Category: Humanitarianism

Woken at 4am by mosquitoes – I left the bathroom door open, and the insect screen only works so far.  None of the hotels here have mosquito nets, so I spend an hour listening and swatting before going back to sleep.  I breakfast on scrambled egg and bananas: my substitute for all the bread products I keep being offered.  The team is here: I watch them do a soil type analysis (using a flowchart: seeing if they can form muddy balls from the soil samples, seeing how long a ribbon of mud they can make (1cm? 2? more?) then finally rubbing in their hands to see if it’s “gritty” or not.  Most of the soil we collected is clay or sandy loam.  It rains.  Msofi and I swap British and Kiswahili words for different types of rain (we both have many of these).

I watch the team do rapid roadside assessments.  In rural areas (red dirt roads), these are done every 2km; on main roads every 20km, and in LOs (regions with a large concentration of surveys) every 4km.  A roadside assessment is just that: the team stands at the roadside, looks at the 50m by 50m area of land in front of them, assesses plant cover, plant types and use, and takes a panoramic photo of the area.  We stop and do 4 of these assessments on a dirt road, and mark the sheet for this with the from and to village names of the road.  I chat with Gervase, who is a seriously good systems thinker – he talks about how he persuaded rural people to adopt more efficient cooking stoves not because of environmental concerns: people living out here don’t understand the concept of save the forest, since the forest is all around them, but do understand the concept of “your eyes won’t get so red from cooking”.  In some areas, red eyes are seen as a sign of witchcraft, and people’s grandmothers have been killed for this.  I have serious respect for this guy and how he gets things done.

We go back to the hotel and go through the data entry and upload process, where data from the paper forms are entered into a tablet, then sent up to the head office servers.  The team are using Samsung tablets which have slots for simcards – available from Amazon, but I’m warned that the Dubai versions have a smaller simcard slot which means cutting down larger cards.  The team uses ODK Collect for its forms – these are available from the main server, but have changed many times already (mostly bug fixes, e.g. not being able to enter lat/longs ending in 0). We start with the soils form, but Joseph has the old version of the form on the tablet he’s holding.  He finds the right form on another tablet (internet here is still terrible), and walks me through the data entry.   There are many forms (and many versions of those forms): Eplot, soils, rapid roadside assessments, water rapid assessments, water lab reports, household surveys, agriculture surveys, farm field and crop surveys, the contents all of which we will have to make comprehensible (along with satellite data, open data etc) to decision makers.

We say our goodbyes and thank yous to the field teams (all Tanzanian, all experts who play games like “name that tree” with each other) who are starting their Easter break after 40 days in the field.  They’ve done roughly 50 of 350 Tanzania sites so far, and have much still to do.

We set out for the bright lights of Iringa, the big town in this area and a 2 hour drive away.  We talk about water sampling methods and the issues of vandalized equipment and data not getting from the basin offices to the central ministry of water.   There are potatoes roadside now, and big boulders that look like glacial moraine, which confuses me – were there glaciers here?  We talk about the timber lorries we pass – there’s a huge need for timber across Africa; there’s much construction, and the people who own woods will become rich on this.  I think about the transitions between old African and new Western-style systems that I’m seeing, and think about the things that get lost in that transition.  Some of these are tragic, e.g. there are many blue babies (brain damaged from lack of oxygen at birth) born here now because the traditional practice of midwives sucking on babies’ noses to clear gunk has been lost in the new Western-style hospitals, but hasn’t been replaced by the Western-style use of suction bulbs to do the same thing.

We reach Iringa, and start hotel-hunting: Avery knows two places in town with wifi. We try the first one: a craft shop, café and guesthouse run by disabled people.  The rooms are beautiful, but their wifi is out.  We hear about a guesthouse by the university, and try that – no wifi, but there’s a strong signal from the nearby internet café.  The five-story glass building opposite seems surreal after a week of one-story houses.  Life now is all about getting wifi, and getting back to ‘normal’ work.  It’s a catholic guesthouse – we’re staying amongst nuns and looking at pictures of the pope, but there’s also a bar in the evening.  We pick up wifi vouchers from the café (5000 ksh for 5Gb) and head off to lunch.  We’re in the tourist zone, and the first café we try is full of earnest young Americans, English menus and high prices.  We go round the corner to a local place and eat lentils, beans, rice and salads off tables with tablecloths and matching cruets then head back to the hostel to get online.  Nicky goes off to get the car fixed – the long fast drives over rough roads have damaged a pipeline and bearing. Which is painful… the bandwidth is so slow that OpenStreetMap goes to the low-bandwidth version, and I can see the titles of my emails (eventually) but not the contents.  I manage a brief Skype conversation with Nairobi before giving up.  Avery goes off to buy her bus ticket (she’s going up-country from here for Easter), and I haggle for local fabrics (blue chickens!).  Then we switch to plan b: the other hotel with wifi has a restaurant, so we head up into the hills above Iringa to a place that even my clean jeans feel a bit underdressed for, and eat Indian food with our laptops in front of us.  I finally get an OpenStreetMap editor open on the area that we were lost in, and show the team how well the red dirt roads and waterways stand out against the vegetation.  When I have good internet again, tracing will happen, so the field team have a small-scale map to start from next time.

We head back to bed – the hostel rooms have mosquito nets, so I sleep (in the trying-not-to-stick-anything-outside-the-net position, waking once to the sound of a frustrated mosquito in the room.

We saw a lot of schoolchildren in uniforms yesterday – in the afternoon, they were walking past carrying hoes.  We go shopping for a cable to charge my phone and  camera: I’ve lost one cable and broken the other, so it’s off to the local shops, each of which points us to another one: general store, electronics shop, phone shops, camera/video shops, some in buildings, others in plywood shacks. Finally have two colourful cables with smiley faces on the ends (the powerstrips in the shape of hearts are tempting too). Avery buys fruit from a lady with a huge basket on her head: it takes 2 of us to the lift the basket back up.  We check the cable: it’s the car adapter that’s broken, not the cable: the team lends me an adapter for the ride.  Add to field shopping list: car chargers, and lots of them.  The car is filthy: I just miss a shot of it parked next to the same colour and model: left car is red; right car is brown.

Gervase and I talked about sample areas last night. One set of sites is in a wildlife area, with twin dangers – from the wildlife, and from the poachers chasing the wildlife.   The team can look “official” in their khakis – Joseph tried to ask a local man for directions yesterday, but the man ran away – a common thing here.  I have huge bruises up my arms, splinters everywhere, torn jeans – but no bugbites. We saw trees, plants, flowers yesterday but almost no wildlife: some insects and butterflies (not many), birds (ditto): nothing larger, although we did find wild pig poo and some elephant damage on a tree.  We also found an animal trap on the path (Joseph got caught in it), but the jungle was eerily silent.

We arrive and walk to the first site (300 metres through open woodland – yay!), but the site’s start point is in a swamp.  This happens occasionally, and there’s a protocol for it: the team moves the corner to dry ground with similar vegetation to the swamp, and marks on the papers that they’ve done this.

Joseph explains the sampling protocol in detail and I take many many photos of soil, buckets and poles.  First tree-counting: they use the ranging pole (long pole with tape markers every 50cm) to find all the trees within 1.5m of the plot start point (“corner”), then they measure every tree over 5cm diameter and 50cm height (knee height).  They do this so the analysis team can estimate carbon sequestration in this area.  Each tree’s diameter is measured with a tape; the small tree heights (5m or less: the ranging pole is 2.5m long) are measured with the ranging pole; the larger tree heights with a clinometer; the tree’s canopy width is measured with the ranging pole (in dense woodland, one person shakes the tree so the others can see where the canopy is).  Once the team starts on a plot, the same team member does the same measurement jobs at every “subplot” point: because people’s estimates vary, this gives some consistency of measurement across the plot.   The team takes soil samples by putting down the metal plate and pushing a corer (also marked up with tape) through the hole in it; putting soil into buckets marked 0-20cm (topsoil), 20-50 (subsoil), 50-80 and 80-100cm (the 80-100 sample is only taken in the first corner): these depths are based on the Afsis (African Soil Information System) sampling protocol.  The team also throws down a quadrant (50cm metal square) and counts the species in it, how much of each it contains, how much bare earth, dung etc; and uses a densiometer (mirror marked into squares) to estimate the canopy cover north, east, south and west of the start point. They also hang a scale bag in a tree, put 500-900g of each soil in a marked plastic bag, weigh the bags, and also weigh marked metal pots containing samples of each layer of soil (the pots are added to a plastic bag attached to the scale).

We hear a shotgun – the team thinks it’s just a car backfiring or tyre blowing out – there are very few guns in this region.  We move on to the subplots.  What we’re doing is called the “E” plot – we’re sampling a 100m by 100m square area, with “subplot” points every 20m in that grid (e.g. there are 36 subplots to a plot).  The shortest path with the smallest errors through that grid is in the shape of a large “E” (eeeeessssswwwwwneeeenwwwwneeeenwwwwn), hence the name.   At each subplot, the team counts and measures trees (trunk diameter, height, canopy width, species) and estimates the biomass only in the quadrant (from the average plant cover and heights).  The team used to do the full works (everything done at the first corner) at each subplot, but the process was long and has now been streamlined.

There are lots of trees here – I learn Kiswahili for “please shake the tree”.  It rains a little, and we have 33 subplots to go.  There’s more noise here than in the jungle: birds are singing, insects are buzzing.  I ask about wildlife protocols – there aren’t any animal, insect or bird protocols yet.  The team were going to set camera traps for animals, but they were too expensive; I wonder if there’s something small and cheap we can hack together with a camera board, motion detector component and microprocessor.

We move over the road to a Pinus Patula (misheard as “Spatula Pine”, which causes much amusement) plantation – well-spaced small trees with soft spines and wild mint growing underneath.  We talk about Apopo’s work on mine-sniffing and TB-sniffing rats, and how this works in Africa but not so much in SE Asia.  At every corner of 100m grid, we take soil samples and do detailed quadrant analysis; we also do this at the centre of the plot, where we take a panoramic photo (the camera has built-in GPS).  I’m still picking thorns out of my head; at the centre, I sit on a fallen log, and Avery rushes to brush the ants off me (I now have ants in my pants).  It rains heavily. The team gets wet, and the car goes red again.

The team has spilt into two, to get two sites done for the day.  We drive over to the other site; it’s raining heavily so Avery and I sit in the car and work on our laptops.  She gets a good data signal for the first time in days (the signal at the hotel is non-existent) and I manage to post an “I’m safe” home.  The flowers here are beautiful, making the site look like an English cottage garden: huge purple mallows and something that looks like clematis overlaying delicate yellow flowers – I wonder how many common British garden plants have come from here.  It stops raining for a while (this rain is monsoon-grade), so I walk out to the site: across a swampy valley, through woods, across grassland to a firebreak between woods.  I find the team’s start point (an umbrella over the sample pots), and track them to the edge of the woods – which look impenetrable: someone has chainsawed the plantation but, puzzlingly, not taken away the fallen wood.  Trees have grown up through the fallen wood, and the whole effect is one of a giant woven basket.  I hear the team’s voices, find a not-so-bad patch, and push through, under, over, across trees to reach them.  They seem surprised: they’ve been pangaing through these woods for hours now, and weren’t expecting anyone to just go through them.   They’re still on the outside legs of the “E” plot – with 22 subplots to go.  These are very different woods than the morning, so I tag along to see how they sample in dense woodland.  There’s a lot of scrambling but not so much wait-a-minute here: today is branch scratches rather than thorns.  Moses the biologist tells me that there are two species of wait-a-minute, and that the one by my head is related to the orange tree – I crush a leaf, and yes, it smells of oranges.  In the firebreaks, I see animal tracks going into the wood – later, whilst crouched under a fallen tree, I hear what sounds like a boar grunting annoyance.  It gets late: 5:30pm in a place where the 6:30pm sunset is a sudden from light to dark.  The team has 10 plots left to do, and push on, quickly measuring trees and assessing the ground cover.  Just before sunset, we finish and rush back to the cars before dark.  I’m wet, cold, muddy, sunburnt (regretting not bagging that aloe) and happy that I understand a lot more about what it takes to collect this data, what it means and what we could do to help.

Today we go into the field.  Woken by laptop charger fizzling – electricity is available but a bit variable here.  Breakfast with sweet milky tea – the tea in it is grown and picked here in Mafinga.  We drive past tree plantations – pines, for their wood. I ask about the rice; Tanzania is a major rice grower, and much of it comes from Morogoro.  I meet the rest of the team, and explain Ushahidi and my own skills to them, as asked, armed with a notebook, pen and much arm-gesturing.   We drive off to the site; on red mud roads, fast.  The team truck has a snorkel and I worry they might be using it.  Finding routes to the sites is an issue, and the gps units fill up if the team trys to track roads: we talk about roadmapping using their GPS-connected tablets and Funf, and about OpenStreetMap traces from Bing’s satellite images.

This area (Mafindi)’s economy is based on trees, tea and maize; I see eucalyptus (grown for its wood) and other trees grown for paper.  I see some cows – we talk about the perceived difference between pastoralists (many cows, with perhaps a small piece of land for vegetables) and farmers (mostly crops, with perhaps few cows), the conflicts between them, and how farmers often don’t count the cows as part of their farming.  We see the first tea plantations of many – I’m surprised that the tops of the plantations are flat rather than tea bushes.

We drive fast for 2 hours on dirt roads – we’re remote but trade is visible; motorbike repair shops and women wearing cloth made in Nigeria.  We see cabbage fields – European missionaries settled here because the climate was familiar to them; there are some churches but also many schools in this area.  It rains again – the roads are getting muddy now.

We stop on the road.  We’re 7km from the site, so we go back and try another road.  The logging trucks are out now (11am) – Mafindi Paper company, and little trucks with offcut bark.  If only we had a roadmap.  We talk about how to improve tea production efficiency, and wonder about mechanical harvesters – just before we see mechanical tea harvesters: giant lawnmowers pulled over the tops of the tea plants.  We reach a dead end: 6km away now.  Really really need better roadmaps for this project; I can feel an OpenStreetMap session coming on.   And here’s how to help:  there’s Internet here in the sticks, but it’s slow – too slow to sensibly edit maps from here.  But internet is good in New York, London etc: so if you help OpenStreetMap make better maps of this region (southern Tanzania), local people can get on with the things that they need to do, like plant monitoring, instead of getting lost.  I’m told it takes 4-5 hours to process each plot, and that each plot will be revisited every 3 years.  If you add getting lost onto that, it becomes a very very slow process.

We stop at a dead end 5km away from the sample site. It’s getting late, so Joseph, the field team lead, asks me if I’m okay with walking 5km (3 miles) – he explains that 5km straight is probably going to be a lot further on the trails, and that it might be a bit up and down.  I look at the tea plantations around us, and think “hey, this is just like walking in Dorset”.   We set out… across the team plantation and down into woodland – walk downwards for a while, then retrace our route because two of the team are yelling that there’s a river in our way.  We set out again… down through woodland, across a small brook (which I’m hoping is the river), up past a small shack with a fire going, and small garden with mint and vegetables (Joseph explains that sometimes the farmer stays with his fields), and up through a maize field.  The field is closely planted – I follow the voices ahead of me.  The field is underplanted with courgettes, which I take care to step around; and then we’re out into another field – some type of wheat?  at the top of a hill.  We go down through woodland again – a slippery muddy path that looks well-used. Someone mentions that we’re going down to the bridge the plantation workers told us about.   I think “ah – a roadway; great’. They don’t mention that the ‘bridge’ is a pair of tree branches across the river.  The guys walk across one of the trees then jump onto the far bank; Avery and one of the guys wade waist-deep across instead.  I take the tree – the guys build me steps down out of their field plates.   This is the first time I hear them say MamaSita; I hear that a lot soon.  We walk along a small trail going up through the trees – and then the trail stops.  It’s panga time: Joseph starts hacking a trail through the jungle, and the walk becomes a long repeat of ducking under vines, picking “wait a moment” (bramble-like plants) off our arms and heads and waiting for enough of a trail to be cleared.  There are holes in the jungle; I step over most of them, but it’s muddy, and sometimes I slide thigh-deep into them.  The guys talk all the time – laughing, teasing each other, talking about politics.  We stop every so often, and call out the distance to the sites.  First stop, it’s still 5km.  Then 4.8 km; we have a
picnic” of samosas (or in my case samosa innards: it’s tough being gluten-free in the jungle).  After 3 hours ducking through the wait-a-moment (jungle: I thought snakes and big cats would be an issue: turns out it’s falling into holes and picking big thorns out of your head), it’s 3:30pm, we’re 3.6km away from the sites, and have 3 hours of light left for the day.  We turn back, planning to return the next day.  The route back takes 1.5 hours: when we return, Nicky has collected local pears for us all.  We drive back past towns with repair shops and chickens, past recently-logged areas and more logging trucks.  I’m exhausted and covered in scratches and bruises – I crawl off to sleep for a while.

Today I meet the team: we breakfast, talk about the plan for the week (travel, measure, measure, rest, travel) and set off on the road to Iringa.  I’ve now been in 2 of the “big 5” wildlife countries and so far have seen: 1 dog.  I’m hoping we might see something else in the parks.  The road is very quiet – most of the traffic has stopped because of the traffic jams around the flooded bridge, which is great in terms of having the road to ourselves, but no so great in terms of being the only car around for the traffic police to stop.  They stop us and show the radar gun (the most common sensor that I see around here) – speeding.  We stop next to one of the communities making woven baskets – I’m tempted to go shopping but know that would just increase our chance of a fine.  I see another dog.  We pass the waterfall where our driver once took a hippy who tried to teach him about transcendental meditation.  All the police want to talk to us, to see the car’s papers – today they must be bored.

And then we enter the Mikumi National Park.  Right away there are baboons – mothers with frisking children, big proud males with bright bulbous bottoms.   Then giraffes, posing tall under shady trees. Impala peeking through the bushes.  Elephants ranging in the distance. Wildebeast and zebra sharing a watering hole.  And more giraffes.   Someone jokes that this is the “safari commute”.  We eat good African food just outside the park (last night the hotel staff made me a late meal: of chicken and chips, then eggs and frankfurters for breakfast), then continue through the plains and along a river.  As promised, there are Masai herding cows whilst on their mobile phones (“they carry two at least”), and small boys with sticks and goats.  I look at an NGO crew in their big white 4×4 and wonder how many of them actually know what it’s like to be poor – not poor as in student, but poor as in having to make the difficult choices you make to survive.  We see onion stands near the river – you can map your location here by what’s on the vegetable stands: onions, mango, tomato, peppers, and finally, in the high plains past Ikaya, potatoes.   We climb, past a crashed lorry, up a mountain road that Nicky tells us once had a phone at each end because it was too narrow for vehicles to pass).   We’re in high meadowlands now, and there are many sunflower fields.

We stop in Mafinga and choose a hotel – the cheaper one that just opened today.  I sign in as their first guest, and put “Webb” in the “tribe” column.   We eat local chicken, plantains, rice and big sweet avocados and talk about maps.

Day 1 – not going well. 5am start in Nairobi – check.  Flight over Kilimanjaro – check. Car waiting at airport – check. Takes credit card as promised – nope. Hack Tanzanian cashpoint to get enough cash to pay.  Buy sim card: simple, but not instant (forms!). And off.. to mall to get cheap camera and supplies.  Cheapest camera = 220000 ksh ($130).  Asked $80000ksh for bugspray and sunblock. Decide to buy supplies in tourist area upcountry.  But first, driver1 has forgotten the car’s id docs – an issue on a road with policement every mile or two.

And so we get into the first traffic jam (10 mins): the president’s mother lives nearby, and visits his mother every weekend (even presidents aren’t immune to this).  Everything stops as his cavalcade drives past – except today it’s the vice-president visiting.  Second traffic jam: 10 minutes for a busy traffic junction in Dar – the traffic policeman routes us onto a lane full of motorbikes coming towards us, and we test the 4×4 going offroad to get round a car stuck in front of us.  And onto the main road out of Dar, and from Tanzania to Zambia, Malawi and Congo.  Third traffic jam: 10 minutes as everything stops, and the president’s cavalcade dashes past.  The rainy season has started, and we hear that the road is damaged ahead.  Fourth traffic jam:  we drive past a long line of stopped trucks, to where the president is inspecting a subsidence (kind of a deep wrinkle across the road) – everyone is walking forward out of their cars to see and hear him.

And then the fifth traffic jam.  Everything stops. For miles.  We drive past lines of trucks, then cars, then buses (there are people everywhere now) then cars.  We stop on a bridge – below us, a group of Masai are stood on the riverbank, watching the trafficjam like tourists.  We’re told that the road ahead is flooded, and a channel’s being cut to divert the floodwater: perhaps a 1-2 hour wait for this to be done.  Motorbikes and big 4x4s with snorkels swarm past us: I make a mental note to hire something with a snorkel next time.  As we drive forward past the stopped cars, a passerby shouts that the problem is “too much crocodiles”. Everyone laughs; we’re all stuck, and this is rapidly becoming a social occasion.  A vanload of traffic police drive past in the opposite direction – perhaps they’ve given up?  It rains again, and we sit and wait for news.  And the president returns – slowly this time because there’s no room left on the road to go fast.  Merchants appear, selling fresh-roasted cashewnuts, which I eat looking at the trees they came from (and smelling the next batch roasting).  I read the boring textbook about javascript that I’ve been avoiding, chat with Nicky (the driver that Esther insisted I had for Tanzania) and people-watch. I learn that Masai carry a stick and an umbrella (and a gust of wind reveals a silky pair of pale-blue boxershorts.  I learn how useful traditional dress is when a girl has to go off into the bush. I learn Kiswahili for “white” and “cashew” and “bugger off, we’re not paying that”.

It’s now 4 hours since we parked on the bridge (we’re now at two lanes of parked traffic, with the 4x4s and motorbikes still swarming past), and we start to move., hopeful. People drive past the jam, taking video of it on their mobile phones.  My own phone dies (add to shopping list: car charger for phone).

We hear that the bridge is clear now, but a line of cars has overtaken the waiting buses, and another line of cars has overtaken that one, on both sides of the bridge. The main route from Tanzania to Kenya now has 3 lanes of static traffic (4 or 5 if you count the bikes; 5 if you count the 4x4s creating a new cross-country trail) with nowhere to go, and no traffic police in sight.  It’s going to be a long time getting over this bridge.

7.5 hours after entering the trafficjam, we see what it’s all about.  First, we drive down onto the river’s floodplain: it’s dark now, and there’s water either side, with crocodile eyes reflecting in knobbly heads just enough distance away to be menacing but hard to see.  The the water – first a static flood across hald the road; not moving, nothing to worry about. Then it’s moving, and quickly becomes a torrent, sideways across the road, getting deeper and deeper as we cross until we reach two trucks stranded on their sides, motorbikes piled across their chassis.  This isn’t a good sign; I start regretting my earlier “just line up the crocs” jokes, and wonder who the god of angry rivers is hereabouts.

Then we’re through, breathless, stunned that we just tried that (but after 7 hours’ wait, what the heck) – and stop again.  We’re stalled on the other side of the river by a mirror of the lines of cars that we just left.  We’re hungry and thirsty – and by luck, stranded in the middle of the night market that many people had come to in the morning before the river broke over the road.  I learn how to jostle for drinks with Masai tribesmen (British hard stare doesn’t work, waving money at barkeep does), and Nicky and I settle into trafficjam number 6 (a van has broken down on the road ahead of us).   We move at last, and drive past the jam on the other side of the bridge. This goes on for miles. People are sleeping in front of their trucks, and in the roadside near buses.  And I’m quietly struck by the thought that this is what a crisis evacuation looks like- the jams, the queues, the rumours, the hope, the decisions: do I stay here with the bus or walk forward in the jams? What do I eat? What should I save just in case? And how can I help other people?  The answer to the last one is “with information” – as we drive, Nicky shouts to stranded drivers about the situation, the bridge, and the length of the queue ahead of them.

Cities are apparently the future. All the predictions I’ve seen for the next few decades show the world\’s population concentrating in cities, but our development indicators and policies are still listed by nation state. Perhaps they should be wider, for instance by including developing cities on the lists.

I said “developing” there – which begs the question “how are these cities developing?”.  This isn’t just a Las Vegas-style spreading of suburbia across the desert: many of the cities I’ve visited in the past year have shanty towns, and these appear, at least from outside, to be where a lot of the city development is happening (btw, I wanted to use a less emotive word than ‘slum’ here: although it’s what Slum Dwellers International uses, there’s still a lot of negative feeling about it).  From Lagos to Guatemala to Haiti, I’ve seen dozens of homes and businesses under tin roofs looking across at smaller numbers of tower blocks, and wondered “how do these economies fit together”, and “where does it go from here”.

Good old BBC gave me a few more answers… and a few more questions (like how does the nation-based world fit with people this adaptable and informal), and case studies, Medellin and London, of both positive and negative ways that the shanty and non-shanty worlds can start to fit together.

Perhaps it’s the way you look at it.  If you look at the Wikipedia links above, you’ll see shanty towns and slums described in very negative terms… impoverished, illegal, lack of services.   If you hang out with people who live or work in shanty towns, they’re communities and neighbours and businesses and services – and quite possibly the adaptable, informal, majority economic future of the cities they\’re part of.  Whichever way you look at it, there are a lot of people in shanty towns, and how (and what) they develop is important.

We needed an example problem set for our current version of Hunchworks (note that this is a very early, i.e. pre-alpha version of the code and a lot of the cool Hunchworks features aren’t in it yet). The UN’s main use for Hunchworks is to gather up the weak signals that people put out about emerging development crises – those small hints that something isn’t right that appear all over the world before they coalesce into ‘obvious’.

Awareness of development crises can happen very quickly. One minute there are whispers of a potential problem – a chat here, an email or text asking for a bit of data there. And then a tipping point appears and there’s suddenly data everywhere. And we have a great example of this happening just at the time that we’re demonstrating Hunchworks to the UN General Assembly.

We had one of these serendipitous test sets before: we tracked the Horn of Africa crisis emerging across newsgroups as one of our early will-this-work paper exercises (this btw is also why we’re suddenly interested in data mining googlegroups).  But the Horn of Africa crisis is well established in the public eye now, and there is both too much online data on it to pick out the early weak signals, and many of the early traces (e.g. anecdotes and messages) have been lost in both human and machine memories (yes folks, not everything on the Internet is logged).  But there’s a new thing starting to happen (which is potentially very very bad for the world and certainly for the people caught up in it) – over the last month or so there have been mysterious messages here and there about something starting to happen in Chad and Niger, across the African region known as Sahel.

So this week we started collecting information about Sahel and turning it into hunches and evidence in Hunchworks. First we had an email from a trusted colleague containing potential places to look. Then an Internet search for news and background information, followed by more digging across the Internet (including recent reports from the UN and other NGOs), a colleague searching in the food-related UN agencies and a Twitter search for hashtags and interested parties.

We could have got a lot more information faster and with more insightful comments into it if we’d crowdsourced its collection, but the EOSG couldn’t be seen getting involved prematurely (i.e. before the dedicated HLWG team) in this crisis. We could have also involved more people in a non-public search if we’d had Hunchworks at its Beta testing stage, but doing the first search by hand is a sane early stage test that exposes bugs early to a small number of people before a larger group (e.g. the Alpha and Beta testers) get annoyed by them.

So what did we learn about ourselves, our information and Hunchworks from this exercise?

We ran the exercise using a spreadsheet (no, we can’t just do this for hunches because it will quickly become overwhelmed – see below). Its first worksheet was a list of hunches: this was quickly populated with a mix of hunches and evidence for hunches that took some time to separate out. We also discovered that the evidence that we gathered often contained new places to look for evidence, suggested new problems that should be proposed as hunches and spawned a whole pile of other evidence-gathering activities.

• Lesson 1: people confuse hunches and evidence. Sometimes evidence is posted on the hunches list; other times hunches are posited as evidence to another hunch.
• Lesson 2: evidence generates hunches. For example, we realized that a hunch about a famine in Sahel also contained hunches about famines in Mali and Chad that the country teams there needed to investigate.
• Lesson 3: evidence generates evidence-gathering activities. We ended up with a to-do list linked to each hunch.

Some things were confirmed for us. We suspected that we could map the connections between hunches as though the hunches were propositions in a reasoning system. We also suspected that there were a set of basic search actions that we would do at the start of most hunches, that some of them would be ongoing (i.e. to catch new information being added to the Internet) and that we could automate many of these. Yes.

• Lesson 4: when we draw graphs using our hunches as nodes, the links between nodes look suspiciously like the links in semantic networks. This should come as no surprise to anyone working on linked data.
• Lesson 5: Google searches, news searches, twitter searches, UN report searches and emailing around likely suspects are obvious first things to do on any new hunch.
• Lesson 6: We can automate some of the above searches, especially if we have search terms (e.g. Sahel) and tags to start from. Our hitlist for Sahel was: twitter stream, food price, migration from/to Sahel and news monitoring agents.

We’ve tried to build a system that doesn’t need much management or moderation. We might need to revise that: in the initial excitement of chasing up leads and links from the original hunch, it was difficult to maintain momentum (e.g. amount of evidence added) and completeness at the same time. I had to do a lot of reading and editing – both to disambiguate hunches and evidence as discussed above, but also in generating tags, thinking about the links between hunches and managing the list of actions that happened most times we added any evidence. Some more lessons from this:

• Lesson 7: Hunches have information-gathering actions attached to them.
• Lesson 8: Once we get textual evidence, it’s pretty easy to create tags from it.

And then we’ve got some very specific lessons about the system.

• Lesson 9: if two hunches are related, they probably need the same people involved in them. Can we start the “involve x” list of one from the other?
• Lesson 10: Some places had 2 locations, e.g. migration from Libya into Chad.
• Lesson 11: We have the same problem as crisismappers with location accuracy, e.g. sometimes we want to mark a region rather than a single point on the map.
• Lesson 12: Using tags brings a set of questions about how we find related things again. This is the same issue we’ve seen in crisismapping and Twitter feeds, and we have tools that can help with this.

There are more lessons learnt, but we’re somewhat busy today. More soon.