WhiteAfrican

Where Africa and Technology Collide!

Category: mapping (page 2 of 3)

Testing Google Driving Directions in Kenya

Gone are the days where you had to have lived in Nairobi for a couple years before you understood all the backroads and neighborhoods in order to get from one place to another. Kenya has an advantage as the only non-sales office in all of Africa for Google is here. When they create new tools, or customize a feature from the developing world, for Africa they do it here in their own backyard first (and sometimes Uganda).

Google Maps in KenyaA couple weeks ago Google turned on mapping directions for Kenya. Like me, most of the people who know Nairobi were shocked and didn’t believe it. Could this really work? It does, and it works well.

I’ve been testing it out for the last week to see what type of results I get, and I’ve been impressed with the results. Fortunately I have my iPhone with me, and it allows me to do things like challenge Google/Apple to find my current location and then give directions from that location to somewhere in Nairobi that I happen know every backroad, alley and footpath between.

Shortcomings

No control for traffic
As omnipotent as Google seems to be, what they’re unable to do is track the vagaries of Nairobi traffic. So, as logical as the directions you get from Google might seem, they are not the best way to go much of the time. While they give accurate directions for new people to Nairobi to follow, they are also the “obvious” route and will cost you hours of sitting in gridlock while you watch the matatu’s clog the road even further.

Alternate route to gigiri

Lack of detail on the maps
Some areas, even large towns like Ongata Rongai aren’t even shown on the map. Below is the failure screen for getting directions from Rongai to Eastleigh. I had to go with Langata instead, as that was the next closest “town” in Google Maps. This pattern holds true for dirt roads and paths that are usable by vehicles, but which don’t show up on the map.

Lack of data - fail on Google map direction in Nairobi

I’ve also seen this in regards to offices and buildings, where they are put on the wrong part of the road, sometimes off by a good half kilometer, as was my father’s office in Upper Hill shown below.

Finding the BTL office in Upper Hill, Nairobi

Final Thoughts

Anyone living or working in Kenya should buy a drink for every intern and Google employee who has done the manual work to get Kenya mapped to the level that it is. It’s an iterative process that only gets better as time goes by and more people work on it.

As Google states:

“This essential tool is by no means Google’s effort alone – we’re enormously grateful to Kenya’s active online cartographers who have helped us build these maps from ground up with the use of Google Map Maker, a tool that allows people to help create a map by adding or editing features such as roads, businesses, parks, schools and more.”

The directions provided by Google in Nairobi (I haven’t tested up-country) are adequate. They’ll get you to and from the locations in Nairobi that you need to go. You’re better off now than you were before, and as someone new to the city you’ll have a lot better luck with Google’s maps and directions than you’ll have with asking someone on the side of the road.

African Connectivity Visualized

Jon Gosier’s Appfrica Labs has put together an amazing infographic on internet connectivity in Africa. Amazing work!

Infostate of Africa 2009

“The African continent is rapidly changing. In the next two years 2 billion dollars will bring 12 terabits of connectivity to the continent. Will africa become the world’s newest outsourcing hub? Will it foster it’s own tech and startup culture? The image above explores the ‘infostate’ of Africa in 2009.”
(Read More)

Flickr set here
Full-resolution version here
Buy it in print here

Tags, Time and Location

On Friday I had a long conversation with Noam Cohen from the New York Times about Ushahidi and Twitter. He was doing some homework for an article he was writing on the increased value that geolocation data can add to the massive streams of data coming out of tools like Twitter, called “Refining the Twitter Explosion with GPS“.

A lot of our discussion was centered around location, especially since he was thinking of the Ft. Hood shootings and the value of location in determining useful information from the Twitter stream during that crisis. This is what we’ve built Ushahidi around of course, the idea that location and even small bits of information give us a better understanding of an unfolding crisis. This is just as true of mundane information, or trending topics in a locale, which is why Twitter is building a new geo infrastructure. It couldn’t be in better hands either, with both Ryan Sarver and Schuyler Erle on the team, what Twitter puts out will be top notch.

What was more interesting than just geographical references for information was the combination of two other big ways to parse this data: Time an Tags. We’ve started to see a lot more apps mixing time and location in the past year or two, and we’ll see more as the visualizations for it improve. Categorizing information, pictures and video by keywords (tags) have been around even longer.

TwitterThoughts

We need to see more combinations of tags, time and location in visualizations and platforms. I can’t think of anyone who does all three really well (if you can, please leave the link), though there are a number who do two of them incredibly well – including Flickr’s geocoding of images (tag + location), TwitterThoughts (tag + time) and TwitterVision (time + location), etc.

We have a widening stream of information. The lowered barriers for entry globally, and the encouragement by social tools, means we’re seeing exponential growth rates. Twitter alone saw an increase from 2.4 to 26 million tweets per day in just the last 8 months. We need some way to make sense of this information. Our ability to create information has far surpassed our ability to understand it in a timely manner.

Chris Blow outlined this best with a visual for Swift River for use in a presentation I did at TED this year:

information produced vs information processed

It’s a serious problem and one that only gets deeper with every month that passes. In most areas, it’s not a big deal, but when a crisis, emergency or disaster hits the misinformation and lack of understanding has very real consequences.

I’d love to see more work being done with all three: Time, Tags and Location.

Traffic Updates by SMS in Nigeria

eNowNow is a service in Nigeria where anyone with a mobile phone can sign up to receive updates on traffic conditions in different areas around Lagos.

How it Works

Traffic via SMS in Lagos Nigeria - mapArmed with a mobile phone, a team of 4-6 motorcyclists ride to different, pre-designate parts of the city. They take pictures of the current traffic conditions and MMS that image to the central office. That image is then geolocated and given a score of “slow”, “moving” or “free”. Anyone who has signed up for SMS or email alerts is then sent a message with the traffic update.

Challenges

I asked Simon, one of the people putting the service into action, what some of their challenges are. His reply:

“Collecting information in this way, although not that technical (lots of people have said why not use stationary webcams it would be technically superior), is turning out to be more difficult than we expected. Finding people who can grasp the concept behind the service, ride well through the crazy Lagos traffic, and are reliable has been tricky, added to that we’ve had lots of issues around harassment and even arrests from the police (many police officers apparently believe you need special police permission to take photos of traffic) and just recently the weather has been in our way as the rainy season has just started in Lagos making operations more difficult and a few phones have been dropped in puddles! “

The business side

eNowNow doesn’t see much value in charging premium SMS rates for their services. They believe margins are low, and they don’t think the uptake would be high enough amongst their target market to make it work. Instead, they have plans to subsidize the service with revenues from licensing traffic information to Sat Nav providers and logistics companies.

“In Nigeria the networks will take anything between 40 and 75% of a premium SMS’s cost to a subscriber for themselves (pull or push) leaving you a tiny margin for profitability and driving the industry standard (and therefore what the networks will allow you) per SMS cost higher. Most people think that traffic only affects those in cars and they can therefore afford to pay for a service, but most of Lagos’ population aren’t in that bracket and those on public transport still have choices about which buses they take, which routes and what time they leave work.”

Thoughts and ideas

Maybe it’s because I’m a motorcycle fan, or maybe it’s because I have a deeply ingrained detestation for being stuck in traffic in Africa’s mega cities, but this application hits the sweet spot for me. I’ve been wanting just this type of thing in Nairobi for a long time…

One additional idea, to make this even more dynamic, and spread it over the whole city is to create a way for ordinary drivers to text into the system when they come across a new or growing traffic problem. I imagine that Lagos has areas with traffic that is not on the pre-designated points that eNowNow operates in currently.

This is a classic locally grown tech initiative, and I hope that they can pull it off. If so, it can definitely be replicated in other major metro markets across the continent.

Map and Stats for Africa’s Undersea Internet Cables

Steve Song has put together a great interactive map that helps you visualize what undersea internet cables go where in Africa. There’s also a helpful table of statistics and data on each of the cables. Head on over to his site and check it out.

A map of Africas undersea internet cables

More on the history of this project.

Location, Mobiles and Social Networks

It’s all beginning to come together, at least on the fringe where all of us technocrats live. Social networks have been humming along quite nicely, many people you know are now part of a service like Facebook, MySpace, Twitter, Bebo or Mxit. On the edges, some applications have started to pair up location-based services around them, thus the rise of smaller applications like FireEagle, Loopt and Brightkite.

What’s always seemed to be missing is a way for location, mobile phones and social networks to coalesce. A way for you to communicate with people, be it updates, comments or chat – and then apply location to that as you chose. Those social networks that tried to do it all couldn’t do it at this level, because they didn’t have critical mass (such as Brightkite). Those that had reach, like Twitter or Facebook, don’t have a simple way to play with location for everyone.

Enter Google Latitude

Just over a week ago, Google Latitude launched. It’s a location-based service that mashes up Google’s own mapping products with Google’s communication products; Gmail and gTalk (chat). One week later, they announced that a million people were already using the service in the 27 countries that they had released it into.

Google Latitude Screenshot

While people are discussing how great the technology works, and it does seem to be quite impressive if you carry one of the supported smart phones platforms (BlackBerry, Windows Mobile, Symbian and Android), I believe there’s something even bigger going on here. Google has not had much success in the social network space, so they are taking a rather nontraditional approach to getting embedded into people’s lives at a much more foundational level. Gmail has a base of 50 million+ accounts, and each comes with a chat service, which has gained quite a bit of popularity. Not to mention, SMS was enabled within chat just a couple months ago, in December.

What Google appears to be doing, is leveraging its massive user base, tied together through email and chat services, and pairing it together into a larger community that works within it’s mapping infrastructure.

(Putting on my Ushahidi hat, this has some pretty big ramifications for disaster and emergency work in locations where Google use is heavy.)

The competition

It also has the potential to change the game for some other large services. What happens if people start using Google Latitude for their status updates instead of Twitter and Facebook? What service do you use to find out what’s happening on a Friday night?

It will be very interesting to see what types of reactions to this service arise out of the large social networks, especially those with a large international footprint. Getting location, mobile and social networks to play together isn’t easy, yet these organizations will not sit by as Google whittles away at their empire.

Here’s something to think about. If you didn’t realize this before, pay attention: the big international showdown in this space is between Google and Nokia in the coming years. They have been gaming each other for over two years, and as the race to the edges begins, you’ll see them come head-to-head more often.

Nokia Ovi

1.5 years ago Nokia bought mapping service Navteq in a mega-deal at over $8 billion. Last summer they launched Ovi, which allows remote sync capability for photos, contacts and calender, gains access to music and games, and marries up their mapping and sharing capabilities. It’s what Nokia is banking on for their consumer value-added services in the future.

I’m not sure who will win out on usage in the end, but I do think that Google’s Latitude is an incredibly strong and under-the-radar type play that should be watched very closely. One thing is for sure though, the organization that opens up for easy third-party development on their platform will have a better chance.

Digitizing Africa: Starting with the Dirt

I was completely fascinated when I heard about the African Soils Information Service (AfSIS) and their goal of gathering detailed digitized soil samples from 42 countries in Africa. That’s a huge project, and it’s propped up by grants from the Gates Foundation and AGRA.

African Soils

Unlike many of you, I have little knowledge of farming and have no agrarian pastimes. So, though the mapping and techie side of me thinks it’s really a neat project, I didn’t know why it was needed. Apparently, one of the biggest problems with food scarcity in Africa is lack of knowledge on soil degradation and the low-yielding crops that these areas generate. This information is critical to identify the types and amounts of mineral and organic nutrient sources needed to increase crop yields.

Digging deeper

I started trying to find out more about soil mapping in Africa, and came across the European Digital Archive of Soil Maps. It turns out they have an amazing number of scanned geological maps for almost every African nation.

I decided to look up my childhood stomping grounds of Kapoeta, a dusty (tiny) town in the Equatoria Province of Southern Sudan. Sure enough, I found a hydrogeological map (circa 1989) with some good information:

Hydrageological map of Southern Sudan - Kapoeta

It turns out that Kapoeta is a bit of an anomaly, in that you won’t find too many areas in Southern Sudan with as much geological diversity. It’s just off the major flood planes, and it there are 3 different types of hydrageological structures within the area. The local Taposa tribe stores their wealth in cows, but they do grow some millet.

Though that data means very little to me, in the right hands it can make the difference between a large annual millet yield, or possibly even the introduction of a new crop that locals didn’t know about. It’s in places like Kapoeta that this project will see it’s true potential.

Digitizing Africa

I was happy to see the following quote in the press release:

“All soil information will be collected and made available via the Internet in a user-friendly manner. AfSIS experts will offer training to agricultural extension agents and others on how to interpret and translate information provided by the soil map for practical application.”

As we get more open and available data on Africa – be it soil, vote counts or census information – more value added services will be created. Businesses can grow up around both the data collection and its use. More importantly, with the use of other tech tools, I think we’ll find that the information that is aggregated and then acted upon, will start to make it’s way back into the hands of those who need it for their daily living. This soil project just might be a greater thing than we realize.

Microblogging, Location and Emergencies

I’ve been using Twitter for a while now, and have thought quite a bit about it in Africa. More, I’ve thought about what the ramifications of Twitter pulling out of the global market means, and then thought quite a bit about Jaiku, Laconica and Mxit and various other chat/microblogging applications. There is, without a doubt, a move towards short-form updating via mobile and web, and it needs to be federated.

There’s something missing in this new mobile + web microblogging movement, and I think it’s location.

Thoughts on location and microblogging...

Why Location Matters

Most of us use these services for updating, and being updated, by our friends and interesting people. That’s the main use, and it will remain so. The truth is, you and I don’t really care to hear what any random stranger is doing, even if they are nearby. However, we do care what is happening on a very hyper-local level in the case of emergency or “big event”.

It’s somewhat like the “pothole theory” that I talked about earlier: you wouldn’t normally care about the pothole on a steet, unless it’s yours. It helps explain why we care about certain things.

If you use Twitter and have an iPhone, you’ll probably be aware of Twinkle – it’s an application that enriches your Twitter experience. In Twinkle, you can set your location and then a certain radius from which to receive twitter updates, even if they’re from perfect strangers. I think that’s the beginning of what we’re talking about.

However, again… I don’t want to just get updates from random strangers in my locale. I want to only receive the ones that are “important” to me. I want to be notified when there is an emergency, major traffic jam or something else pertinent to me.

The “What if…”

What if we created a way that a greater federated system of microblogging applications could also use location as an alert point?

Of course, my current world is colored by Ushahidi, crisis and emergency news coverage. I think of the ability to anonymously send in reports to a system like Ushahidi running in any country, and those who are part of this greater, extended and federated network would be updated – even if that person was unknown and anonymous.

Federated Microblogging, SMS and Location

Here’s a use case:

John is a Twitter user in Accra, Ghana. Anne has setup a local Laconica server with 5000 users in the greater Accra area. Eddie is not part of any of these networks, just an average guy with a mobile phone. Ushahidi is running in Ghana.

Users from the Laconica group can setup an “alert” for a specific radius from their location using Ushahidi, linked to their Laconica account.

An earthquake happens and Twitter and the Laconica server are ablaze with dialogue about what is happening. Eddie (our normal guy), sends an alert into the Ushahidi number, along with hundreds of other Ghanians who are not part of Laconica or Twitter. Anne, and the other Laconica users are receiving alerts (web and mobile) from within their set alert radius automatically, from completely anonymous people. Alerts on where people are trapped, who is missing, who is found, where not to go, and where help is needed most.

John, our Twitter user is updating Twitter, but it has no little local implications due to not being able to be used in Ghana (except via web). Local mobile users aren’t receiving his updates, and he isn’t receiving theirs.

I recognize that there are a lot of things going on in this scenario, and it’s imperfect, but it serves as a good setting to discuss some of the shortcomings of the current situation and the possible growth areas for them. It also talks to even bigger ideas and the greater impact in Africa of a real social mobile network that can connect people using only mobile phones and do it as needed.

There are some interesting things to learn and apply from location-specific alternatives to global SMS gateways (like FrontlineSMS), and I wonder where tools such as InSTEDD’s SMS GeoChat can be used here too.

More to come on “getting updates that matter” later, this is just some initial thinking on it. I’d love to hear your thoughts too.

Ushahidi in the Congo (DRC)

When we pushed the first version of Ushahidi live in Kenya, I was trying to juggle that as I spoke at a conference in New York. Today, we’re deploying the new Ushahidi Engine (v0.1) into the Democratic Republic of the Congo (DRC), and I’m in Rhode Island speaking at another conference. I’m starting to see a pattern emerge…

Reporting Incidents from the Congo

The DRC deployment can be found at http://DRC.ushahidi.com, and the mobile number to send SMS reports to is +243992592111.

Ushahidi Deployed to the Congo (DRC)

Note: This is the alpha software for Ushahidi. If you find any problems, please submit them to bugs.ushahidi.com.

How you can help

Get the word out. Let people know the mobile number (+243992592111) and website (drc.ushahidi.com). Help get word to the Congolese on the ground in the DRC of this tool, that’s who needs to know about it.

Things are serious in the Congo… They are bad, very bad. As Sean Jacobs states:

“Since August this year at least 250,000 people have been left homeless in Eastern Congo in the latest outbreak of a civil war described here as between government troups and a rebel group claiming to protect ethnic Tutsis. At least 2 million people are refugees from that war which dates back to 1996.”

It’s a difficult situation, with a swirling mixture of militia and armed forces, compounded by particularly brutal and confusing activities. External military forces, years of displacement and a misinformation mar the landscape.

A new Ushahidi, a new test

To be quite honest, we’re a little nervous, just as we were the first time. The new engine still has a few bugs, and there are some process flow issues that we’re still trying to get figured out. This time we’re backed up by a group of competent developers who are working to get things straightened out. Want to help us make it better? It’s an open community, and we’re looking for your input.

We are VERY interested in hearing from you on how we can make the system better. If you have ideas, thoughts, comments – tell us. Leave them in the comments here, on the Ushahidi blog, or on the Ushahidi contact form.

This is a test of the system, albeit a very difficult one, but it will affect the way the software is changed, modified and upgraded in the next version. What we get right here, we can make work for you in your area when you need it.

How SMS messages route through Ushahidi

This simplified graphic was created to show how SMS messaging moves through the Ushahidi system – it’s a 2-way communication cycle.



SMS Reporting Through Ushahidi, originally uploaded by whiteafrican.
  1. An SMS gets sent to a local number
  2. It passes through FrontlineSMS
  3. This syncs with Ushahidi
  4. The message shows up on Ushahidi
  5. Admins can decide to send a message back to the original sender

We use FrontlineSMS so that we can provide local numbers in areas where the larger SMS gateways don’t operate. For instance, if you were to try to run this in Zambia, you’d probably get a UK phone number if you went through Clickatell. However, we do use Clickatell for the messages that we route back to the original sender due to cost savings. They also have a very nice, easy to use API.

Ushahidi Funding and a New Website!

Most of June I spent in Kenya, much of that time talking to developers and getting ready for the next big Ushahidi push. During that time there was a new article about Ushahidi being one of the “Ten Startups to Watch” in the Technology Review, which was exciting for us to say the least!

July and August have been spent working hard on getting the application rebuilt, the site redesigned and creating partnerships with other organizations. September is about launching the NEW Ushahidi.

A New Website

Now we’re off and running with a new website design, live today, that shows how our goals and focus have changed since things blew up in Kenya. (get a new Ushahidi button for your site.)

Funding

I’m very happy to announce that we’ve secured more than the $25,000 prize money from NetSquared (which has allowed us to do so much already). We have also just secured a grant of $200,000 from Humanity United!

Humanity United is an independent grantmaking organization committed to building a world where modern-day slavery and mass atrocities are no longer possible. They support efforts that empower affected communities and address the root causes of conflict and modern-day slavery to build lasting peace.

There is an obvious fit between Humanity United and Ushahidi, after all, we were founded on the same beliefs back in January in Kenya. Though we’re creating the Ushahidi engine as an open source project, our goal remains to see it used to better understand, give warning of, and recover from mass atrocities.

The Vision

Ushahidi is moving from being a one-time mashup covering the post-election violence in Kenya to something bigger. We are setting out to create an engine that will allow anyone to do what we did. A free and open source tool that will help in the crowdsourcing of information – with our personal focus on crisis and early warning information.

We see this tool being used in two ways:

  • First, to crowdsource crisis information by creating an online space that allows “everyday” people all over the world to report what they see during a crisis situation, and whose reports are generally overlooked or under reported by most media and governments.
  • Second, make that software engine free and available to the world, so that others can benefit from a tool that allows distributed data gathering and data visualizations.

We’re aiming to release an alpha version of it in just a few weeks for internal testing, and for alpha testing with pre-screened pilot organizations.

Volunteer Devs, Designers and Others

One of the reasons Ory and I were in Kenya was to talk to developers about helping with Ushahidi. We were overwhelmed with the amount of interest and the quality of the people who stepped up. So far we have a team working on mobile phones, a designers group, and a number of PHP experts. Go ahead and take a look at the development wiki as well.

If you’d like to play a part, get in touch and we’ll see where you can best fit in. You don’t have to be a developer or designer either.

[Credits: Richard “Ochie” Flores for the excellent design, Kwame Nyong’o for the beautiful illustrations, and Ivan Bernat for the spotless HTML/CSS markup.)

Press Release: Ushahidi Funding & New Website (PDF)

Older posts Newer posts

© 2021 WhiteAfrican

Theme by Anders NorenUp ↑