Stuffing the Internet in a box and shipping it to schools in Africa


This page in:

there are multiple options for moving forwardOver the past decade or so, increasing numbers of groups have been working on answers to variations of the following question:

How can the wealth of educational resources on the Internet be brought to the majority of African schools that are today 'un-connected'?

While the Internet has not wrought the similar types of profound, broad societal changes in Africa that it has in other parts of the world, the connectivity landscape in Africa is in fact changing very quickly in many places (for the better!), with (for example) macro-level announcements about progress with new fibre optic cables coming on what seems like a weekly basis.

(For those who like such things, here's a great map to track technical progress in this area.  For acronym fans, here are links to announcements about some of the major backbone connectivity initiatives in Africa: Glo,  RCIPEASSyTEAMS, Seacom and LION2.) 

Earlier this year the total number of mobile phone subscribers in Africa (over 300 million) passed the total in North America and, while access to the Internet via mobile phones is still low across the continent, it is growing quickly.  In Nigeria, for example, published reports now have mobile phones as the primary access device to the Internet in Africa's most populous country.  There is even increasing talk (and some action) of connecting African educational institutions to the 'cloud' in various ways. 

That said, it also undeniable that improvements in connectivity are not coming fast enough, or at a high enough speed or quality, or cheaply enough, for all citizens and schools, especially outside major population centers -- and won't any time in the near future.

For years, groups have been attempting a variety of grassroots solutions and approaches across the continent to provide better options for schools than simply waiting.  Tools such as loband, for example, were designed to strip out the 'extra' stuff in web pages so that they download quicker over slow connections.   Some groups have supported training activities related to bandwidth management and optimization, recognizing that many universities (for example) could do a much better job of managing the current bandwidth that they already have. Other have advocated for more attention to designing web pages for faster access in low-bandwidth contexts.

While the connectivity environment for many may not be changing fast enough, one thing that is changing very quickly for everyone is the cost of data storage.  Back in 2007, a vice president at Google noted that, since 1982, the price of data storage has fallen by a factor of 3.6 million, and that "if this trend continues, and the cost of storage continues to decrease, we estimate that somewhere around 2020, all the world's content will fit inside an iPod, and all the world's music would sit in your palm as early as 2015."

Recognizing this trend, some groups have championed approaches to 'cache' content locally and/or to provide content on physical media like USB sticks or DVDs (one well-known example in ICT4D circles is the Wizzy Digital Courier), physically delivered to remote communities (this is affectionately known as the 'sneakernet', as a nod to the early sneaker-clad 'geeks' who would walk floppy disks between unconnected computers and computer networks). 

The eGranary Digital Library has been involved in different aspects of these sorts of activities in Africa for almost a decade.  Cliff Missen of the University of Iowa (USA), the driving force behinf the eGranary project, recently stopped by the World Bank to share lessons from experiences across Africa and the results of a recent evaluation report. For those unfamiliar with the eGranary initiative, here's a short description from the project web site:

The eGranary Digital Library - also known as "The Internet in a Box" - provides millions of digital educational resources to institutions lacking adequate Internet access. Through a process of garnering permissions, copying Web sites, and delivering them to intranet Web servers INSIDE our partner institutions in developing countries and other places around the globe, we deliver millions of multimedia documents that can be instantly accessed by patrons over their local area networks at no cost.

The eGranary currently includes about 14 million documents across multiple sectors (this equates to about 2 terabytes of data), with a heavy representation of resources from the health sector.  Included in this document cache are full text papers from over 250 academic and scientific journals (interestingly, it turns out that most places don't use them all that much; by far the most popular resource is ... the off-line version of Wikipedia).  It has been recently re-architected, so that it can work more seemlessly with multiple hard drives, and new security tools have been introduced. Researchers at the University of British Columbia (Canada) are studying the impact of programs like the eGranary helping to promote information and digital literacy (in Uganda and elsewhere).  Currently the eGranary is installed at over 300 educational institutions in Africa and elsewhere, with ambitious plans for expansion.

One criticism that I hear of projects of this sort goes something like this: 'Stuffing the Internet in a box and shipping it to Africa -- do you think that all we are need is your 'knowledge', and that we have nothing to contribute and share ourselves?' 

This is a sentiment that the eGranary project has apparently heard, and content creation and management tools like Wordpress and Moodle are now bundled as part of the eGranary as a way to help enable local content creation and curation.  However, as the folks at eGranary will be the first to tell you, providing the tool alone isn't usually enough to catalyze local content creation -- a variety of other things need to be done as well.  There is no reason, of course, that initiatives like this can't be used as triggers to help with the creation/digitization and dissemination of digital content in local communities. One thing that the eGranary folks highlight, and which they are clearly passionate about, is the need for additional capacity building for librarians throughout much of Africa around issues related to digitized resources.  They note that it is not enough to help support better access to, and the development, of digital resources if sufficient energies are not also expended around creating related competence and expertise around the curation of such resources, so that users can find them in ways that are useful and relevant. 

Reflecting on some of the lessons emerging from the eGranary and other projects of this sort, a few questions come to mind:

  • Given that storage costs are falling more more rapidly than bandwidth costs, how might off-line storage tools of this sort be considered as a matter of course within larger, more holistic considerations of access to information that are an important driver for action to bring better connectivity to schools in poor connectivity environments?  
  • There is a trend toward increasingly 'open data' around the world.  How can projects like the eGranary take advantage of these, incorporating newly released data sets into the document warehouse in ways that are useful?  How, for example, might World Bank open data be marketed and made more attractive to be included in off-line document and data respositories like the eGranary? 
  • One of the clear messages from previous versions of the Africa Tertiary Interner Connectivity Survey (ATICS) was that there was a great need for African universities to better manage the existing (expensive, often poor quality) Internet connections.  Wouldn't it be great if bandwidth management and optimization tools could be integrated into resources like the eGranary, as a way to help expose network administrators to them in advance of the arrival of more copious broadband connectivity?
  • As storage costs continue to fall, what are the opportunities for things like an 'eGranary on a USB stick' -- or even better, given their increasing ubiquity even in some of the most remote or poor communities around the world, on a mobile phone?

For what it's worth, another criticism I hear in some quarters is that creating these sorts of solutions for places that do not have reliable Internet lessens the urgency to bring connectivity to them, relegating certain segments of society to 'sub-optimal' solutions compared with other, more privileged groups.    (In a similar vein, some people have criticized an earlier post on the EduTech blog about the so-called One Mouse Per Child project -- which enables up to 50 children to use the same computer at the same time, each with her own mouse -- complaining that giving prominence to such 'gimmicks' distracts us from more important goals, like providing equal access to all children to educational computing devices.)

While sympathetic to such sentiments, I don't see these as binary, 'either/or scenarios'.  In the words of Voltaire, there is a danger that 'the perfect can be the enemy of the good'. Given the opportunity to have an affordable, always-on broadband connection, I doubt few places would opt for tools like the eGranary as a 'replacement' -- nor would the use of tools of this sort lessen the degree to which institutions advocate for better, cheaper, and more reliable bandwidth.   Tools like the eGranary, and others covered in an earlier EduTech blog post (A (digital) library ... in your pocket?) perhaps should not seen as detours from such larger aspirations and needs, but rather as useful (and inexpensive!) way stations along the various developmental paths that connectivity-poor education institutions of various sorts in Africa (and beyond) are following.

Note: The public domain image used at top of this blog post ("there are multiple options for moving forward") is of the AutoNov1, a 1970s concept car creation of the late Nigerian professor, engineer and inventor Ayodele Awojobi that could be driven forwards or backwards with all four pre-existing gears.


Michael Trucano

Global Lead for Innovation in Education, Sr. Education & Technology Policy Specialist

Wayan Vota
October 29, 2010

To those that say great tools like eGranary lesson the desire for "real" broadband content, I challenge them to ask the users of such offline data repositories.

They will quickly find, as I have, users who were scared of online content grow familiar and comfortable with it by using this "safe" offline medium first. As the eGranary is mainly a replica of the Internet, its a controlled first step. But just that - a first step. Users quickly wonder what else is out there, and expect actual online experiences to be as fast and rewarding as locally hosted sites. Or to put it another way - when eGranary is your first "online" experience, slow dial-up just isn't gonna satisfy.

So kudos to eGranary & others like it (SOS Childrens Village Wikipedia, Moulin) for helping wet the appetite for online knowledge.

Edward Mokurai Cherlin
October 26, 2010

Yes, One Mouse Per Child is a silly gimmick. Why no mention of a genuine campaign to bring education to all of Africa and everywhere else?

What is this nonsense? "That said, it also undeniable that improvements in connectivity are not coming fast enough, or at a high enough speed or quality, or cheaply enough, for all citizens and schools, especially outside major population centers -- and won't any time in the near future."

I deny it absolutely.

First, how fast is "fast enough"? How long did it take to get railroads across Europe, the Americas, Asia, Africa? Telegraph? Telephones?
Optical fiber is moving much more quickly. Not only are multiple cables being laid all around Africa, but spurs have reached several landlocked countries, and there are plans in motion to connect all the rest.

Second, we are talking about broadband to the village. Within some of the newly-connected countries, most notably Rwanda, there are plans to get all cities and towns connected, and discussions about the best combination of technologies to reach all of the villages, with WiMax an important contender.

Thirdly, you write as though ICT were an expense rather than an investment. The economic growth from universal broadband, particularly for education, will repay the initial costs and ongoing maintenance many times over. All that is required is the political will to continue to bring the poor out of poverty, rather than to further enrich the already rich.

Lastly, advances in technology generally happen about as fast as possible, when they are possible at all. It's no good whinging about how fast it is. Just do what you can to move it along, and to help others to take advantage of it.

For example, we know that fiber optic and wireless communication are growing out at great rates. We know that computers now cost less than printed textbooks in all but the poorest countries, those where the textbooks are completely inadequate. So we need to get cracking on free e-learning materials in a hundred or more languages, by engaging the already educated who speak those languages, and not leaving it to foreign "experts".

Michael Trucano
October 26, 2010

Hi Edward,

Many thanks for your comments.

I guess we can agree to disagree about what 'fast enough' means. The speed of the spread of ICTs throughout the world is certainly unprecedented in recorded history. That said, I find that this fact is often of little solace to those who still don't have them (and I would hardly characterize such folks as 'whingers').

I certainly don't mean to write, and I don't think I do, as if 'ICT were an expense rather than an investment'. This certainly isn't my view of things. (I will note parenthetically that it appears to be the view in practice of many ministries of education, despite their public pronouncements.)


Cliff Missen
November 12, 2010

Almost half of those who have eGranaries also have some Internet connectivity, but it's painfully slow, unreliable, and won't scale to serve hundreds, even thousands of users.

Given the high cost of Internet bandwidth, savvy librarians adopt mutiple strategies to deliver information to their patrons, asking the key question: "what's the best way to serve my patrons using the least amount of Internet connectivity?"

Those who still cling to Internet-centric "oh, let them buy bandwidth" sentiments need to stop and ask themselves: why would we expect the world's poorest people to use the most expensive medium to access static information? What has changed in the last year or two that renders obslete what we've known about nursing, mathematics, and rural development?

A hybrid system makes critical distinctions about what information can be stored and served locally, what information requires frequent updates, which communication services can be rendered asynchronously, and which information access and communication requires real-time connectivity.

As nearly one-third of American households have demonstrated: people living at or below the poverty line do not belive that spending on high-speed Internet is more important than meeting other, more critical needs. For all of the current, laudable efforts to lay fiber cables across Africa (a land mass three times the size of the United States), it will be generations before redundant connections and competition drive the prices down to anything resembling what is known in richer countries. And it will be at least as long before a growing middle class can support such infrastructure. For the forseeable future, Internet connectivity will serve the elite.

So it remains for us to focus on both paths: grow the local capacity to afford and use bandwidth; and find immediate ways to improve access to information and education for everyone.

Think hybrid...