There’s a great resource that Built Visible has just written about the history and power dynamics of the internet’s undersea cables. It’s well worth a read, if even to just get a snapshot view of the surveillance programs that governments use to tap information, as well as seeing a map that plays by year to see how it all got built out.
I keep meaning to write blog posts on each of these reports on tech, most of them on Africa, but can’t seem to get it done. Instead, I’ll just post a link to each, a visual, and why I think it’s worth reading.
1. The Akamai “State of the Internet” Q3 2013 report
Has good information on overall usage globally, and trends. In Africa, even though they have a node in Kenya, all we’re seeing is stats on South Africa, Egypt and Morocco. However, there is a really fascinating chart by Ericsson in it on wireless usage.
2. GSMA’s “Digital Entrepreneurship in Kenya” report 2014
The GSMA puts together some fantastic reports, due to the amount of data at their fingertips due to their association’s membership. Alongside the iHub Research team, they’ve done a deep dive into the tech entrepreneurship side of Kenya, and you can see the results here.
3. Deloitte’s “Value of connectivity” report 2014
The Deloitte folks do a study and argue that an increase in internet penetration could have a large impact on an emerging market country’s GDP.
“Deloitte estimates that the resulting economic activity could generate $2.2 trillion in additional GDP, a 72% increase in the GDP growth rate, and more than 140 million new jobs.”
4. infoDev’s “The Business Models of mLabs and mHubs” report 2014
I’ve had a front-row seat to infoDev’s work starting and supporting places like the m:lab in East Africa. After doing it for 3 years, here’s their indepth report on what’s working, not working, how much money has been spent and what the future might look like.
5. McKinsey’s “The Internet’s transformative potential in Africa” report 2013
Mostly useful due to the interest large corporates and banks put in McKinsey, this report makes that the greatest impact of the internet in Africa is likely to be concentrated in six sectors: financial services, education, health, retail, agriculture, and government. What they’ve done particularly well is gather a large range of numbers from diverse and various sources to make better sense of what’s going on.
5 years ago I wrote about the news that Google had invested, along with others, in this new internet connectivity via medium-orbit satellites for the parts of the world that were hard to reach with terrestrial cable or even mobile phone towers, called O3b Networks.
Last week O3b finally launched.
A Russian Soyuz-STB rocket launched from Kourou in French Guiana today, 25th June 2013 at 19:27 UTC.
The rocket carried the first four satellites of the O3b Constellation. O3b will provide internet access for hard to reach parts of the world. 8 more O3b satellites will launch in a further two launches later this year and then in 2014.
Who are the first users?
First is Telecom Cook Islands, who will receive the first commercial signals across the network this summer and then Maju Nusa, soon to roll out a state of the art 3G backhaul network in Malaysia built on O3b’s low latency capacity.
The plans originally were to have these over Africa as well, let’s see if that happens.
Why do we rely on equipment made for the Berlin, Orlando and Tokyo when the conditions we have in Nairobi, Lagos or New Delhi are completely different?
Today we’re announcing the BRCK: The easiest, most reliable way to connect to the internet, anywhere in the world, even when you don’t have electricity.
We have a BRCK Kickstarter going, where we’re asking for your on taking it from prototype to production.
The BRCK is a simple, and it came from us asking:
“How would we design a redundant internet device for Africa?”
It would need to do the following:
- A router for 20 people
- With 8+ hours of battery for when the power goes out
- That fails over to 3g when the Internet goes out
- That travels, so you become a mobile hotspot
- With cloud-based backend that supports every country
- On device with both a software and hardware API
As a web company, being connected to the internet when you need it is a big deal, small outages cause lag that ripple through the organization. Even in Nairobi with it’s 4 undersea cables and growing tech scene, we still have power and connectivity problems. Could we do something to scratch this itch of ours that would help others too?
Since we travel a lot, we decided that it needed to work in every country. The BRCK had to work when the power was off for a full day (8 hours), had to fail over to 3g internet when the ethernet didn’t work, it also had to work in any country we were in, by just changing the SIM card. At the same time we wanted it to be accessible for both software and hardware extensions by others.
Having a BRCK cloud means that you can login to your device from anywhere in the world, load apps and services on to it, such as a VPN, Dropbox or other services and also control sensors and other devices connected to the hardware. We think that the BRCK model of both a software and hardware API represents the glue that will make the internet of things work.
As Ushahidi we’ve always used simple technology to create tools and platforms that work for us in Africa, and which is also useful globally. This holds true for the BRCK too. We’re redesigning technology that’s been around for years, but making it work for our needs in Kenya.
A year ago I jumped on a plane from South Africa back to Kenya without my book and my phone battery almost dead. Funny enough, these happenstances which leave me bored and with nothing to do but think have lead to my most interesting ideas (I’m sure there’s a lesson in there somewhere…). I subsequently broke out my notebook and started sketching out what I thought would be a fun hardware side-project for Ushahidi’s core team that would give us something to work on, when we were too fatigued with the normal coding/work.
We live in possibly the most interesting time for technology in history, where we’ve created this incredible thing called the internet, connecting us globally while at the same time getting to the point where the people who can code software can also “do” hardware. An era where analog and digital are democratized and the making of both attainable by anyone with a computer.
Making things is hard. It’s harder in Africa. I can’t overnight an order of processors, boards or 3d printing filament here. There aren’t an over abundance of local fabrication facilities or tools, and the milling machine you find might be in disrepair and take you two days to calibrate. We’ve got our work cut out to create the right spaces for prototyping and small-scale fabrication on the continent.
We actually started with Jon Shuler doing a lot of the early builds being done by him at his home in California. I’d bring these builds back to Brian Muita and team in Kenya where he was hacking on the firmware to make the system work. All the while hoping that air travel security would let me through with what to all appearances looked like a remote detonation device…
By prototype version 5 we were in Nairobi with a bunch of plastic, using the University of Nairobi’s FabLab to mill the body. There was a fair bit of repair and adjustment needed on the machines to make it work. Like most things in Africa, you either fix what you have or you don’t do it, because there isn’t another option. After a couple days we got it within close enough allowances that we could do it. It still wasn’t pretty, but we knew it would work by then.
That was all just the hardware bit. Concurrently we wireframed the software side, ensuring that this device was much more useful than just a MiFi on steroids. The BRCK Cloud falls directly in Ushahidi’s software development wheelhouse, so we set about creating a simple responsive interface that would work on both phones and big screens.
The software side does three things:
- A simple setup interface with only 3 form fields. Router setup is scary and hard, so we’re trying to take the pain out of it.
- A dashboard, so you can see if your BRCK is running on backup or primary power, how fast your current internet connection is, your provider, and how all of these have done over the last hour, day, week and month.
- A marketplace for free apps and services, as well as the place for others to offer up their own creations to the rest of the BRCK users around the world.
While having a device that was remotely programmable and that could run its own apps and service is important, we realized this was only half of the equation. We would need to create a similar interface for hardware creators and users. This means we needed the device to have hardware ports for everything to connect to, from temperature sensors to Raspberry Pi’s (as an aside, I want to get a Raspberry Pi hooked into the BRCK, thereby making a small, working server). We also decided to put special hex nuts at the top that would allow you to pop the top and get into the guts easily to do your own re-jigging.
The plan for the future is that you’ll be able to stack components under the BRCK like Legos, so that if you need an additional battery pack, a temperature sensor, solar charger, or other product you could do so with ease.
For a full rundown of the all that the BRCK can do, check out the Kickstarter. If you want to get into the real details, see the spec sheet.
This week I’m in Berlin to speak at re:publica – and as this post goes live I’m finalizing my talk. I find myself driven to tell the story of Africa’s great potential and growth, tempered by my experience building companies, communities and products here. I see the other entrepreneurs, hungry to create new products and driven by the same powers that are seen in their European and American counterparts. Here, it’s a harder road to hoe in many ways, it takes more grit, more determination and more belief in a future that is not yet realized to do it.
I look at the success we’ve had as Ushahidi and what this new hardware product means to us, and I’m humbled that we have the luxury to self-fund the R&D to get it to this stage, while so many my peers are struggling to take great concepts to even the prototype stage. The opportunities afforded us by our international awareness, the advantage of attracting and hand-picking the top talent that come through the iHub, the ability to have funds that we can risk on a half-baked original idea, a Board who believes in us and trusts our decisions – these are what I’m grateful for.
For this same reason, we’re committed to making a difference for our friends and peers in Nairobi. We’re going to build a makerspace through the iHub that allows others to start from a better position. A place that will give hardware hackers and entrepreneurs a chance to get trained on tools and machines, meet their peers and take risks on their own crazy half-baked hardware ideas. We’re calling this Gearbox.
We’re looking for corporate, academic and other partners right now to make it a reality. I’ll write about it more at another time (as this post is already too long). However, if you’re interested in being a part of this initiative, do let me know.
99% of Kenya’s 6.5m internet users access it via mobile, of which Safaricom owns 77% marketshare.
In Kenya, when you buy a 1.5Gb internet bundle from Safaricom you pay 1000ksh (~$12). You’ve paid for the data, and there is no additional cost to Safaricom if you were to use that data today or a year from now. The whole concept of data bundle expiry is ridiculous, as noted by Safaricom CEO Bob Collymore when he visited the iHub:
“When you go into a petrol station and fill up your car, does the owner of the petrol station tell you to bring it back on Wednesday to take back what’s left in the vehicle? Of course not. So I ask, why the hell are we doing that?”
Bob goes on to say that he isn’t going to be an apologist for this practice, that there is a problem with leaving the data there ad infinitum. That 60 days is probably too short and that Safaricom does need to change how they handle this.
- Until recently they just held your data hostage. If your data expired, you could recharge with just a few shillings of data, this would re-trigger your “old” data that was past the expiration, and have that available to you again.
- Today, it is “data gone, money stolen” after expiration. They cut you off if you haven’t used all of your internet bundle in the nominal 7-90 days, no matter how much is remaining.
I brought this up with Bob Collymore, and his chief executives when they visited the iHub earlier this year (see video), at which point he admitted that it was indeed a dubious practice that would be changed to something much more open to users. You’ll see what Bob says at the 1:17 mark in the video below.
Here Bob is on video speaking to this point (I’ve saved the link to go to the right point in the video):
The other day I caught a Tweet from Sunny Bindra about some surprising changes:
Safaricom is actually very responsive on Twitter, probably the best big company on social media in the Kenya. They followed up with Sunny with this:
So, Safaricom didn’t broadcast this significant change in the way data bundles are handled broadly. Apparently, “publicized on our website” means quietly posting a PDF somewhere in the morass that is their website to notify the data using public of the changes.
If you follow the links to the PDF, you’ll find the following:
What is the Validity period?
This is the time frame that you have to use the bundles, when this period elapses it means that any remaining bundles will have expired and will not be available for use.
(Note: there is conflicting information on how long bundles will last, you can only find out by topping up a bundle. I did this for 1.5Gb and found that it’ll last 80 days, not the 30 that they say in the PDF. I don’t know if it’s more/less time for other bundle amounts.)
It’s in Safaricom’s best interest for you to keep buying more data, over and over, even if you haven’t used it. It costs them nothing to let you use it over a longer period of time, or to keep recharging it.
I’m disappointed with Safaricom, especially after Bob Collymore came to the iHub and said he was going to fix this, not break it further.
This is an outright fleecing that the Safaricom team should be questioned on. In a country where they are the monopoly player on the primary source for people to access the internet, this makes them appear like a bad actor.
Basically, we’ve gone from a bad system that was promised to be made better, but which had a corrective option, to a worse system that has no option.
Other Safaricom Data Miscellany
While I’m at it, let’s go ahead and talk about a few other ways that the data service that Safaricom raises the bar for bonehead usability: buying data bundles themselves.
You used to be able to send airtime to a SIM card on your Safaricom modem. Then, using the inbuilt Safaricom Broadband app, send an SMS to 450 with the amount of the bundle that you wanted to buy, now 450 only seems to work for checking your balance.
With the new service updated in the aforementioned PDF you can now only use the USSD code to update it.
Take the SIM card out of your modem, load it in your phone and do the USSD code. Once confirmation is received, switch that SIM card back to the modem.
Yes, that’s correct. Instead of being using the software that comes native with your modem, you now have to use a phone to update your bundles. Why would you change your system to not work with everything that people use? I’m quite curious actually. I can’t understand this decision from a either the business or the product side at all.
Safaricom wanted to make it easier for people with modems, iPads, Android tablets and smartphones to be able to update their bundle (good idea). They created http://portal.safaricom.com/bundles for this purpose. Let’s say you’re out of data, you have no credit on your phone. How do you get to this page?
There are none. You’re stuck because this page isn’t zero-rated. This is mind-boggling in it’s oversight. I have no data, therefore I cannot go to your page to load more data. Seriously… who is the genius that thought this up? Or, probably more accurately, what form of bureaucracy is in place that allows this mediocrity to persist?
Further, if you’re Safaricom who controls 77% of the consumer internet access in Kenya, why wouldn’t you zero-rate your whole Safaricom.com domain and make it free for anyone to surf, even if they don’t have a single shilling on their phone?
[As a resource, here is the latest quarterly Communications Commission of Kenya (CCK) PDF report on the tech scene in Kenya.]
The iHub Research team has worked up an infographic on Tanzania to match their past ones on Kenya and Uganda. We’re looking at 50% mobile phone penetration in Tanzania, with about 22 million connected, where Vodacom has the largest market share at 42%.
The crazy stat is online: In Tanzania, only 2.5% of the population has access to the internet, 80% of those on mobile phones.
Hats off to Patrick Munyi (@ptrckmunyi) for the great design!
In January I wrote about the way the Google Global Cache is affecting Uganda – how local web caching is completely changing the internet user experience for that country. We’ve known for a couple weeks that this was underway in Kenya too. Well, here are some numbers on that.
Here’s the aggregate month:
We’re seeing the overall traffic increase 300% from around 100Mbs to around 400Mbs. Those are some pretty impressive numbers, no matter how you look at them. Why is KIXP/TESPOK not making some noise about this significant achievement?
How does it look across the ISPs that are using it?
KDN hosts the cache:
In the words of Steve Song, Village Telco is “an easy-to-use, scalable, standards-based, wireless, local, do-it-yourself, telephone company toolkit”. He’s just put out a new video making it very clear just how useful this system is.
The team over at Blinktower has done an exceptional job of creating a short, concise and eminently understandable video of what Village Telco is.
Often, we get caught up in our high tech wizardry and get overly excited about the newest Android app or the best new web app built in African Megalopolis #5. And by “we”, I mean “I”, since I too am a tech guy who is endlessly intrigued by the latest, newest and shiniest.
What we forget is the village. “Up country”. What happens when we get comments like this last week from the new CEO of Safaricom, Bob Collymore, threatening to do away with their rural network:
Weâ€™re OK with losing market share (faced with unrealistically low rates) and focusing on Nairobi and high-income communities. The people in remote districts are receiving calls (more than making them). If rates decline, why should I continue to do that?â€
Some rural communities have never had connectivity of any kind, voice or data. Others have it now, but could lose it if their revenues don’t prove to be high enough for big operators. Who is going to fill that niche?
I think the answer lies in technology like Village Telco. It’s a business, not an aid program. Where an entrepreneur can get a link to the network started (or not), and then mesh out from there to the whole community. People pay for access, and profits can be made.
For the last few years, a dedicated team of enthusiasts have been building the initial hardware and software. Both of which are open source. It’s a low-cost way to get into the telco business. Here’s to hoping that more entrepreneurs take a serious look at rural connectivity.
If you’ve been wondering what the numbers look like for Kenya’s mobile and ISP space, look no further than the latest CCK Report (Communications Commission of Kenya). It’s one of the best documents that I’ve seen, compiling information that you just can’t seem to find anywhere else.
Highlights of Q4 2010:
- There are 22 million mobile subscribers in Kenya
- 9.5% mobile subscriptions growth, which is increasing over the previous quarters
- 6.63 billion minutes of local calls were made on the mobile networks
- 740 million text messages were sent
- Prepaid accounts for 99% of the total mobile subscriptions
- The number of internet users was estimated at 8.69 million
- The number of internet/data subscriptions is 3.2 million
- Broadband subscriptions increased from 18,626 subscribers in the previous quarter to 84,726
Everyone recognizes the impact on SMS and voice, due to the price wars brought on by Airtel last year. The average, people are paying 2.65 Ksh per minute for voice representing 33.4%
reduction on pre-paid tariffs. It comes as no surprise that there was a 68.4% increase in traffic during this period, nearly triple the norm.
Interestingly, a decline in total number of text messages sent (4% less) was recorded. It’s an indicator that given the choice of lower cost voice, people would rather use that, and they do.
Safaricom lost 4.8% market share, from 80.1% to 75.9% (still massive). Surprisingly, it wasn’t Airtel who benefitied, as Orange made up for most of that with a 4.4% increase of their own. Airtel did lead the market by recording 1,143,353 new subscriptions, about 3x their closest competitor.
A whopping 99% of the internet traffic in Kenya is done via mobile operators, meaning 3G, Edge or GPRS. It’s to Safaricom’s credit that they moved on this early, not dithering around on data as their competition did, effectively taking the whole market.
My theory is that there are only two major players in the ISP space in Kenya. The first is Safaricom, supported by this report, who will own most of the country due to having an island strategy (mobile towers). This allows them to own all the rural areas and anyone who needs decent speeds and has to be mobile.
The other is the fiber bandwidth provider (ISP) who figures out and cracks the consumer market. The closest to doing this is Zuku (Wananchi) who started rolling out 8Mb/s high-speed fiber-to-the-home internet connections in Q4 2010 at only 3,499 Ksh ($45). These numbers aren’t reflected yet. My guess is that we’ll see Zuku tying up all the home internet connections in the major urban areas.
Estimates for those with internet access in Kenya is closing in on 9 million users, and at over 22% of the population, we can say we’re getting a lot closer to the critical mass needed for real web businesses and services to thrive.
Overall, the numbers on both mobile and internet are trending up, and at a very favorable rate. The indicators here prove that you should be paying a lot of attention to mobiles and data connectivity in Kenya.
If you’re a business, what’s your mobile plan? How are you providing and extending your services over the internet (and no, a website is not enough)?
If you’re an entrepreneur, how are you going to use this information to decide what to build? Are you paying attention to the wananchi, building apps for the upper class?
PDF of Report: CCK Report download – Kenya Q4 2010
I get culture shock every once in a while, and it’s not the normal type where you’re coming to a new country and everything is completely different than your own country. This is more subtle, I’m at a conference with a lot of people who look and sound like me, but when you actually listen to their conversation you realize that they define themselves and the world in a way just slightly different than you do. That’s what happened to me over the last 3 days here in Boulder, Colorado at the IxDA 2011 – the big Interaction Design Association annual conference. I’m surrounded by 600+ designers, people who think deeply about why you and I do things, and ways to make us do it better, differently or for more money.
Africa’s Digital Design Constraints
I was fortunate enough to meet Jon Kolko, one of the organizers, at PopTech a couple years ago, leading to this invite. My role was to talk as a practitioner, and I covered everything from AfriGadget to Maker Faire Africa and Ushahidi. I then delved into the constraints around design and building in the African tech space, by breaking down the three main areas that I see:
Specifically, I covered how bandwidth has made it difficult for people to create new sites and services, but more importantly, how the uptake of those is limited by consumer use of the internet due to costs and speeds. This is changing though, as tracked and evidenced by the lowering data costs and increased bandwidth being piped into the continent each year.
I also covered the swiftly blurring lines between Mobile and web. How due to the fact that mobiles are the primary device for Africans and usually the first device that people have a meaningful interaction with the internet on, is creating a different type of user. How the entrepreneurs in Africa’s web space are thinking of it from a mobile context and how they build services to address their audience. Here I got into the argument of diffusion of internet penetration via the big international players like Facebook and Google through mobiles, which then open up infrastructure and cultural use making it more accessible to local startups.
Finally, I talked about culture. How this culture of mobile first plays out. Where the phone number trumps the email address on user signup, and where transactions happen due to that norm. It’s here that I also got to bring up one of my favorite people, Jepchumba, the creator of African Digital Art. She is creating a community, and a movement, to get African designers talking to each other and showcasing their work to the world – breaking down the stereotypes and building up new personalities across the continent.
Jepchumba helped me come up with some of the content behind my talk due to running her African web design survey last week (it’s still open). There’s a lot of information in that survey, much of which is still being gathered. As an example though, is this chart showing the percentage of African web designers who are self-taught as opposed to having a formal education. I wonder if this is normal globally?
Designers as Explorers
Getting back to my starting point. Sometimes this culture shock leads to great conversations, and it allows me to see the world that I live and work in a slightly different way.
Erin Moore is a designer and a storyteller, usually through video and blogging (see her newest project on Kickstarter). She introduced me to this terminology of “designers as explorers” – something that might be very apparent to the IxD field, but foreign to me. It’s a phrase that fits. Where we see designers as a new generation of what we thought of as National Geographic explorers a century ago. They’re best embodied by the Jan Chipchases of the world, who spend a great deal of time watching, listening and understanding how design interactions work, and then translating those discoveries to the rest of the world.
It fits because I have a hard time with a lot of the well-intentioned design community thinking that they can parachute into places like Africa, usually with a solution already in mind, and change the world. There is a place for designers in Africa, but the greatest value lies in recognizing the expertise at the local level, the inventiveness and ingenuity already there, and rubbing shoulders with them in a way that both gain value and maybe even build something new.
Ana Domb is another of the unique people that I met here at IxDA, she’s studied at MIT and has a good steeping in both digital technology, mixed with a focus on media and understanding fans (the people kind). It was this background that took her to Brazil (she’s Chilean) to study Technobregas – a crazy hodgepodge of fans, artists, sponsors and DJs all banding together to create their own music reality, outside of the traditional music industry’s grasp. It takes someone with a distinct design focus and understanding of how social interactions happen to be able to translate that to someone like me (paper here).
We need to see more of this. Where American designers do parachute in, but not as problem solvers, instead as explorers. Where their expertise rubs off on those they meet, and those they meet rub off on them. Both benefit. Equally, we need to see more African designers going abroad and using their expertise in shaping the way the Western world uses technology and understands community. Design interactions go both ways.