About two weeks ago, there was a good deal of buzz around Nathan Eagle’s talk at the ETech Conference (or more properly: O’Reilly Emerging Technology Conference). His talk was about txteagle, a new venture that he has started in Africa that could potentially change a great number of ways that people work in Africa.
While I was very much aware of the talk, I didn’t have much of a chance to really sit down, watch it, read some reactions, think about it, and toss together some of my own thoughts on it due to Maneno getting a bit of press (thanks again to everyone) around the same time and taking up my days because of it. Now I’ve had a chance to finally get back around to all of this. Maybe you didn’t see the talk either? It’s almost 39 minutes long, but if you’re having a slow day, I highly recommend giving it a watch as whatever you may think about txteagle, Nathan brings up some good facts:

All done? Great, because there are a great number of points in there that are worth discussing.

Good Stuff

Eagle is creating a very innovative system. I like that. I like that he’s found an area where there is a need and a possibility for growth. Obviously, paying people to do small tasks isn’t an original idea. Amazon really fired up the idea by starting their Mechanical Turk system. Eagle admitted in the talk that he was told the best way to explain what he was trying to do was a “Mechanical Turk for the mobile phone”. It’s an apt comparison, although I got the feeling he wasn’t too keen on it.
There is a massive amount of growth for a system such as this. Mobile phone usage is exponential in the developing world and as he pointed out, the majority of mobile phone users in the world are indeed in developing nations. txteagle takes advantage of a perceived abundance of time for people in these areas, a lack of work, a need for people to work small, low-paying tasks, and a dearth of extra network capacity (although he obviously didn’t want to get in to beeping which takes up some of that perceived “extra” network.) This all seems like it’s pretty much a win-win for everyone involved, but there were a couple of things that stuck out a rather nasty thorns which Eagle seems to want to pretend aren’t there.

But How Good is it?

Like I said at the beginning of this article, I was interested in seeing what other people said about this talk. It was surprising to see that all of my regular reads, whom most anyone in to Africa and technology would know, didn’t mention a thing about it. Maybe they were busy, or maybe they just didn’t want to say some of the harsh things that probably should be said. Luckily for me, Steve Song summarized just about every large-scale issue I had with that talk:

Nathan makes an important point when he says the fact that no one in Kenya can afford not to have a mobile phone. Even if you are digging a ditch by the side of the road, day labour is now organised via SMS. This means that mobile operators have Kenyans by the throat.
He gives another example about a water pump manufacturer in Kenya who, by combining a mobile-mPesa-enabled, solar-powered metering system with their water pumps, have completely changed their business model. They are now able to give water pumps away for free (if I understand correctly) and then make a profit by selling access to water via Safaricom’s mPesa service. Send the pump 20 Ksh and it pumps 20 litres of water for you.
…Am I the only one who finds this a little disturbing? When a single mobile operator is a gatekeeper to water supply, something is wrong. For any village in this situation, Safaricom can charge whatever they like.
The failure of communication regulators in Africa to either license sufficient new market entrants or to curb the excesses of incumbents with significant market power has led to a situation where existing operators collude to maintain high profits.

Steve perfectly nailed the biggest issue in all of this to me. Basically, it is allowing monopoly control of resources. This is nasty. Obviously, there is no way for txteagle to function without working with the major mobile operators, but I have to question as to whether this will ultimately be a good thing, or as Steve says, “have Kenyans by the throat”.
I also have to say that I’m a bit perturbed by txteagle’s eventual acting as a small fringe bank. One of things that Eagle proposes is to have people be able to collect credits and then roll them over in to an actual bank at some point. At no time during this will txteagle be considered an actual bank as they are actually just classified as a “financial creditor”. He says that you don’t want to be a bank in Eastern Africa because of all the undue regulation. Sitting here in the crumbling economies of the developed world, I have to ask, how is regulation of a monetary holding company a bad thing? These regulations, while burdensome, exist to aid the citizen, not the institution. Allowing txteagle to work outside this is rather shady in my book.

The Long Term

A question that I’ve been asked on one or two occasions (and probably more in the future I would hope as it’s a good question), is, “What is the incentive for people in Africa to use Maneno?” To this, I usually give the same response, which is, “The same as for anyone in North America, Europe, or Asia.” People find their own incentives to blog whether they be response in the form of comments, visibility, informing a greater audience, activism, or any number of other items. Within a community, there will always be individuals who wish to act at the historian, chronicler, crier, or journalist. Those are the people I want to be able to give access to Maneno so they can act on those desires if they wish.
I bring this up because we’re approaching Maneno from a very hand’s on, functional point of view, not an academic, theoretical point of view. This difference is rather critical in that I think that txteagle is based too strongly on the theory and not enough the practical. Don’t get me wrong, I still love the innovation, but it appears to have been created first to try and fill a need (the SMS Bloodbank example he mentioned in the beginning), which it couldn’t fill without introduction of a practical element–payment. Eagle then went on to develop the theory out further without taking in to account the fact that there were two needs to the equation which were constantly uknown and changing variables: a source of work/tasks and a source of revenue. He was focused mostly on the primary issue that there is an abundance of waiting labor. He said that in the pilot, they’ve “run out of tasks”. This is an tremendous issue because if people start to rely on these tasks for income and then they suddenly dry up, then this is in some ways worse for people than their current situations.
It is true that txteagle could end up with a good deal of work from companies in need of such small task work. Like Eagle pointed out, medical transcription is a huge market, which I guess someone would listen to, write down, and then text back. Could work. The surveys seem a lot more promising, albeit limited. Same goes for the advertisement calls. But, the one big, big area with the most potential for short task work like this is object identification in videos and images (or any binary, non-textual file.) In a past “sometimes office” I worked for, they had to send the piles of video they had off to the Philippines for matching the products which were shown in the video. There is huge potential in this, but it is simply not possible over SMS. In fact, it’s barely possible over modern mobile data connections. So, I have to question as to where this is all going and if it could potentially be much more harmful than beneficial given the collusion with large mobile operators, possible shortage of work, and the technological hurdles that are simple not possible to overcome at the moment. If txteagle can get past all of these issues, then it will indeed be a fantastic service. I suppose only time will tell at this point.
Thoughts about Nathan Eagle and txteagle