If you could ask just one question [5]

Segbwema, South East Sierra Leone
Wednesday, June 14, 2017—Dispatch 5

By Randy Thomas Jones

It is typical of this kind of fieldwork that keeping a daily log of activities is a challenge. I’m actually writing this a week delayed, attempting to capture the mood, but to also share all-important introduction for those who are not initiated into the complexities of the world of humanitarian work and conflict response. I am one of those people myself, though I’m forced to play a fast pick-up game to get up to speed as quickly as possible; the demands of working on a project like this force everyone involved to become multi-faceted and interdisciplinary.

Ivan Gayton, the project originator and CEO, is able to move seamlessly through the varying demands on his attention and knowledge. He’s got internal radar for so many of the things that to me, as an outsider, would demand attention, discussion, and thoughtfulness. It is not that he doesn’t give thought to the various issues that arise, it is just the fact that his experience is so broad, it often looks to an outsider that he doesn’t even have to consider what to do in any given situation. Add to that a native drive for action, and by Saturday morning, less than 48 hours after we have arrived, a crew is assembled at our guesthouse for training on the surveying process.

We manage to get an independent off-line local server operational, at least temporarily. It is a small gadget (called an “Intel Edison”) that easily fits in the palm of a hand, which functions as a repository of the ODK (Open Data Kit) software that we use. Ivan is able to ignore the chaos that surrounds the scene much better than I. News of our presence has already travelled, and even though we are only offering about 4 or 5 positions for surveyors, attempting to state clearly that anyone who wants to apply must have an Android phone which will successfully take a GPS point, there are at least another ten people milling around who it seems do not have phones, or at least do not turn up in our verification process.

The first step to filling out an Open Data Kit Android phone survey

Anyone who has a phone is instructed to download the app ‘ODK’ from our Server, and then fill out our “Surveyor Registration” survey. Open Data Kit: for data collection of all sorts worldwide. It is a wildcat procedure—part of the test is whether a potential candidate can focus on the instructions and complete them. It is not completely Darwinian; Ivan is more than happy to help if someone gets in the queue and asks for help. (In fact, that is one hallmark of a likely good candidate, we are happy to contribute time to training, but there is a limit to the amount of energy that can be devoted to it). The experienced surveyors help also to get the phones loaded up with ODK and the registration survey.

Ultimately, the surveyors will go out with a similar form to fill out on their phones, one that has been designed by us using the background development tool, XLSForm, and that is how we will gather information about the all-critical proper village names and locations. Since the GPS is critical, the surveyor registration form includes the task of gathering a GPS point. Not all of the phones are able to complete the task properly, and discussion ensues about the possibility of showing up with a working phone and continuing. Yes, some allowance for this will be made, but it had better happen fast, we are planning to get on the road on Monday for our first field test.

Through the Open Data Kit survey software, and its associated support, what shows up on a mobile device (Android based only) is a simple form than can handle much more complexity than we need for this project. For our purposes, we are only attempting to disambiguate place names for the purpose of healthcare. In fact, Ivan will repeat himself several times over the following week in various meetings: “We are a public health project using mapping, not a mapping project. I love maps, but what I care about is saving lives.” The ODK software has been around a while (and there are also other versions with slightly differing specifications since it is an Android software), and it seems whenever anyone is doing a project, everyone has a curious desire to get an extra question added to the survey.

Ivan resists as best he can all comers. His general answer is a polite reminder that the phone number of the village chief is one of the critical pieces of data that we are gathering. When the data is uploaded, humanitarian agencies will literally have the ability to give a phone call to any of those villages, which, much more often than not, will have a skilled English speaker available. He has to bow good-naturedly, of course, to the Red Cross request to add a question about the presence of village markets. MSF likewise gets a question added about the availability of working water pumps. In my enthusiasm, after the first field day and we accidentally discover a story about the meaning of the name of the village, I argue to include that as a question as well. Our colleague Rupert Allen, who will be joining us shortly, has an academic interest in the way people are dealing with new technology. He would like to add a few questions as well, Ivan gives him one, he is not, as it turns out after all, a complete ogre!

Back at the training day, there are enough people who have sent a form with a correct GPS point to the Server (which can immediately be verified y reviewing the database as it emerges) to begin training. It is not a completely logical step-by-step process, but once the basics are introduced, everyone gets a copy of the real survey and we begin practice. Just as the survey design is critical, or perhaps why the survey design is critical, the asking of questions can’t be a completely rote process—some attention is given to explaining the intention behind the questions and the idea that people’s point of origin coming to a health facility will be a good thing.

Our general instruction is to ask for a knowledgeable villager (often the village chief). This is not as pompous as it might sound, it is simply a paying respect similar to the idea that you probably wouldn’t immediately walk into someone’s back yard in a Western context, you would knock at the front door and introduce yourself first—the chief is the front door of the village. This provides an opportunity to do some role-playing, with one of the experienced surveyors playing the role of the chief, and the other two coaching on the side.

After about three or four rounds, it is starting to go fairly well, and I offer to step in as temporary chief, planning secretly to mix it up and cause a little trouble to test the resilience. I know Stanley, Alberta and Richard can handle it.

The first task is saying hello and explaining briefly that we are trying to make the map accurate because it is kind of insulting that the villages are not even known to exist in the outside world. It is mostly going to be an easy sell, but during the Ebola outbreak a few places had to be forcibly quarantined, we have to account for the possibility of refusal, and indeed, consent to the process is a gold standard. My first response in the role-playing is to proclaim that I don’t want my village mapped, and to tell the surveyor prospect to leave the village! This causes a little laughter, but there is also a good kind of confidence, “Can I just explain what we are doing?” says the surveyor-in-training. “Yes, that is acceptable,” I reply. With a bit more chance to explain, the chief gets convinced, the survey can proceed, but I am not out of ideas to stir up a little trouble yet.

Like Information, Maps want to be Free [4]

Segbwema, Kailahun District, Sierra Leone
Tuesday, June 13, 2017—Dispatch 4

By Randy Thomas Jones

Friday afternoon, after our visit to the Nixon Memorial hospital, we hit the ground running again, getting some possible candidates for the positions of surveyors and drivers. The surveyors have to have a working Android phone that can take a GPS point. We get them to download the Open Data Kit app (ODK for short) and install it. Then they have to find a simple survey: “Surveyor’s Registration.” Name, address, phone number, GPS point. If they can do that, they are still in the field. Five of them are successful, though the group is salted with a couple of ringers.

Alberta and Stanley were both involved in mapping projects that happened following the Ebola outbreak in 2014. They know the ropes, and it is important to get them assisting with training and informal leadership. That provides a level of redundancy, the opportunity to refine instructions in the nuances of the local language, and provides aspirational models for the other new candidates. With us also, from Freetown, has been Richard Bockarie, a Sierra Leonean with a variety of relevant skills. He’s done mapping before as well, and even knows how to do some of the programming behind the scenes; he can be an asset to the project.

Along the way, I’ve again had the pleasure of learning from Ivan some of the history of computer programming. It isn’t directly relevant to the daily tasks of painstakingly asking for a receipt for every purchase, the obligatory defeat of dehydration that forces a stop to pick up bottled water, the endless need for a catlike agility to deal with simple unexpected circumstances, but it is good to know a little bit about the subject because it is an important thing philosophically for the project and our partners.

The argument is structured around perceptions of the value of information, technology, and programming. Early on, most people in computing thought that programming would be a part of computers that no one would pay for, and that the value was in the hardware. Later, it became apparent that the programming had economic value after all. But one of the early designers of computing wanted to buck the trend that was developing toward proprietary programming.

‘Free,’ yet ‘priceless’ maps, can save lives.

getting some shade
Arrival in a village, shade is both sought after and freely offered.

There was one strong ethic in the early computing development of free-wheeling creativity—if you want it, go ahead, make it up for yourself . . . and for everyone else as well. If someone adds to what you have done, fantastic, you will probably borrow something that they have made next week. It is a share and share-alike community. Along the way, the phrase “information wants to be free” emerged as a slogan.  Information wants to be free doesn’t mean free of charge, it means free to move, be utilized, be accessible. Here is where we loop back around to the value of humanitarian mapping. Maps, Ivan would say, also want to be free. They paradoxically get more valuable as time passes when they are available to use as a part of public infrastructure, especially when used as a foundation for the improvement of public health. “Free,” yet “priceless” maps, can save lives.

Returning from one of his MSF missions in 2013, Gayton was ready for a desk job in London, England. (He had also acquired a family by this point). The year was 2014. He walked to work with enthusiasm the first day and settled in looking for the same background base maps that he had seen in Haiti 4 years earlier. But by the end of the first week, he realized that there was no background maps like there had been in Haiti from the intense contribution made by the digital humanitarians of the Humanitarian OpenStreetMap Team and Haitians living in the US. In most other areas, there is no strong expat group of diaspora who had taken up the task. The situation was unacceptable to Gayton, and after a little more cross checking, his suspicions were confirmed—the early seeds of a plan started forming. Let’s return later to that, and jump back to Sierra Leone where we are beginning to implement some of the ideas that started developing that day back in London.

It is only our second day on the ground in Segbwema, a town of perhaps a thousand people, but we must return to Kenema in our first attempt to deal with our technical problems. There is actually no electricity during the day, though the Nixon Hospital is able to light up the town in the evening because they have a field of solar panels. We pay extra for a generator, but somehow the fuel, much to my dismay, it doesn’t actually last until the promised 7:00 am, only until 5, when I happily awake ready for coffee that I can’t make because I have neither coffee nor hot water. During a later sojourn for supplies, I manage to find powdered instant coffee (typical in West Africa), and powdered milk. It is at the electrical equipment store, which is tracked down with the usual quick few referrals, that I come upon a plug in kettle. I also manage to secure my packets of instant Nescafé coffee.

We’re not at the electrical store for the kettle that I covet; we are in a rush to set up an independent power supply so that our high tech equipment will work. Happily, they have a battery Ivan is content with, and an inverter. The inverter is a 12V to 220V Alternating Current unit. When we do have power, we will be able to plug in and recharge the battery, (a 12 V 100 amp hour unit, which is twice the size of a car battery, that proclaims its ability to operate at 25 degrees Celsius). I think we will certainly be able to manage the temperature requirement here in tropical Sierra Leone, and when we don’t have power, in theory, we’ll be able to plug in phones and computers to recharge . . . to some degree, as it is said. They also offer us a solar panel, and it does cross my mind to enquire whether it would provide enough juice for my kettle, but my attention is quickly drawn towards the next step in our adventure.

Where do we send the Ambulance? [3]

Segbwema, Kailahun District, Sierra Leone
Monday June 12, 2017—Dispatch 3

By Randy Thomas Jones

Today I’ll return to the story about how contact tracing, maps and public health all connect the dots that form an arrow pointing us to come here to the verdant and vibrant atmosphere of West Africa, to gather people with Android phones, to find motorcycles with their drivers, and make maps of places that can literally be “off the map.” In many cases, being unrecognized means that voices from the hinterland are not noticed, and this is the first step to helping people say to the administrative systems of the modern world, “We are here.” In a way, an analogy could be that if one called for an ambulance in the modern world, one would expect to be asked, “Where are you?” And given an answer, to expect them to arrive. In low-income settings, sometimes the difference between being known and unknown is a difference between life and death.

With Nigel Jagbwem OSM
Screen shot, indicating example of geotagging photos as part of a data set. [Photo Credit: Rupert Allan]
Yesterday I was discussing the need for mapping in Sierra Leone, and how the lack of accurate maps and gazetteers hampered the response to Ebola in 2014. In Nigeria, on the other hand, extremely aggressive contact tracing did effectively stop the outbreak before it became an epidemic. Nigeria had the infrastructure available to a wealthier state that made it possible to ask the key question of where patients were from and receive an unambiguous answer. Sierra Leone has been declared Ebola-free for 18 months now, but Sierra Leone is still considered “At Risk,” because the virus is in the environment and could again transfer to the human population.

One of the things that becomes apparent after some consideration is that the creation of maps and an unambiguous gazetteer (which lists not only village names, but the association with higher administrative districts) is much better done before a crisis hits. And of course, registrars at clinics and hospitals have to understand why it is necessary to use them, and be insistent on a full answer when asking, “Where are you from,” or, as sometimes is the case when someone brings in a friend or relative, “Where is this person from?”

Dr. Monk is developing his portfolio to focus on both UK-based medical practice and Global Health for Alternative Certification in Core Medical Competence, enabling him to tailor his studies to his interest area of epidemiology, infectious diseases and microbiology. He’s a congenial ball of fire when he invites us to the local bar in Segbwema Thursday night, and he and Ivan happily trade stories, both of them enjoying the meeting of minds with similar obsessions. Ivan considers him a great resource as a friend to the project—the more typical situation is that doctors don’t pay that much attention to public health (again, that old bias of the consequentialists). Doctors, and quite rightly, tend to be Kantians.

Friday morning we walk up the hill from our guesthouse to meet Monk. Or, more precisely, he arrives at 8:00 in the morning with a typical British enthusiasm and a warm invitation to the canteen at the Nixon Memorial. A breakfast that is unusual, for the Western palate, composed of fish and rice, but I do remember the “Egg and Bread” from my visit to Ghana last year, and determine to track that down whenever possible!

motorcycle track
Typical of some percentage of village access trails, impossible to transverse with even a 4 wheel drive vehicle—walk or bike! [Photo Credit: Rupert Allan]
After breakfast, we take an extensive tour with Ed through the hospital find out that the Nixon Memorial Hospital has surprisingly good practices for registration. “Daddy” is the registrar at the outpatient desk, and takes great pride in his system for keeping track of patients. To the inexperienced eye, it looks like a haberdashery of file cards in randomly placed boxes. In fact, Daddy passes every inquiry regarding how individual patients’ records can be traced through his system. He has aspirations, and borrows a computer to practice his Excel at every opportunity from the hospital lab. Similarly, Betty, who manages the intakes on the wards, confidently pulls out scrolls of old records and points to one. “That would be in that one,” she says. Ivan rates them in the top 10% of facilities he has seen in low-income countries (the term he prefers to the usual terms “undeveloped / underdeveloped / developing”).

What started out seeming to be a quirky interest in records turns out to be only the “responsibility of care.” I’m getting enthusiastic about records myself at this point, and there is a certain kind of devotion to the practice that is as important as remembering to feed all of your children, and not leave one hungry.

Leveraging the goodwill that Monk has built, we make some very casual inquiries as to the possibility and interest for adding any columns to the intake books. Currently, the disambiguation of towns with the same name is done informally, depending on Betty and Daddy’s admittedly formidable memories. There are complications to this; one of the master books is a federal record that has a specified format that doesn’t meet standards for current best practice. But the answer is a tentative and provisional, “Yes, possibly.”