Read my write-up from the Mobiles for Data Collection Technology Salon:
You might think that the topic of collecting data via mobile devices would be a rather dry discussion of data management and statistical methodology. You would be very, very wrong. The Technology Salon all but came to blows as we wrestled with privacy issues, total costs of ownership, and other elephants in the room.
When you combine some of the brightest mobile-for-development minds from projects stretching from agriculture to health to democracy, all of whom are facing increasingly common problems, perhaps that's to be expected. Stories were shared around the basic challenges of data collection, picking the system to use, and the complications of different sectors.
Read more: Mobiles for Data Collection Technology Salon
I missed the recent Technology Salon on Mobiles for Development (I was kind of busy with Global Youth Service Day), and was already scheduled to make it to the Mobile Active m4Change unconference the next week.
A prime outcome of the Tech Salon was a challenge from Vodafone's Terry Kramer:
Mobile network operators (MNO) want increase revenues and market share by expanding into rural areas, and see partnerships with the development community as a key market entry strategy. Specifically, Vodafone is looking to the development community for key applications that solve a common need for many and can be scaled into commercial activities.
Here's my response, an idea that's been rattling around in my head for quite some time.
A common "form" application to send out encoded questions and receive back answers via SMS.
The core would be an agreed-upon encoding/compression format for sending form questions and receiving the answers via SMS.
This standard could be coded to using existing outreach tools like EpiSurveyor, FrontlineSMS, RapidSMS, and the like. It would have a "server-side" component where you could set up a form, specify the answer types (y/n (or t/f), multiple choice, very-short-answer?) and compose the SMS messages to send.
The key (and the hard part) is having this app as common as a calculator tool on deployed phones. The app would capture the coded SMS form, present it as a user-friendly form, and take the answers and reply (again via SMS) using as few outbound messages as possible.
Anyone with an SMS-sending tool could code questions and use their tool to distribute them; and any phone with this (pre-installed) app could "decrypt" the compacted form and present it to the user.
This would make data collection much smoother, eliminate distribution of questionnaires and codesheets prior to each new questionnaire, and improve data quality.
I see immediate applications in election obersvation, human rights monitoring, and health field-worker reporting, not to mention census, and even for-profit ventures for immediate customer satisfaction surveys, more complex SMS-voting (imagine what BravoTV would do with this were it widely deployed!)
Check out some updates -- props to @MobileActive!
During the last breakout at the Mobile for Change (#m4change in twitter, a good writeup by Development Seed's Will White: Recap of Mobile Tech 4 Social Change BarCamp ) open conference yesterday, we began to get into some of the problems that had been bugging me all day.
During NDI's Ian Shuler's presentation on the state of Mobile in Development (especially in Election Observation) I twittered:
what happens when govt catches up, blocks sms during unrest? we need p2p networks #m4change
9:38 AM Apr 29th from web
Now that I'm not constrained by 140 characters, let me unpack that a bit.
Repressive governments are getting smarter. They've enjoyed the ability to censor, control and manipulate traditional media for quite some time now, and have been caught blindsided by the Internet and cell phones. We cannot hope or expect that to continue. We've seen the first, blunt reactions to this - shutting down SMS service during elections to prevent the spread of information, and during the #pman protests SMSs were severely throttled. In #pman the protesters were able to continue exchanging information on their cellphone's data connections. I wouldn't count on that working the next time around.
But it's unlikely that SMS would be totally cut off in a country for more than a few days around a big event, so why worry so much?
Because of course shutting SMS down temporarily is just the blunt response of a government that hasn't had time to do anything more devious.
Forcing all SMS messaging to run through a series of scans for keywords, tracking political dissidents using SMS to spread the word, and taking actions against people based on these types of logs would destroy the use of mobile phones - not only in the spread of democracy, but also in any form of activism that was not strictly in line with the current regime.
Due to the centralized structure of cell phone systems, scenarios that make China's Great FireWall look irrelevant are pretty easy to imagine even for a regime without much tech savvy.
So -- how can we fix this?
Encryption on cell phones has two big problems - it's not built in on most handsets and it's very clear that you're trying to hide something, which is just as bad as doing that something. Unless encryption was built into the protocols and implemented by default, it's useless -- and even if it were implemented, the telco could probably still access the texts, so we're no better off than we were before. Sure, end-to-end encryption would fix that, but key exchange and signing via SMS? I'd like to see a workable RFC on that.
I think the most valuable solution is a peer-to-peer mesh network that can carry voice, SMS, and data on it. The downsides of these is that their range can be extremely limited by lack of density of users.
The technology is simple: Terranet outfits a special Erricson phone with peer-to-peer wireless networking ability. In its pure form, there is no need for base stations, antenna installations or infrastructure. With this phone, a user can call and text anyone at no cost within two kilometers, or up to 20 kilometres in a mesh network.
The trick of course is getting these mesh phones (or adding mesh capability into enough existing phones) rolled out in enough numbers to make a difference. There are some clear use-cases in development for disaster-prone areas, areas without telco service (they could be a great business model for micro telcos), or as a low-cost phone + data connectivity tool.
However, there are quite a few barriers against this. The price point is higher than your basic cell phone, and you have to have rapid local adoption for it to even work. From a business side, it must fight against the existing install base of cell phones (and their providers). The business plan for the network is itself hamstrung by the very power of the technology - in a mesh network, how do you charge for calls that you don't know about?
If however the mesh technology could be embedded on new phones as part of a disaster-mitigation scenario (even in the US, cell networks are notoriously overloaded during crisis), a powerful technology would find its way into daily usage that was immune to many of the censorship and oppression problems of centralized communications networks.
There's a project Comm.Unity at MIT's Media Lab that runs as a software piece on top of cells, PDAs, and laptops:
Comm.unity runs on mobile phones, PDAs, and regular old laptops and PCs, allowing them to easily communicate with each other and build networks of interactions for their users without the need for any centralized servers, coordination, or administration.
There's also Fluid Nexus which is an application for mobile phones that is primarily designed to enable activists to send messages and data amongst themselves independent of a centralized cellular network
Fluid Nexus requires bluetooth, and for people to be within bluetooth range (though it is a store-and-forward style system), so it has its own limitations, but at the same time could be a powerful tool in the right situations.
It's a start.
My bitter, cynical hope had been to demonstrate that the conversation switched from a small Romanian-language conversation about the actual protest events to a self-congratulation festival in the English-language twittersphere. Good thing we’ve got data to prove me wrong. [...] I’d expected to see “twitter” emerge as one of the most popular terms by Wednesday or Thursday, and to see the conversation shift into English. [...] But by Thursday, Twitter’s out of the top 20 entirely and “comunistii” ranks behind Moldova and Chisinau. So yes, the conversation on Wednesday - the busiest day with over 1,000 authors - included lots of non-Moldovans. But the conversation quickly shifted back to the political standoff.That being said, there are under 200 reported actual twitter users inside Moldova; so while the conversation avoided turning into the twitter version of back-patting, it also is not the twitter flash-mob we're looking for. Worse, governments are getting more sophisticated in limiting the utility of mobile phones for this kind of disruption, as Evgeny Morozov at ForeignPolicy reminds us:
I've just spoken to a Moldovan friend who is himself a big technology fan; according to him, there is little to none cellphone coverage in the square itself (turning off cellphone coverage in protest areas is a trick that was also used by the Belarusian authorities to diffuse 2006 protests in Minsk's central square), so protesters have to leave it to post updates to Twitter via GPRS technology on their mobiles.It seems likely that next time around, the government will also make sure GPRS is hobbled as well, and there were reports that the government was strong-arming local ISPs into restricting outside connections. So while Twitter was involved, it seems too early to claim it's victory, as both Evgeny Morozov and Ethan Zuckerman seem to agree on. There was no sign-in form at the protest with a "Where did you hear about this? ( ) Twitter ( ) Facebook ( ) SMS (non-twitter) ( ) Friend ... " so we can't really be sure of the impact of any one social utility over another (though we could do some interesting things with Facebook photo tagging perhaps?), and this will continue to haunt any attempts to link online social media movements with offline action. That's not the only story here, though. While I'm excited about turning online interaction into offline action, I strongly believe that the lower-hanging fruit in social media sites is real-time, mass reporting of events. You may get a thousand different viewpoints, but you're guaranteed to not just get one filtered and sanitized report. As Evgeny Morozov notes;
There are also a few moving English-language Twitter posts like this - "in #pman a grenade thrown by the police has torn apart one of the protester's leg"- that would surely be perused by foreign journalists.We saw the role of SMS and Twitter in getting the news out about the Mumbai bombings in November 2008. As microblogging sites get increasingly sophisticated (or their users settle on hashtags and location update formats) I think we can expect to see fast local news coming not from traditional media but from our peers. Without editorial oversight or research/verification, we'll have to rely on mass numbers of twitterers reporting on each event to present an evenhanded view, but overall I see this move towards instant sharing of information as an amazing development that will only getbetter and more interesting, both in the case of free speech and media, and for mobile possibilities for development.
I recently saw Ken Banks present at a local speaker series run by IREX. He gave an updated version of this presentation from POPTech, on the power of mobile phones in citizen empowerment, NGO communication, and a host of other amazing stories of using the available, appropriate technology in remote and rural locations which are often off-grid and without Internet access. By attaching a computer (Linux, Mac, or Windows) to a cell phone with a data cable and installing his (free, open source) software, FrontlineSMS, that computer is turned into a messaging hub; sending and receiving text messages via the cell phone to hundreds of contacts.
That's pretty amazing. Three reasonably available pieces of hardware and you have a tool to send alert messages out, receive election monitoring information through, or communicate with field medical workers to coordinate and track supplies and treatment information. Or track corruption. Or report human rights violations. Or share news and tips in places where the media is not independent, as one of the FrontlineSMS success stories shows:
Are Mobile Phones the Winner?
February's Technology Salon was on the (false) dichotomy of mobiles versus computers in development. Thankfully due to the high caliber of all the attendees, we were able to establish and move quickly past the problem that so often plagues the actual projects and "real world" debates - which is better? Some people will claim mobile phones are better due to their low barriers to entry, but then you see low-cost computing and netbooks providing that same promise to computers. Others will argue that you'll never write a school paper on a cell phone.
The reality is, the entire frame of this argument is off on every possible angle.
First, there are clear cases where one technology is better suited to a task than another. I'd no more write long papers on a cell phone than I would carry around a laptop to use as a personal communications device. However there's a large chunk of tasks where either tool will suffice, and which "should" be used is more a factor of the local conditions than the features of any one technology.
Secondly -- and more importantly -- this discussion is tool-centric. We have a hammer (two, in this case) and are going around the development landscape searching for nails we can drive home, and it's a race between the two hammers to see who can hit the most nails. This is inherently the wrong way to apply ICT in development.
We shouldn't be arguing about mobiles vs computers, or even OLPC XOs vs Intel ClassMates, or Windows vs Linux, we should be arguing about specific problems in development, what tools could help, how, and for what costs (training time, implementation and infrastucture gotchas, as well as equipment costs).