Over the past six months or so I have been talking with about a dozen high-profile mobile companies, IT companies and mobile specialists to gauge interest in a new project, one which builds on the work of Nathan Eagle in East Africa (see http://web.mit.edu/epromfor details). I have also been speaking with a major US Foundation, who have expressed interest in funding the initial scoping phase of the project.

The challenge was this. How do we empower individuals in developing countries to develop and build their own mobile applications? EPROM has already been teaching computer science graduates in a number of East African universities, but how can we scale this initiative, allowing universities in other parts of the developing world to do the same? And with software development largely taking place on desktop computers, how can we empower users to build applications on the phones themselves? And if we could, what would a mobile-based programming environment look like?

Within the next couple of months or so, “mobility” hopes to start exploring these questions. The potential is huge if we can find appropriate, sustainable solutions.

http://www.pambazuka.org/images/authors/ken_banks.jpgKen Banks, founder of kiwanja.net, explains the idea and history behind FrontlineSMS – a technology that was used by an organisation, Human Emancipation Lead Project (HELP) to monitor the Nigerian elections.

An idea is born
The FrontlineSMS concept came rather suddenly a couple of years ago during one rainy Saturday evening in Cambridge, UK, a long way away from the country that inspired it. The idea now seems like an incredibly simple and obvious one, but it took its time to dawn on me.

A few months earlier, in the autumn of 2004, I was working in South Africa and Mozambique with a South African NGO, Resource Africa, on a contract with the oldest international conservation organisation in the world – Fauna & Flora International.

We were looking at ways national parks could better communicate with local communities – something which has traditionally been rather problematic – and the project I was working on at the time had a specific technology angle.

I was already working on another mobile phone project. With SMS usage just beginning its astronomical climb, it seemed like an obvious tool to consider. Things were beginning to happen back in 2004, but it was early days in the mobile revolution, particularly in developing countries. So, as a starting point in the solutions evaluation process, local companies were asked to put in tenders to help develop a service, and existing services were trialled and tested.

There were two specific issues which, several months later, became central in the thinking behind FrontlineSMS.

Firstly, everything we were looking at was web-based. This was fine for a parks authority, and fine for this particular project. But I’m always looking for ease of replication and scale, and I didn’t see the perfect solution as being a purely web-based one.

Secondly, everything we were looking at was one-way – top down if you like – and I had trouble with this. The parks could send stuff down to the community, but the community voice was silent. After a short period of research and evaluation, a trial was started with a web-based service. For me these two external issues remained unresolved.

http://www.pambazuka.org/images/articles/302/41128b.jpgFrontlineSMS, which finally addressed these shortfalls was launched in 2005. It was the first text messaging system to be conceived, designed and written firmly with the needs of the non-profit sector in mind.

Up until then, the majority of systems did not take into account the nature of non-profit work, nor the specific conditions – financial and physical – which many work under. Since the non-profit sector isn’t considered fertile ground for most for-profit companies, this wasn’t surprising. The software was picked up by a number of news sites, and trials began. Most were small-scale grassroots initiatives, however, and little news got out to the wider community.

But then came the Nigerian elections…

In African election terms, it doesn’t get much bigger than the Nigerian elections. Two months ago, back in February 2007, I was contacted by the Human Emancipation Lead Project (HELP) Foundation, a Nigerian group interested in establishing a team of volunteer election monitors to report on their forthcoming presidential elections.

HELP are a non-profit group of young professionals in Nigeria, advocating for social change through good governance. Their goal is to encourage the Nigerian electorate to participate in the electoral process. Since 2005, HELP has been in the forefront of employing available mass communication technologies in their work.

According to the group, the 2007 elections presented a 'vital opportunity to truly change the cause of things for good for the common Nigerian by ensuring that a transparent and acceptable general election is conducted'. With the proliferation of mobile technology in Nigeria, the group chose SMS as their communications medium.

Their initial search for a software and hardware solution led them to a series of mobile guides published by MobileActive, a global network of people, tools, projects and resources focused on the use of mobile phones for activism, campaigns, and civic engagement.

One of the guides specifically deals with elections and voting, and FrontlineSMS was featured in the guide as a tool worth considering. Two months before the presidential elections were due to be held, HELP contacted me and asked for kiwanja.net’s help.

As with many organisations looking to use text messaging for the first time, they were confused over issues of available solutions, short codes, licenses to operate, costs, applicability and ease of adoption. As an organisation with little or no budget, all of my services were offered for free. I still need to work on that business model!

HELP installed FrontlineSMS onto a single machine, obtained a phone and a new SIM and began their tests. There were local elections planned in the weeks leading up to the presidential elections, and they were to be used as a dry run. During the testing process I was in occasional contact with their team, but I generally left them to it. FrontlineSMS is designed to be a simple, works-out-of-the-box solution and require little or no support. Other than a couple of emails and the odd call at 3 o’clock in the morning, HELP managed to take the software and run with it with little help. The local government election monitoring was a success. The main event now loomed.

Next was the launch of a website – – where HELP promoted their work under their 'Network of Mobile Election Monitors of Nigeria' (NMEM) banner. The site’s primary purpose was to encourage the general public to register as volunteers, and tell them how they could engage in the process. Individuals registered their mobiles by texting their name, location and polling station to the new NMEM election monitoring hub. Each volunteer was then registered on the FrontlineSMS system.

On election day itself the volunteers were asked to send in two reports – the first to contain details of when the polling station opened, voter accreditation and the ballot box delivery times. The second was due when the polls closed and was to contain information on the result, counting processes, turnout and general conduct. Things were slowly falling into place.

Up until now, I was generally oblivious as to how things were going in Nigeria – I was just quietly getting on with my other work at Stanford. The first news I received that things had gone so well was an email which landed in my inbox around midday on Tuesday 17 April, which came with an accompanying press release. HELP were now ready to monitor the main election, and it was just four days away. The press release was for immediate circulation, so I put on my PR hat and started shooting mails off to my various friends, contacts and acquaintances in the social mobile world. One wrote for the BBC.

On the Thursday morning I woke to find that I’d missed seven calls on my UK mobile. On checking my email, I realised that the calls were from the BBC World Service. I also had an email from one of the BBC website technology editors. After some frantic conversations, I provided them with contact details for the Nigerian team. I could answer the BBC’s technical questions, but the real story was NMEM’s and I was keen for them to have the chance to profile their work themselves.

For the rest of the day I watched as more and more sites picked up on the story, hoping that the BBC would manage to make contact. Friday was the last chance – once the weekend passed it would no longer be a story. Perhaps, more importantly, if the BBC did manage to get the story out then suddenly there was much greater potential to recruit infinitely more volunteers.

I got up at 6am on the Friday morning – 2pm in the UK – in case the BBC had mailed and needed any final information. I needn’t have bothered – the story was already up.

The next time I spoke to HELP was on the Saturday, just after the polls closed. Despite general disquiet about the overall election process, they were very happy at the response to their call for volunteer monitors.

What’s more, FrontlineSMS worked exactly as they hoped. A result all round – after all, this hadn’t been attempted in Nigeria before - and was maybe a first in Africa?

HELP are now working through their data which will be presented to EU monitors and other monitoring groups. Sadly, in this particular case, problems with the electoral process are already well documented.

In their initial report, released a few days after the polls closed, NMEM commented:

'As has been highlighted by both local and international observers, the elections in Nigeria leave little to be desired.'

However, amidst the widespread report of fraud and rigging there were pockets of hope. In communities like Ibiono Ibom in Akwa Ibom State, 80 per cent of the SMS received indicated calm, orderliness and a free and fair exercise.

The same was indicated in reports from Kano GRA in Kano State, and Ward 3 & 4 in Calabar Municipality of Cross River State, among others. We believe that these communities should be identified and commended as an encouragement to others to imbibe fair play and transparency in subsequent elections.

It should be noted that most international observers were trained and equipped to spot and report in places where things did not go as they should. They were further sent to major urban areas where most of the heavy rigging took place.

Our observers, on the other hand, were instructed to report on everything, both the good and the bad. As a result, we documented many remote, rural communities where polls were orderly, materials arrived on time and polls were relatively free and fair.

In total, over 11,000 messages were received from the volunteer monitors, a great response.

NMEM are now looking at how SMS can be used to engage Nigerians in the everyday political process.

For its part, FrontlineSMS was just a tool in the process. It doesn’t do anything on its own, but it does empower. It was NMEM who had the mission, NMEM who had the passion and NMEM who had the commitment to drive their vision forward. NMEM also found FrontlineSMS, and they took the software and ran with it. Anyone else can do the same.

Kiwanja.net believes that all non-profits, whatever their size and wherever they operate, should be given the opportunity to implement the latest technologies in their work, and actively seeks to provide the tools to enable them to do so.

* Ken Banks is founder of kiwanja.net.

www.kiwanja.net
www.frontlinesms.kiwanja.net

* Please send comments to [email protected]