Those bright folks at Geekcorps Mali have come up with another project that I’m vicariously proud of. (By “vicariously proud”, I mean, “I had nothing to do with it, can claim no credit for it, but am really glad people are still doing cool things in the name of the NGO I cofounded.“) The project is called Moulin – French for “windmill” – and it’s an interactive, offline version of Wikipedia.
The idea of offline wikipedias gets a lot of chuckles, including this cartoon from Iliad at UserFriendly. But the idea makes a great deal of sense: there are low-end PCs even in very poor parts of the developing world, like secondary cities in Mali. But because it’s extremely difficult to provision internet access in rural areas, these PCs go unused or underused. With a French-language Wikipedia on CDROM, this PC becomes a library of sorts for a community that’s otherwise information-starved.
It’s a neat technical hack as well – the Wikipedia articles, sans images, are compressed onto the CD, which also includes a lightweight webserver and software to unpack and configure on Windows, Mac or Linux. The resulting version of Wikipedia can be edited locally, and can be updated with future releases via CD.
I’m seeing an explosion of interest in content for the developing world, linked in part to the One Laptop Per Child project. As publishers and open content developers think about what content they might try to put on laptops, it’s worth thinking about what other sorts of open content we’d like to make available to developing world computers, both online and offline. Alex Weir, a former Geekcorps volunteer, has worked on a project called cd3wd – a CD-Rom of useful open content materials for the developing world, including a wealth of agricultural and technical information. The main limit to building such collections is finding content that’s available under open licenses.
It’s possible that enthusiasm for One Laptop Per Child will inspire some publishers to make their content available either under truly open licenses like GFDL, or under more constrained licenses like the Creative Commons Developing Nations license. But the impact of opening this content could go far beyond just the countries in the OLPC project. This content can be distributed using systems like Moulin, and in conjunction with inexpensive recycled computers (and, possibly, printing capabilities), could help fill the “library gap” that most developing world schools and communities face.
I am glad you posted about this project.
The cartoon, which I used in a presentation I made in Nigeria, did not so much told me how illogical it was to print Wikipedia, but rather reminded me that what we should aim at doing is not necessarily a great english speaking wikipedia to spray over all nations, but rather to propose content in the people mother language.
i think offline wikipedias have more potential than one can initially imagine. if one is installed at a location like a library, and the community is provided with proper training, the ‘wikipedia model’ could be reproduced locally, draw that small percentage of active writers, and lead tond an improvement of the wikipedia – say on subjects locally significant, but globally underrepresented.
the trick then would be, to ensure that this knowledge is at some point fed back to the online wikipedia. it must be possible to develop some tool enabling initial comparison of articles to determine, which one is better – based on word count, if nothing else.
i am wondering, will there also be cases where meshes of OLPC laptops will be disconnected from the online internet? if so, then this issue is even more crucial.
Anthere, I think one of the exciting things about the Moulin project is that it might help seed other language wikipedias, like the Bambara wikipedia that some Geekcorps Mali volunteers have already been involved with. Once people see the opportunity to create content, a logical next step is creating content in their mother tongues.
tarkowski, as you point out, this leads to questions of how you sync this offline content with the existing online content. The scenario you raise with OLPC is a very realistic one – there will certainly be some OLPC installations that have low or no bandwidth, and will largely use the mesh for local networking. Figuring out how to resolve edits from those communities with the larger wikipedia community isn’t just a tech problem – it’s a big social problem. Wikipedia works, in part, because of the social dynamics around editing. How do we continue and replicate that social dynamic with participants in conversations who might only be able to weigh in once a month? Is it fair to put up an article into the deletion process if the author might not be able to get online again to defend her work?
Moulin is très cool. Getting offline-editable versions of global collaborations to work is a useful meterstick for how mature a global project is — beyond the adolescent stage of a hosted global project is a stage of distributed organization which has a great deal to offer. Moulin will help provide Wikipedia’s excellent material to the offline francophone world. I hope it will also be another prod for those working on social and technical platforms to support asynchronous participation in Wikipedia and similar projects.
Also: recent difficulties with combining text from two ‘freely-licensed’ sources compels me to point out that the GFDL, and other Share-Alike licenses, are not quite examples of “truly open” licenses. I would very much like to see a version of Wikipedia available under a simpler license that removes barriers to incorporating such knowledge into other materials.
Great idea. One suggestion – include a little script that checks if the computer has a globally routable IP address, and if yes, groks the Wikipedia recent changes XML file and updates the offline content.
Comments are closed.