3 Numbers that explain the digital divide

Written by Michael Kende

Thanks to the growth of the mobile internet, there are now three numbers that are relevant to the global digital divide – 94, 50 and 36. All three are amazing, and underlie a new way to approach the digital divide that focuses on availability, affordability and relevance.

  • 94% of the world’s population can receive a mobile telephone signal, which represents growth that almost no one could have predicted 15 years ago.
  • 50% of the world’s population can receive a mobile internet signal, because a mobile network can be upgraded to offer internet with far less investment than building the original network.
  • 36% of the population has subscribed to the mobile internet, from a standing start just six or seven years ago, thanks to the widespread availability of and access to smartphones with millions of apps.

These numbers mask significant regional variation, of course. In developed Asia Pacific, 99% of the population has a 3G signal, and 109% have subscribed (some people have multiple subscriptions). On the other hand, in sub-Saharan Africa, where 82% of the population has a mobile signal, 35% have a 3G signal, and so far only 11% have subscribed to mobile internet.

The common thread in all regions is that availability of the internet is no longer the limiting factor – mobile internet is always available to more than those who have adopted it, and can grow relatively easily to cover the entire mobile network if needed.

The key question should be why potential users who could access a service have not done so. For instance, in sub-Saharan Africa, why have only a third of those who could access 3G taken advantage of its availability?

The answer has two parts. First, of course, affordability is a key issue: in some countries, broadband still costs 10% or more of average monthly income and is thus out of range for most. Second, though, is relevance. Is there content available in the local language? Is it of interest? Useful? If the answer to these questions is no, then chances are many who could afford internet access will spend their time and money elsewhere.

In many countries, relevance is now the most significant issue. For instance, in a recent survey in Brazil, subjects were asked to give one or more reason why they had not used the internet. Of the respondents, 25% indicated availability as a reason; 26% indicated cost; while 70% indicated it was a “lack of need or interest”.

So what can be done to help regions clear these final barriers and fully embrace the internet and its advantages?

Of course, everyone would benefit from lower costs – not just those who aren’t online today – and governments can help by removing any barriers to connectivity, such as high costs for deploying infrastructure, and high taxes on equipment, devices and services that act to depress demand.

However, at least as importantly, increasing the amount of content in the local language, with relevance to local needs, is critical. Governments can help promote content creation by developing their own mobile services, hosting them locally and promoting capacity-building to support these activities. Governments can also encourage the private creation of content by removing vague or restrictive laws with respect to content and liability.

As we collectively celebrate the amazing numbers already achieved by the mobile internet in closing the digital divide, we should also work hard together to make sure the remaining challenges are met so that existing and new users enjoy a mobile internet that delivers the hope and promise the internet can bring to everyone.

For a broader discussion of these issues, please check out the Internet Society’s Global Internet Report 2015 which delves deeper into mobile’s impact on the digital divide and a host of other issues related to the mobile internet, and the World Economic Forum’s Global Information Technology Report 2015.

Have you read?
How is mobile changing the world economy?
Which countries have more internet connections than people?
How better connectivity can transform Africa’s economies.

Sixteen fellows to be at AfPIF this year

Sixteen technology experts have been sponsored for this year’s Africa Peering and Interconnection Forum (AfPIF) conference, taking place on August 25th to 27th at the Girassol Indy Congress Hotel in Maputo, Mozambique.

The fellows bring the number of fellowships to a hundred, in the last six years. Since inception, AfPIF has endeavored to bring more African technology experts together, in effort to develop more solutions. In 2010 there were 5 fellowships, 14 in 2011, 17 in 2012, 28 in 2013, 20 in 2014 and 16 this year.

The fellows are drawn from the public and private sector, and are expected to deepen their understanding and develop solutions for peering and interconnection in their respective countries.  The Internet Society convened the first AfPIF in 2010 and it has become an important annual event for peering coordinators, ISPs, regulators, content and infrastructure providers.

Among other benefits reaped by the African Peering and Interconnection Forum (AfPIF) recipients will be;

  • Sharing experiences on ways to improve running and administering of a new, or existing IXP
  • Business opportunity to meet potential IXP members
  • Promote public awareness and evangelism of IXPs and peering in general at the national and regional level
  • Advance and influence national/regional policies on peering and cross-border Internet interconnection
  • Provide a face-to-face networking opportunity for peers and experts

Now in its sixth year, the three-day event will run on different themes; first day being “Peering Coordinators Day” that will highlight on interconnection and data center operations. The second day’s theme will be on “Lesson Learnt” where the impact of peering in Africa will be discussed. The last day of AfPIF will run under the theme “Regional Content Producers and Distributers” where Internet exchange point (IXP) and measurements will be discussed.

The Internet Society called for application end of April 2015, giving a one-month duration for individuals to apply for the programme after which the committee took three weeks to sort the applications and announce qualified individuals. To attend the AfPIF fellowship, persons must be actively engaged in Internet policy, business and technical development relating to peering and transit. Also, the applicant in question should be African who has previously worked towards the development of an IXP and is able to demonstrate proof of work done in either of the fields.

The participants will be eligible to Air ticket to and from Maputo, Mozambique Hotel accommodation for the duration of the event.

Meet the 2015 fellowship recipients

Read on the Internet Society News Release announcing AfPIF 2015 Fellows to Attend Africa’s Premier Peering Event in Maputo, Mozambique

Google sees value in AfPIF meetings

In 2007, Google entered the Sub Saharan market. At that time, it looked like a risky move, given that Africa was still grappling with expensive satellite connectivity and efforts to lay fiber optic cables were underway. To some, Africa looked a long way from affordable connectivity while others saw it as just ripe.

By 2010, fiber optic cables had landed and connectivity costs were beginning to fall considerably. That year, Google supported the first Africa Peering and Interconnection Forum (AfPIF).

Mike Blanche was part of the Google team that was in Nairobi, talking about Google and its efforts in Africa. We asked him three questions:

1. AfPIF: Google has supported AfPIF for the last five years, how has the peering and interconnection landscape changed?

Five years ago, the modern high-capacity submarine cables that link Africa to the world were in the process of being laid, and the challenge was to develop demand for Internet services, terrestrial infrastructure and local interconnection points.

Now, we are seeing successful regional interconnection happening within Southern and Eastern Africa, with an increasing proportion of African traffic staying “local”. Lack of infrastructure and competition is still an issue in places, but the main challenge now is replicating the success stories to all corners of the continent, so everyone can benefit from fast, inexpensive, reliable Internet access.

2. International companies have different perceptions when dealing with Africa, what are some of the lessons learnt?

Google was one of the first international companies to invest in infrastructure in Africa, and over the years we’ve seen more international operators attend AfPIF and learn about the fast-developing African internet ecosystem. I think we have all learnt that Africa is a diverse set of different opportunities and challenges, and each country requires a slightly different approach.

3. I know the GGC is not openly spoken about, but how has AfPIF participation affected Googles entry into African countries? (I spoke to a techie from Benin who told me that speaking to Google team at AfPIF helped understand what is needed and probably sped up the process. I am looking at instances where Google maybe entered a country out of their initiative.)

AfPIF provides great value for Google, bringing together in one place those focussed on developing Internet connectivity and interconnection. AfPIF has helped to build a community of people who are building the fabric of the Internet across Africa.

Without the contacts made and relationships forged at AfPIF, we would not have been able to extend Google infrastructure across Africa.

AfPIF Participation Helps Get the first Google Global Cache in Benin

By Vivien Assangbe Wotto

In 2010, African techies, ISPs, content, and infrastructure providers met in Nairobi, Kenya, for the inaugural African Peering and Interconnection Forum. This event was envisaged to be the forum where African in the tech sector exchange ideas and develop lasting solutions to some of the problems.

The ability to develop solutions became a reality for Vivien Assangbe Wotto, an Engineer in Telecoms and Networks Mobility at Benin Telecoms SA.  After participation last year,  Wotto was able to link up with the Google team,  and now Benin Telecom hosts the first Google Global Cache in the country.

Wotto shares his experience after the meeting.

What was the goal for attending AfPIF last year?

My goal for attending this forum was to better:

  • Know more about Internet Exchange Points (IXPs), interconnection and peering
  • Know its mode of operation, these benefits and see any opportunity for my company BENIN TELECOMS SA, and in particular for my country, Benin.
  • Know and make contact with the different operators of the IXP in Africa, the West African sub-region and the world.
  • Know what promotes the development of an interchange.

Admittedly, the development of local content is a major challenge that promotes the proper operation of the IXP.

The government allowed the implementation of an IXP, which will, operational by the end of July 2015. In Benin, we currently have five GSM operators and Internet service providers present at BENIN-IX, hosted at BENIN TELECOMS SA.

During the meeting in Senegal, I made contact with the Google Africa team, and initiated the procedure for installing Google Global Cache Servers (GGC).

Today it is a reality, since the servers of Google Cache are currently installed at the BENIN-IX, located in the offices of BENIN TELECOMS.

On the technical side, the experimental network BENIN-IX is set up and managed by a technical team composed of engineers from each operator or supplier present at the IXP. We had a lot of debate and exchange on how our IXP.

On decision level, the discussions helped me to understand the benefits of each member of an IXP, and the various models to be adopted for the sustainability of an exchange point.

Why attend AfPIF?

I go every year to:

  • Capitalize experiences
  • Share my experiences and to exchange among government policy makers, managers and technical team ISPs and IXPs.

Internet Society Brings African Diversity to AfPIF-2015

Selects 16 Fellows to Attend Africa’s Premier Peering Event in Maputo, Mozambique

Addis Ababa – The Internet Society today announced that it has selected 16 fellows to attend the 6th African Peering and Interconnection Forum (AfPIF), which will be held from 25–27 August 2015 in Maputo, Mozambique. This year’s AfPIF Fellowship programme attracted more than 280 applications from around the world, with the 16 selected fellows representing 16 countries.

AfPIF facilitates discussions on African Internet infrastructure challenges, including terrestrial capacity, national and regional Internet Exchange Point (IXP) development, local content development, connectivity costs, and international peering. The AfPIF Fellowship programme provides qualified applicants the opportunity to attend the event, share experiences and exchange ideas to advance and influence national/regional policies on peering and cross-border Internet interconnection.

“The AfPIF Fellowships are a significant part of our work. They aim to support attendees who would not otherwise have the opportunity to attend the event. As a result, this helps to increase the continental representation at the forum and to further the vision for an interconnected Africa,” said Dawit Bekele, Internet Society African Bureau Regional Director.

Mr. Bekele continued, “We are thankful to our sponsors who have enabled us to bring more people by supporting this fellowship programme, and to the Fellowship Committee for their diligent work in selecting these highly-qualified individuals to benefit from this year’s program.”

The 2015 AfPIF Fellows are:

•    Abdoulie Sowe (Gambia), Serekunda Internet Exchange Point
•    Adelard Kenese (Burundi), BURUNDIX
•    Foncham Denis Doh (Cameroon), Ministry of Posts and Telecommunications
•    Ikusan Charles Abimbola (Nigeria), Internet Exchange Point of Nigeria
•    Chris Oleke (Uganda), National Information Technology Authority
•    Emmanuel Antwi Kwarteng (Ghana), Ghana Internet Exchange
•    Ghislain nkeramugaba (Rwanda), Rwanda Information and Communication Technology Association
•    Ismail Settenda (Tanzania), Tanzania Internet Service Providers Association
•    John Bettie Gizea (Liberia), Liberia Internet Exchange Point Association
•    Koffi Kouadio Armand (Cote D’Ivoire), Cote D’Ivoire Internet Exchange Point
•    KOUMA Cyriaque-Didier (Gabon), Gab-IX
•    Mamothokoane Tlali (Lesotho), Lesotho Internet Exchange point
•    Mohamed Ben Yahia (Tunisia), Tunisian Internet Agency
•    Nico Tshintu Bakajika (Democratic Republic of Congo), Kinshasa Internet eXchange Point
•    Roderick Esquibal (Benin), Isocel Telecom
•    Yassia Savadogo (Burkina Faso), ARCEP

Following the formalization of the fellowship programme at AfPIF-2014 in Dakar, the Internet Society African Regional Bureau received encouraging feedback from the fellows. Many delivered tangible benefits to their local peering community and have shared their experiences in the articles posted on:

“We hope that the experiences shared will inspire others to contribute towards the vision of realizing 80 percent local and 20 percent international Internet traffic in Africa by 2020,” said Michuki Mwangi, Internet Society’s Senior Development Manager for Africa.

Details of the Internet Society’s AfPIF Fellowship are available on the AfPIF 2015 website.

About the Internet Society
The Internet Society is the trusted independent source for Internet information and thought leadership from around the world. It is also the organizational home for the Internet Engineering Task Force (IETF). With its principled vision and substantial technological foundation, the Internet Society promotes open dialogue on Internet policy, technology, and future development among users, companies, governments, and other organizations. Working with its members and Chapters around the world, the Internet Society enables the continued evolution and growth of the Internet for everyone.

Media Contact: Betel Hailu, hailu@isoc.org

France-IX – Lessons learnt

The French context was very specific. We had many IXPs in France when we decided to launch France-IX in 2010. So, before starting it was decided to run an enquiry to make sure there was a real demand for a change. Most of IXPs were run by operators (which may represent an issue regarding neutrality and independence), and were based on free business-model (scalability issue): Paris was not seen as a strategic peering place, as the peering market was very fragmented. The consequence was that many local players were peering abroad on wider and more efficient IXPs and content providers didn’t come to Paris to exchange traffic.

The market study shown that there was a global agreement within the French Internet community to have a federative IXP in France. A workgroup has been conducted to foster the change. Then France-IX was born.

Neutrality and independence

Since the 1st day, it was very important to make sure we would run an independent and neutral organization. So that it was decided to setup a dual core entity: an association “France IX” and a private company “France IX Services”, with the association as the unique shareholder of the private company. We could have run either the association or the private company but it was decided to run both: customers sign a contract with the private company and at the same time they become members of the association with one right of vote for annual meetings (to vote for closure of accounts, elect board members…), and the private company brings more flexibility for operations, commercial aspects, marketing…

To guaranty neutrality, the rules voted and approved are the same for all the members, independently of the size of the member.

Importance of board and board members

The founding members did constitute the first board of directors, to make sure that the founding members were able to help and bring either human or financial resources to France-IX. Some founding members also made equipment donations to help France-IX to start and avoid spending too much money. To maintain independence, it was decided to reimburse by the time the board members who had decided to support financially France-IX when it started.

After 3 years, it was planned to renew the board by third, with members being able to apply for a board seat: to avoid any dispute, the board members to be renewed have been designated by a draw. The very first board was composed of 6 seats, only with people representing their company, and since the board was upgraded to 8 seats, with still 6 seats for companies and 2 seats for individuals.

Partnerships

It was very important to consolidate the French peering market and solve the issue of the fragmented market. It was decided to setup partnerships with some other IXPs: partnerships have been established then with other French IXPs or IXPs very close to France (setting up cross-border interconnections).

Large footprint

When France-IX was launched, it was decided to setup directly 6 PoPs around Paris, to make sure this footprint would maximize our capability to collect the existing networks. We decided to run directly the PoPs interconnections using dark fibres and then integrate progressively WDM (using DWDM passive equipment) for scalability. Choice of passive WDM was particularly pertinent (versus active DWDM equipment) as we were running a metropolitan infrastructure where the distance between PoPs is less than 30km usually. It was very cost efficient.

Building a backbone and added-value Services

An Ethernet layer2 infrastructure was built initially, and it evolved to a VPLS backbone in order to deploy easily new services.

We decided to start with basic services (Unicast IPv4/IPv6 services, NOC support), to avoid complexity at the very beginning, but as soon as we grew it was important to be able to provide additional added-value services such as routes servers (to ease peering for new members), develop a nice an friendly portal for members and build an information system, make sure to ease access to DNS resources (national DNS extensions, TLD’s, DNS root servers), provide private VLANs (to allow private interconnects), provide a looking-glass, setup a reseller program… Now, after 5 years, we are about to release 100Gbps ports offers, start to provide SLAs…

Operating an IXP

We use to hear that it is easy to build an IXP from the technical side. It is partly true, but when you grow (by adding additional PoPs) engineering may become more complex, and your members will ask for more professional services.

Nonetheless, if you want to operate in an efficient way an IXP, it is important to define some rules/procedures to be able to grow fast and smartly:

  • Of course, when running services, you need to have a real NOC, available 24/7. In our case, we decided to outsource level-1 support and keep internally level-2 support (based on very skilled engineers having experience on WAN operators backbones or previous IXPs).
  • Pre-cable positions between the customer patch-panel and the switching platform in order to save time and avoid DC human works.
  • Automate configurations whenever possible.
  • Define/write procedures to connect members and apply it. In our case, we defined a very clear procedure allowing a new comer in the technical team to connect a member easily.
  • Define the Network Technical Specifications and share them with the customers. This will permit to have clean configuration on customer side and will facilitate the operation of the network.
  • Create a knowledge database and share it with the connected members.

Keep in mind that an IXP can grow very quickly and the time you invest initially is time you will save in the future.

Cooperation and community

When you run an IXP, you must always keep in mind 2 things: you can get support from the community (useful to start with), and when you grew it may be appreciated that you share your experience. We decided to join Euro-IX association, which allow us both to get support and share at the same time. However, as soon as the IXP is growing the best-effort model for human resources is not scalable. In the case of France-IX, at the very beginning human resources were brought by founding members and we were also getting some help from the community. Now we have a strong technical and commercial team, and all the people are employed by France-IX. However, the number of employees is not directly linked to the number of members connected to the IXP (even if of course when you have more members it does generate more work): the staff growth is also driven by the diversity of services we propose by the time.

Community is very important. When running an IXP, you must listen to your community and meet the people also. That’s why we participate in many events: any NOGs meetings (FrNOG, MENOG, NANOG), but also peering events like EPF, or events directly concerning IXPs (AfPIF, Euro-IX), global community events (Africa Internet Summit, Afrinic, RIPE…), and we also run our own events.

As mentioned above, we are now in a phase of sharing our experience: that’s why we are happy to contribute in project like AXIS (with great cooperation with ISOC) to help African countries (especially French speaking countries) to get their own national IXP.

Designing the network

The very first France-IX design was a pure layer-2 backbone, very simple. Since came a second round when we decided to deploy new PoPs and to integrate core PoPs to have a dual-core networks to optimize the mesh (and the cost of backbone links) and still propose a resilient infrastructure. During this second phase we also generalized the use of passive DWDM equipment to bring scalability for interconnecting PoPs. The very last phase we deployed in summer 2014 consisted in introducing a third core PoP and make sure then that all others PoPs were connected with fully redundant dark fibres paths to 2 of the 3 core PoPs and to generalize VPLS everywhere. We also did select equipment bringing much more density for 10Gbps ports and allowing 100 Gbps connections.

Last time we decided to renew the full backbone to have equipment with much more port density and also integrate more flexibility within the network by generalizing VPLS, we did setup a lab which did simulate the whole new core network.

However running a Proof of Concept before deployment does not prevent from bugs or outages, even by generating some traffic during a large period of time. So it is very important in our case to get a direct and efficient support from the equipment vendor (avoiding intermediaries), as we are concentrating a very large numbers of connected networks. In such context, when we run our tender to select the equipment vendor, a focus was not only done on equipment features but also on support.

In order to optimize equipment life duration, when we renew equipment from core PoPs, we globally try to re-use them on edge PoPs or keep them for brand new PoPs (and then we upgrade the equipment when a new PoP grows and does reach a critical mass).

We also decided to deploy our latest backbone based on 2 vendors. For the choice of optical modules we also work with 2 vendors.

At the very beginning of France-IX, we were pretty flexible regarding connections of new members. However flexibility may please a new member but it does not scale. Now, we apply very strict engineering rules (allow only single MAC address, filtering rules, check for unauthorized protocols…), but to make it work it is important to publish them so that everyone knows the rules. Strict rules could displease some members, but it’s the way to ensure the stability of the IXP when you have hundreds of connected customers.

Transparency

The IXP community covers a large amount of Internet players. It is important to be transparent with this community providing information about how France-IX works on the technical side. For example, a network weathermap and traffic statistics per PoP are available on our website. We also send to our members (trough a dedicated members mailing-list) a quarterly report detailing the main actions undertaken during last months.

How to make sure the IXP will last?

There are key points to be matched, to make sure your IXP will last:

  • Organization
    • Keep and guarantee neutrality and independence
    • Keep diversity of representatives and members type within the board (CDNs, operators, Carrier Neutral datacenters, etc)
    • Remain transparent about the way the IXP is run and managed (including accounts presentation to members)
  • Services
    • Provide a professional service as soon as the IXP is increasing (and not only a best-effort service), with a real 24/7 NOC
    • Listen to your members and their expectations, but also be proactive to propose new services
  • Business-model
    • Avoid free models (scalability issue)
    • Apply fair and transparent prices (public prices)
    • Keep in touch with the market and the real players
    • Identity new players/prospects
    • Integrate a marketing/communication process (public relations, website, social networks, events…)
  • Think about extensions
    • Check about possible partnerships with other IXPs
    • Development of new PoPs
    • Check about cables infrastructures (especially submarines cables when there are)
    • Develop reseller program to bring more members

Future of France-IX

France-IX is now 5 years old (we did celebrate our 5th year a few weeks ago). When France-IX was launched the idea was to have a strong footprint on the 1st day (by starting with 6 PoPs) and to gather French Internet community, and of course bring back France as a visible international peering place. Since, we have more than 275 connected members, around 400 Gbps of traffic, 10 PoPs in Paris and 2 PoPs in Marseille.

Significant developments have been done on France-IX backbone infrastructure during the last few months, in order to prepare for important growth of traffic and number of members. During summer, we will insert 100 Gbps ports linecards in several of our chassis. This offer will be available on core PoPs in Paris, as well as in Marseille. Paris and Marseille will continue to be run as 2 differentiated IXPs, even if a gateway is available for small members who need to peer on both by running one single connection. For larger members they will need to go through one of our resellers if they want to get connectivity to Paris when connected in Marseille, or even simpler connect directly to both Paris and Marseille.

We are also working on new services, like SLAs, market place… to make sure we match our community expectations. They will be available by the end of the year. Of course we will develop new services by keeping fair pricing, because of the model of our organization (not for profit association owning a private company).

We also keep in mind to either develop additional PoPs in other French regions, or establish partnerships with new IXPs in France when developed outside Paris or Marseille, as our goal is of course to continue gathering French community.

We have received requests to setup PoPs outside France. However, we also keep in mind the reason why France-IX was created, and the aim is not to develop everywhere in the world just because some others are doing it. By default, we prefer to work in a cooperation mode, in which we help and support other IXPs initiative and potentially partner with them. Even when setting up new partnerships we also consider the mutual benefits rather than only looking for new commercial opportunities. Access to French contents and partnering with French speaking countries is also a key point in our strategy. We receive many requests for partnering or even setup interconnection with us, but the point is not to collect as many partnerships or interconnections as possible. You need to have a global strategy and serious reason to do it.

Meet France-IX

France-IX will be present in Mozambique for AfPIF meeting, and will be one of the sponsors of the event. This is a great event to share experience and talk to other people who are willing to setup IXPs in Africa. Our team will be happy to meet you there!

More information about next AfPIF meeting

Additionally, France-IX is organizing its annual meeting in September in Paris. Anyone interested in France-IX can attend the meeting.

More information about this meeting and registrations

Liquid Telecom makes major strides in local content

By Rebecca Wanjiku

Africa’s consistent investment in ICT infrastructure and an uptake of e-government services has allowed Liquid Telecom to increase local content passing through its network to 50 percent.

Three years ago, only 20 percent of the content carried by Liquid in East, Central and Southern Africa was local. In the next few years, the regional operator expects to reach the 80 percent target set by the Internet Society.

“Three years ago, we had 20 percent local content in our infrastructure, this year, we have 50 percent, a 10 percent increase every year. This shows its is possible to achieve 80 percent local content,” said Ben Roberts, Liquid Telecom Kenya CEO and CTO of Liquid Telecom Group.

Roberts expects the overall local content in Africa grow, with video playing a major role in the demand for increased bandwidth. He projects that the demand for online streaming services will increase, as has been shown by the expat and local community in Kenya, that has found a way to bypass internet blockage and get services from companies such as Netflix.

“Netflix doesn’t offer services in Kenya, but it doesn’t mean that Kenyans cannot access services. Some people are using Virtual Private Network (VPN) services to access content on Netflix, this shows there is demand for video services, which will drive up the demand for higher capacity,” added Roberts.

In January this year, Netflix indicated its intention to grow globally to 200 countries. Netflix CEO Reed Hastings and CFO David Wells made the statement while announcing Q4 2014 earnings.

“We already offer Netflix in about 50 countries and have learned a great deal about the content people prefer, the marketing they respond to and how to best organize ourselves for steady improvement. Acceleration to 200 countries is largely made possible by the tremendous growth of the Internet in general, including on phones, tablets and smart TVs. We intend to stick to our core ad-free subscription model. As with our initial round of international expansion, we’ll get some things wrong and do our best to fix them quickly,” the statement said.

Africa has had a burst in growth of video production, spearheaded by Nigeria, Ghana and South Africa. Although these local productions target Africans in diaspora, with availability of affordable bandwidth, local audiences are expected to rise.

Nigeria’s iRokoTV, Pana TV and Ibaka TV are probably the most popular online but Bozza and Wabona have also been growing their audience online. A UN report estimates that Nollywood injects about $500 million to the Nigerian economy.

Roberts identified piracy and lack of enforcement of Intellectual Property laws as a major hindrance for companies interested in investing in streaming services; whether local or international.

“If you can get the latest Hollywood movies at a cost of a dollar in downtown Nairobi, it becomes hard to convince one to pay five dollars to stream the same movie, this points to a weak IP law enforcement, which can kill business,” said Roberts.

Roberts notes that an increase in local productions at whatever quality and level will help grow the demand for video, while stricter laws and enforcement of international IP laws may discourage people using VPNs to bypass restrictions imposed by streaming providers.

Regarding growth plans in the region, Roberts said that each country is adopting country specific customer care services to ensure better delivery and client satisfaction. Given that different countries offer various levels of infrastructure services, Roberts said it would be hard to have one approach in all the markets.

Digital TV migration to provide business for Kenya’s cloud provider

By Rebecca Wanjiku

Digital TV migration and the regulatory requirement of 60 percent local content is expected to increase demand for local hosting solutions, as businesses seek to reduce costs and beat competition.

Angani Limited, Kenya’s growing cloud services provider is expecting its business to grow, propped by migration from analogue to digital TV and the demand for faster and affordable digital services by media owners.

“Local content has recently experienced a shot of adrenaline because of the Digital TV migration. This has spurred great interest in producing and distributing local content, simply because, for the first time in a long while the barrier to entry for television broadcasting is very low,” said Phares Kariuki, Angani CEO.

The Communications Authority of Kenya (CAK) has developed a roadmap aimed at increasing the local content aired on local channels from 36 percent currently to 60 percent in 2018. The Authority has partnered with Kenya Film Commission (KFC), which is responsible for development of the local film industry, and the Media Council of Kenya in order to help local investors take advantage of convergence technologies.

“Angani, seeing the opportunity to further reduce the costs of operation for the various television stations has now provided a platform that allows producers to distribute their content online/through regular digital TV/satellite or cable. This is opening up a world of possibility for producers whose only mandate now is production,” said Kariuki.

The agreement between CAK and KFC provides for collaborative policy interventions in addressing challenges faced by local content producers such as piracy and unhealthy competition from foreign film products; crafting of the minimum quality standards for local programmes and partner in forums to promote increased production of local content.

“Mutually beneficial partnerships such as what these we have signed today, will tap into each organizations strength and this will go a long way in facilitating the growth of the broadcasting sector,” said Francis Wangusi, CAK Director General.

The CAK is also working with the ICT sector to identify solutions that will make transition to more digital content less expensive, efficient and provide job opportunities.

The growth at the Kenya Internet Exchange Point has given content carriers, hosting providers, ISPs and other infrastructure providers to lower operational costs by peering locally.

“KIXP has provided the platform that allows for Angani to exist; by ensuring local peering, we ensure that it’s cost effective for ISPs to use the exchange as it saves them international transit. Our offering then becomes compelling for both users and ISPs as it offers lower latency links whilst saving the ISPs money, hence increasing the number of peers and the amount of traffic being exchanged at the KIXP,” added Kariuki.

One of the challenges is getting ISPs to open up local infrastructure, especially for clients backing up multi media content on Angani cloud. Providers that allow free local infrastructure connections have been attracting more customers, which means more players will soon open up free local loop connectivity.

My experience and my participation at AFPIF

My name is Nico Tshintu Bakajika. I was engaged in the field of the Internet in 2006 while working as permanant secretary for Internet Service Provider Association- Democratic Republic of Congo (ISPA-DRC). This commitment was motivated by the desire to see the Internet contribute to the development of the Democratic Republic of Congo.

The mission entrusted to me had two components. The first was to do everything so that Internet operators speak the same language for the development of the Internet industry. The second part consisted in making ISPA-DRC accepted by government institutions in the Democratic Republic as a partner in the development of the Internet industry.

This dual role was not easy due to the fact that I should work with companies that are present in the same market with the same products. This forced me to learn a lot in order to:

– Acquire knowledge about the Internet industry and its evolution;
– Allow me to ensure good communication;
– Be a true leader, a good negotiator, a good initiator and a project manager;
– Apply transparency.

The fulfillment of this mission had two phases. The first phase covers the period from taking office until achieving some results listed below and this lasted for 6 years. When I started out, ISPA-DRC had three members representing at that time 42% of suppliers of Internet services providers in the Congolese market. Under the guidance of ISPA-DRC Management Committee, I contributed to the achievement of the following results:

Internally.

– Over 80% of providers operating in DRC are members of ISPA-DRC, 8 suppliers;
– The leaders of the member companies communicate each other and Internet service providers do self-regulation;
– ISPA-DRC has adopted regulations allowing it to adapt to the local environment and to technological changes;
– ISPA-DRC has established a framework for dialogue between engineers operating in member companies and some other resource persons;
– The initiation of projects of common interest such as training and implementation of an Internet exchange point.

Externally.

– ISPA-DRC has been considered a partner in a certain measure by institutions such as the Ministry in charge of telecommunications and the regulator and by associating the work involving the elaboration of some regulations in the telecom sector in the DRC and the development of the action plan for the future of information and communication technologies. I have attended to the telecom conference organized in 2010 by the Ministry in charge of telecommunications. In the same year, I took part in the  preparation of  the National Action Plan for the development of ICT in the DRC. I was one of the delegates of ISPA-DRC to work on the re-delegation of the domain name .cd and negotiations with the government for the removal of certain taxes considered as one of the obstacles to the development of the Internet industry in DRC.

The second phase started in the second quarter of 2012 and is presently on-going. A new dynamic is underway, ISPA-DRC has 13 members and controlling more than 89% of the Internet market in the DRC. This is the phase of major projects and the confirmation of ISPA-DRC as a true partner in the development of the Internet in the country. In this phase, with the confidence and the means made available to me by the Management Committee of ISPA-DRC, my contribution was remarkable at the internal as well external level through the following actions:

– Participation in the development of the naming charter for managing the .cd domain name;
– Participation in the drafting of the memorandum requesting the deletion of the 5% tax initiated by the Congolese regulator on the price of bandwidth;
– Participation in the drafting of the memorandum demanding the removal of Internet excise duties in the DRC;
– Active participation in the development and realization of the KINIX project, raising funds for the start up, organizing the working visit to KIXP in order to build on the model of Kenya Internet exchange point and organizing training workshop on the BGP with the technical, financial and educational support of ISOC;
– Active participation in the organization of the training on IPv6 in Kinshasa with the technical and educational support of AFRINIC.

To advance the Internet industry while drawing on the experiences of others, the ISPA-DRC Management Committee has opted to involve some resource persons of ISPA-DRC to the activities of exchange and transfer of knowledge organized internationally. It is in this context that I participated for the first time in AFPIF in 2014 in Dakar. This participation was an extraordinary experience, it added quite a lot in my way of seeing the world of the Internet and how I behave as an actor in the development of the Internet industry in my country.

During my stay in Dakar for the AFPIF event I was able to:

– attend the high-level discussions  on the future of the Internet,
– attend the high-level discussions  on innovations in the field of the Internet;
– discuss about the future of peering in Africa and the world;
– attend the high-level discussions  on the future of the production and development of Internet content in Africa;
– meet and exchange with experts  in the field of interconnection, production and development of Internet content;
– be advised on the behavior to adopt in order to play a role in the development process of the Internet in my country;
– establish contact with potential partners like Google, Akamai, France-IX for the production, development and hosting of Internet content;
– establish contact with experts in management and development of domain names,
– bring my reflection on the future of content in Africa,
– boost dialogue with Google for hosting GGC in Kinshasa.

With the knowledge gained and the people contacted in AFPIF, I am more than equipped to make my contribution to the development of Internet in the Democratic Republic of Congo.

My current involvement in the work initiated by the Deputy Prime Minister and Minister of ICT PT on reform of telecom legislation was remarkable for the gathering of reference documentations obtained through certain contacts in AFPIF.

My wish is to see AFPIF continue to bring together experts for the development of the Internet on the African continent. My contribution to this forum would be to share with the African Peers my experience on Interconnection and development of African content.

Africa Records Major Increase in Domestic Bandwidth Production

By Rebecca Wanjiku

African Internet Exchange Points have recorded a big increase in domestic bandwidth production, as a result of growth in sharing of Google cache, e-government services, local hosting infrastructure and services.

According to Packet Clearing House (PCH), Africa’s domestic bandwidth production grew by 145 percent, from 113Gigabits in April last year to 277 Gigabits in April this year. The number of IXPs also grew from 25 last year to 37 this year, a 48 percent increase.

“There is a general observation of significant traffic increase at IXPs where members have mutually agreed to share Google Cache and other CDN cache traffic; there is also considerable traffic being generated from e-government services, growth of local hosting services supported by the availability of local hosting infrastructure,” said Michuki Mwangi,

One of the fastest growing IXPs in Sub-Sahara Africa is NAPAfrica. It has three locations in South Africa; Johannesburg  (Est. 2012), Cape Town (Est. 2012) and Durban (Est. 2014). NAP Africa Johannesburg records 20Gbps peak traffic, Cape Town has 5Gbps, while Durban has 100Mbps peaks traffic. Two NAPAfrica IXPs have recorded significant growth within a very short period. On the other hand, the INX operated by the South Africa ISP Association (ISPA) and also hosted in data centers in Johannesburg (JINX est. 1996), Cape Town (CINX est. 2009) and Durban (DINX est. 2012) have equally high traffic at JINX (14Gbps peak) and CINX (3.8Gbps peak) by the regions levels. However, it is of interest to observe that NAPAfrica’s two facilities have achieved higher traffic levels over a shorter time compared to the INX in similar locations.

“NAPAfrica is an IXP located in one of the few carrier neutral data center facilities in Africa operated by Teraco. As a result, NAPAfrica is in a prime location to attract membership from a diverse range of businesses collocated inside the carrier neutral facility. I believe that, the carrier neutral data-center factor has played a significant role in the impressive growth seen at NAPAfrica over a short period,” added Mwangi.

Considering that most carrier neutral DC’s are often served by major operators. It is likely that NAPAfrica’s growth is buoyed by its ability to easily connect and cross connect providers within the data centre and at high speed, without the need for procuring additional links with infrastructure operators.

According to preliminary data from research being conducted by Africa IXP Association (AF-IX), 35 percent of the IXPs charge port fees (monthly/annual), which are considered a global best practice to ensure sustainability of the IXP operation. This position is enforced by the fact that 35% of the IXPs that do not charge are planning to implement fees in the future. If this would be considered it would be safe to say that soon, at least 70% of all the IXPs in region would be self-sustainable and capable of establishing themselves as regional hubs.

The survey also highlighted that majority of the IXPs (55%) have small networks and content friendly peering policies. These policies appear to be in line with the current level of development where most of the IXP members are small networks and are looking to attract content providers. Fifteen percent of the IXPs have bilateral peering which is friendly to large networks that prefer to have the choice of whom they interconnect with at the IXP. The remaining 30% of the respondent IXPs have less favorable peering policies that enforce peering for all IXP participants. The mandatory peering policies are often favored by startup IXPs to develop the peering culture. These policies tend to be reviewed as the IXP grows and members have a better understanding of the benefits of peering.

AF-IX was formed in 2013, to provide an enabling environment for IXP operators and to help IXPs maximize their value, to improve connectivity within the region and increase Internet’s value for all.

Even though the research shows growth and stability in terms of power back up and security in the IXP facilities, many IXPs are struggling to grow the number of networks peering and the capacity exchanged locally.

“To grow their capacity, the African IXP operators need to consider expanding the target market of the IXP membership to include a diverse range of non-traditional members such as banks, government networks, media, academia, research and education networks,” said Mwangi.

With the massive investments in ICT infrastructure, Mwangi says IXPs need to develop strategic partnerships with terrestrial infrastructure and submarine cable operators to provide suitable packages for connectivity to the IXP facility. This will serve as an incentive to connect the new diverse range of (local and cross-border) businesses to the IXP and with higher capacity.

Mwangi concluded that the notable growth in traffic exchanged at IXPs is a clear indication of the regions potential and future growth potential is dependent on the stakeholders ability to nature and leverage on the relationships formed within the IXP’s ecosystem.