is the choice of "skin" and the way it works depends on the underlying software).

So, to make a dramatic change in the appearance of the page, either (1) the content itself must be changed somehow, or (2) the proportion of the page devoted to content must be reduced. (1) would involve changing, or editing, all the millions of actual pages. (2) would be antithetical to the mission of Wikipedia.

The skin and software underlying Wikipedia have undergone significant changes (such as the change in default skin from Monobook to Vector, dramatic improvements in search, etc.) but these don't affect the overall look and feel much because a Wikipedia page is primarily content.

Another related point is that unlike, say, Facebook, the "content" part of Wikipedia is pretty much stored as text rather than in the form of structured semantic information. Facebook has a lot of structured semantic information about relationships and interactions and can use these to change the algorithms underlying what content to display and how. Wikipedia pages are displaying what's mostly plain text (with some markup) with very little semantic information that could be subjected to new and clever modifications in how it's displayed. The only things that can be changed to cause large-scale changes are the heavily used Wikipedia templates, but even these wouldn't have much effect.
Quora UserQuora User, Contributing mostly by trackin... (more)
7 upvotes by Steven Walling, Tom Morris, Jeff Forbes, (more)
Wikipedia works great IMHO. What's the problem ?

They focus on content and that's exactly what we need from them. UI is not an issue when it comes to reading.
Write an answer