Can $890,000 Make MediaWiki Useful?

Written by Steven Walling / June 12, 2009 4:30 PM / 11 Comments

Late last year, an $890,000 grant was awarded to the Wikimedia Foundation (the non-profit behind Wikipedia). It was dedicated solely to a new Usability Initiative for improvements to MediaWiki.

Now, the first designs and prototypes have been made public.

Though only a fraction of what's in store, the work done so far is showing real promise not just for Wikipedia's future, but for MediaWiki in all instances. In a down economy, a new-and-improved MediaWiki could likely compete with more expensive and cumbersome enterprise collaboration solutions.

Enter the Usability Initiative

The Wikipedia Usability Initiative exists to draw in new volunteer editors to the free encyclopedia by making MediaWiki more user-friendly. The first step was an exhaustive study that revealed the weaknesses that make the current version of the software intimidating and frustrating to newcomers.

Unveiled June 2nd, the initial designs are of basic improvements to skinning, navigation tabs, search results and the edit toolbar. They're currently up and running on prototype sites, and should be available as an option in Wikipedia user preferences by July.

What the Future Holds

The first mock-ups may not exactly be mind-blowing, but they're just the beginning. We spoke recently with Usability Initiative Program Manager, Naoko Komura, and she outlined some bold next steps. Concepts being explored include easier media uploading, a real-time preview system, and (most importantly) reducing or hiding the extremely complex wiki syntax that creates tables and templates in MediaWiki.

One idea that's not being entertained is moving away from syntax altogether.

Nearly all enterprise wikis have implemented a WYSIWYG editor. Even ones that began with MediaWiki, like Mindtouch, quickly replaced wiki syntax with XHTML. The Initiative has attempted to draw on the experience of companies like Mindtouch, whose CEO, Aaron Fulkerson, told ReadWriteWeb he was impressed with the Initiative, but that he felt, "wiki text will always and forever be inferior to XHTML."

WYSIWYG editors are already available for MediaWiki through extensions, but the potential for corrupting the data that makes up Wikipedia's encyclopedic content is very real. Avoiding that scenario is primarily what lead the Initiative to discount a switch to WYSIWYG, at least within the scope of the project.

MediaWiki in the Enterprise Again?

Even with the retention of wiki syntax, the Usability Initiative may see a revival of not just Wikipedia's openness to contributors, but MediaWiki's suitability for more business-like use.

Licenses and subscriptions for enterprise collaboration platforms are looking more expensive than ever. In 2009, more companies are clearly seeking to upgrade and improve software they already use, rather than start fresh. Despite its quirks, organizations like Intel have maintained large and productive installations of MediaWiki for years.

The Usability Initiative's core goal is to improve the software with potential Wikipedians in mind. But an unintended consequence of the project may very well be that MediaWiki regains respect in the enterprise.

Comments

Subscribe to comments for this post OR Subscribe to comments for all ReadWriteEnterprise posts

MindTouch has converted many thousands of Mediawiki installations to MindTouch with our Mediawiki converter. The most recognizable and publicly viewable Mediawiki converted to MindTouch is http://developer.mozilla.org , which is obviously very high traffic and massive in terms of registered users and amount of content.

By migrating to MindTouch from Mediawiki Mozilla collapsed ~20 Mediawiki instances into a single multi-lingual polyglot MindTouch instance. What I mean by polyglot is that MindTouch can be configured for many languages in a single instance; thus allowing cross language search, relationships between articles in different languages and more... The maintenance is significantly lower in cost and the MindTouch platform is remarkably more extensible and customizable. Of course, there's also all the benefits of native XHTML and a 120 method REST API that MindTouch offers.

Mike Shaver, Mozilla VP of Engineering was interviewed about the migration to MindTouch (from Mediawiki) here: http://www.ddj.com/architect/207600577 Also, @sheppy (Mozilla Dev Center Doc Master) blogs about his work on MindTouch frequently. Finally, other Mozilla projects have migrated as well, such as http://wiki.songbirdnest.com .

Posted by: Aaron | June 12, 2009 9:54 PM

The choice of headline for this story is bizarre. "Can $890,000 Make MediaWiki Useful?" includes the direct implication that MediaWiki is not useful as it is now... news to the vast numbers of people using it every day, I suspect. I guarantee you that far more people find MediaWiki useful in its current form than have ever heard of ReadWriteWeb. A touch less arrogance might go a long way.

Posted by: Matt Andrews | June 13, 2009 12:21 AM

That's quite a lot of money!

Posted by: converter | June 13, 2009 1:07 AM

Ha, let's just see.. keep us posted RWW re developments. With this kind of funding, I'm sure there's a love of things to come.

Posted by: ITrush | June 13, 2009 6:16 AM

In terms of usability, wikis are falling further and further behind the rest of the rest of the web. It is a trend that doesn't bode well for the entire wiki ecosystem. So I'm delighted to see the good work that Naoko and the rest of the Wikimedia Usability team are doing.

I'd also like to point out the good usability work that Wikia has been doing. Wikia has integrated the FCK WYSIWYG editor into their new wikis and it works impressively well. You can edit this page:
http://twilightsaga.wikia.com/wiki/Renesmee_Carlie_Cullen
to see how it compares with the old wiki syntax version. That said, the Wikia folks have not yet integrated this into their older, more popular wikis like the Star Wars Wiki and the World of Warcraft wiki. Putting a robust WYSIWYG on legacy Mediawiki wikis is the next challenge. We've been experimenting with adding the FCK WYSIWYG editor on wikiHow, but there are plenty of challenges that are currently holding us back.

Posted by: Jack Herrick | June 13, 2009 9:18 AM

Jack: you hit the nail on the head. Putting a robust WYSIWYG editor like FCK on legacy MediaWiki is absolutely the next challenge. Thanks for chiming in!

Matt Andrews: Inaccessible software is not useful software. I've made more than 30,000 edits to Wikipedia, and several thousand more to other MediaWiki sites. There's no doubt in my mind that the software is functionally inaccessible to most people. If you'd like to see hard evidence, I suggest you take a look at the usability study I mention in the article, wherein even a young person says that trying to edit makes them feel stupid.

Posted by: Steven Walling | June 13, 2009 2:52 PM

The new prototypes and mock-ups don't appear to be much of an improvement. Instead of trying to improve the usability, the Wikimedia Foundation should partner with leading providers that have already addressed the challenge.

The Foundation should focus on much larger initiatives like ensuring the MediaWiki content is accurate and up to date and that access is given to 3rd world countries so that we hear their voice.

Posted by: Mark Fidelman | June 13, 2009 6:03 PM

Lots of money. i think $890,000 Make MediaWiki Useful .and use $$$$$ to explore marketing .and ad .

Posted by: Alex | June 14, 2009 5:12 AM

Ha, let's just see.. keep us posted RWW re developments. With this kind of funding, I'm sure there's a love of things to come تعب قلبي

Posted by: دردشه | July 15, 2009 3:58 PM

Mark: If the Foundation wanted to use MindTouch's solutions, they could just do so without having to "partner" with you (whatever that means); they could just use your code, as you did with MediaWiki.

Posted by: GreenReaper | November 4, 2009 10:25 AM