Africa Has Computers; What They Need is Software
Dr. Joel Selanikio has a great overview on the BBC about why writing software for cell phones is far more important that writing software for PCs when it comes to making useful applications for the developing world.
The question we should be asking ourselves, then, is not "how can we buy, and support, and supply electricity for, a laptop for every schoolteacher" (much less every schoolchild), but rather "what mobile software can we write that would really add value for a schoolteacher (or student, or health worker, or businessperson) and that could run on the computer they already have in their pocket?"...
Unfortunately, as of this morning a Google search for "educational software for Windows" got 41,300 results, while a search for "educational software for cell phones" got exactly 9 hits.
...
After all, who is more likely to come up with innovative software based on the centrality of the cell phone, a programmer in Silicon Valley surrounded by beautiful desktops and laptops, or a programmer in Nairobi who lives in a world in which almost all contact with the network is via cellphone?
The invisible computer revolution [BBC]
Previously: EpiSurveyor: Saving Lives with Open Source (our interview with Dr. Selanikio) [BBG]
the latest
latest episodes
He really, really oversimplifies the situation. A cell phone is NOT a handheld computer as much as some people think it is or would like it to be. The overwhelmingly vast majority of cell phone have neither the processing power, video power, storage capacity, or internet throughput potential to provide the necessary tools he suggests - regardless of the brilliance of the programing.
I disagree with his supposition that a common cell phone available most anywhere, much less Africa is somehow an alternative to a Teacher's version of a One-Laptop...
David B.
I suspect that the lack of educational software for cellphones is at least as much because there is no way to run it even if someone writes it.
(It's not just educational software that suffers from this limitation, of course.)
@David B: although a cellphone is certainly not as powerful and as capable as a computer, I can likely think up some useful small apps that don't yet exist. A scientific calculator, perhaps? A student oriented time management app? An addition/multiplication flash card practise app? For the much younger set, there are any number of small games - matching, memory, alphabet/number flash cards. Even a simple alphabet slide show of objects/animals, including a sound track for hearing the words?
I also think David B. might be missing the fact that Selanikio's company has already written and deployed mobile software for phones that is quite useful. Sure, an OLPC (or any other PC) would be have more capability, but will it have an internet connection in some remote area? That's actually a bigger issue than power for most of Africa, which is now under a fair amount of GPRS coverage if they're lucky. Nigeria and some of the bigger, more developed countries generally have GRPS and possibly even some flavor of 3G, but for the rest of the continent it's GRPS at best for the foreseeable future.
You can't overthink what "useful" means in this context. Just the ability to collect and transmit basic text or survey numbers is important.
The issue I've always run into with programming for phones is that despite the fact most use some version of JVM, it is extremely difficult and time-consuming to validate compatibility on the host of handsets available.
Additionally, many inexpensive handsets suffer from difficulties getting software onto the phone. Of course, this is my experience in the US, so if Africa is using a different system, then it becomes even harder for me to adequately test and deploy an app.
Which is too bad, because I really want a Scheme interpreter for my phone.
When I first said that "every child has a mobile phone; why to spend millions on to give them a computer too?" they looked at me in a way that it was a miracle I survived.
A decent mobile phone runs about 200-300MHz, and has a MIPS processor - its performance is roughly equivalent to a mac 10-15 years ago.
It's not a question of capacity: a usual phone has enough capacity for basic information systems. It's a bit slower to type, and has less screen space, but that's all.
The thing is, that the west world seems to connect the "network" idea to the networked computer. An average african, or even an average hungarian connect this to the mobile phone primarly.
Never forget that for computers, it was the internet which was the killer app for most people. And internet is about one thing: there are people on the other side. And this is probably the same for phones, and it's possble with them.
Of course, giving kids a laptop instead of a cell phone gives them the tools to write their own programs rather than relying on outfits like Selanikio's company for whatever software they want to push out.
Regarding the comments on my BBC article ("The Invisible Computer Revolution"), just wanted to make a few points:
On the one hand, David B is right in that MOST cell phones in developing countries are not high-end phones at all, and lack video capability, etc. He has probably forgotten, though, how much computing we were all able to do back when all we had was a command-line interface: all that "primitive" stuff was amazing, and very useful, back in the 80s.
In the same way, with these phones there is tremendous potential for information sharing simply through the use of SMS and other simple technologies. On my Mac I can type a word and see the definition, or the translation. Why can't you send that word as an SMS and get the same thing back to you as an SMS? Why can't a clinical officer in a rural location SMS the name of an illness and the age and weight of a patient, and receive back the recommended drug and dosage? In these cases, the phone isn't the processor, it is the connected gateway to the information on the internet.
With certain subsets of workers, however, the phones are typically much higher-end: many if not most provincial- and national-level health workers I've met have phones that run Java, play video, have basic browsers, etc. These capabilities raise the possibility of even more enhanced tools for these subsets of phone owners.
Halloween Jack is right in that without a desktop/laptop it's not possible right now to right software. But remember that the vast majority of the world's population alive today will probably never even touch a laptop or desktop in their lifetimes. By concentrating on the cellphone, we can bring useful, and connected, tools to this majority.
Joel Selanikio
DataDyne.org