Indian telecom regulator’s idea of free data is centred on an aggregator model. But for critics it is a new form of zero rating. In such a scenario can the model of equal rating for open and accessible internet be a distant dream?
Mozilla is working on two separate goals in net neutrality. One is to bring everyone online; ensure that everyone has access to the internet. The other is to ensure that the network should remain open and diverse. We want people to have access to the whole diversity of the internet and not just in some parts. The debate on neutrality has been playing in India for two years in a more developed way than anywhere else in the world.
What we see in the ‘zero rating’ is anti-competition and anti-consumer concerns that harm neutrality. I believe, we need more action, both in terms of solution and regulation. But the promise and premise of equal rating [where consumers choose content based on the quality of that content, not the financial power and business partnerships of the provider] has always been that we can and we must do better. So the equal rating comes from the derived rating principle that traffic should be treated equal, irrespective of the sender, receiver, content, device, application and website. There shouldn’t be differential treatment or pricing.
We have been doing this through three ways: policy, research and innovation.
For policy we have been in talks with the Telecom Regulatory Authority of India (TRAI) on what should be the regulatory framework which would guide telecom service providers (TSPs).
We have wrapped up a study in India, wherein we particularly approached the demand side, met the users, and asked them about their expectations, motivations and behaviour. We asked them how they use internet differently when data is subsidised and why they use subsidised services. We commissioned a research in Sub-Saharan Africa (Kenya, Nigeria, Rwanda and South Africa), South Asia (India and Myanmar) and Latin America (Peru). What we saw in other markets is that people are not coming online to use subsidised services but for rather sophisticated data management.
Look at some of the statistics put out by Facebook: 50 percent of Free Basics users buy a paid plan within 30 days. But what does this mean? Fifty percent of SIM cards have some amount of credit added to them. We knew nothing about their connectivity situation before. Users in India and in many other developing markets have two, three or four SIM cards. They have some very sophisticated strategy to use the best SIM for unique purposes: one for calling, another for data, yet another for bundled services.
These are predominantly prepaid markets. So some amount of credit is added to them in 30 days. We don’t know if that is a meaningful amount. We don’t know if that amount went in voice or data.
The further implication of this for users who can’t afford more data is that they will be forever restricted to limited internet services.
Facebook is one the most popular services in the world, so it’s not surprising that people who are familiar with it will find a SIM card which will help in getting that service for free. We are trying to ascertain the actual user behaviour, rather than making assumptions on what users want and why and how they use services.
Equal rating solutions
There is no one company that has the silver bullet, which can solve the connectivity challenge. We need diverse and new solutions. A few months ago, we launched the equal rating innovation challenge to catalyse new thinking and innovation and provide affordable access and digital literacy. We received 98 submissions from 27 countries around the world. The one from IIT Bombay, focusing on White Space technology, won the competition.
I was impressed with the tremendous quality of the solutions available, which means that there are a number of ways to bridge the digital divide. This shows that the popular notion of zero rating as the only solution to provide internet is not true.
Moreover, the equal rating challenge was not just about selecting the finalists, but the community of solution providers. We want to build and nurture that community of people.
One of the common themes in the challenge was community networks. Of the two African semi-finalists, one is AFRI-FI. They are working with municipalities in South Africa to provide free Wi-Fi and make the initiative viable. They offer 500Mb data per day free to users. That’s tremendous amount of data. I live in the US and I don’t have that kind of data!
So they [AFRI-FI] are looking at how to use advertising to meet some of the cost. Their cost is already low as they are working with municipalities. They serve 2.6 million users in South Africa in a couple of different municipalities. They have Wi-Fi routers throughout the municipality. Now, they are looking at the sustainability of the model. Can they have a model where people pick up the litter, something which municipality wants, and exchange it for data credits?
There is another interesting municipal effort, called Smart Kigali, from Rwanda, though it is not a semi-finalist. They have partnered with telecom service providers for 4G connectivity on public buses in Kigali, the capital of Rwanda. Now as part of your ticket, you get free 4G connectivity in Kigali. It is a municipal effort.
An interesting way of looking at it is if you don’t want to pay money then is there something which you can give up to create some flexibility in the system that allows it to be sustainable? You can use it on the bus, or may pick up trash. Or you can watch an advertisement.
In Bangladesh, we partnered with Grameen Phone and Telenor by which users can watch an ad for 30 seconds on the market place and subsequently get 20 Mb data for free. In Sub-Saharan Africa, they are still doing it.
Another model we pioneered in 25 Middle Eastern Sub-Saharan African markets. This was a $40 Firefox smartphone device that comes bundled with unlimited voice, SMS, 500 Mb data every month for three to six months, depending on the market.
The argument is to give people a taste of the full internet rather than giving access to some select sites and services.
TRAI’s free data proposal
There are two models. One was building on to the Digital India initiative to connect rural communities. It’s a scheme that envisions offering 100 Mb free data for a limited period. This is a great idea. This is what we have been saying for equal rating.
From a user’s perspective, I think one of the major concerns is providing the Aadhaar number. We need to consider the privacy implications of it, though we understand the government’s legitimate need to prevent double-dipping. India currently lacks a privacy protection law. We have said multiple times to the government and TRAI that privacy legislation needs to be a policy priority. From the privacy point of view, that merits attention. From the neutrality perspective, that’s a pretty great scheme and good use of the universal service obligation fund.
The second scheme considered under free data order was around the third-party intermediaries like Gigato and I think it’s really a question of implementation. You know the differential pricing recommendation given by TRAI [which disallows zero rating]. If this is used to circumvent the differential pricing then obviously it would present some significant concerns.
If one bit costs differently than the other bit then definitely that would also be concerning. It’s essential to know what that scheme would look like and so we are waiting to see what the implementation looks like before passing any judgement.
A common definition
Saying that there is no single definition of neutrality (as pointed out in the TRAI paper on net neutrality) is a misnomer, although the wording changes from jurisdiction to jurisdiction.
We talk of best efforts principles. This is the idea that TSPs have to make their best efforts to deliver the packets. Second is end-to-end principle, which says that every user should be able to reach all points of network. Third is innovation without permission. This is the core protection for competition and innovation in a sense that the TSPs should not act as gatekeeper to allow new services and new content. A user should be able to reach anywhere on the internet and not just what a telecom services provider permits.
If you look at the neutrality legislation around the world you have various definitions but in the end everyone has a very strong non-discrimination principle. Actually it’s rare that the word ‘net neutrality’ show up in statutes. Net neutrality as a phrase rarely shows up in regulation. Normally, what you see is non-discrimination, non-interference or non-restriction.
To the question of Indian context and should net neutrality be treated differently here, I don’t think it should really matter. If you compare India with Latin American countries, though there are differences, there are many similarities too. Like more of wireless connectivity than wired, road connectivity problems and similar economic status. But more than 80 percent of Latin America is already covered by net neutrality protection.
The last mile
Many Latin American countries have national broadband plans. For the last mile, we are seeing a lot of interest in community networks. We have a Brazilian solution (Free Networks P2P Cooperative), as part of our top five semi-finalists of the innovation challenge, which is focused on training people to build and manage their own network instead of going village to village to help people to build that infrastructure.
There are different ways to connect the middle mile and backbone connections with the last mile. Generally, once you are in the local village, we are talking about Wi-Fi or radio. You do have wireless internet service providers as well. I think there is also lot of traction in the upstream, between the internet cloud of the TSP and the national gateway.
A participant proposed to lower down the internet cost by laying down greater optical fibre cable. This participant was in-charge of Google’s optical fibre infrastructure. He basically created a non-profit to build national optical fibre capability. Telcos have very little incentive to build the backbone with optical fibre cable, which in turn dramatically affects the wholesale price of data at the top.
According to the participant, if you look at the wholesale cost of the data in Pakistan it’s about $300 per Mb. In London, it’s 30 cents! [Here per megabyte price is not what a user pays. It is the cost of one Mb data in the pipe, which caters to several users.]
If you can reduce the upstream cost – where fibre enters the country at the gateway level – then it trickles down to the end user level. Several governments are thinking about how they can get more fibre cable into the country and reduce the upstream cost.
As countries get closer to the internet backbone the cost will go down.
The backbone exists everywhere around the world. We used to have a very hierarchical model. In the 1990s, we had backbone providers, regional and local ISPs and the traffic had to flow to local to regional to backbone. That’s not how we are today. We have a flatter network structure, which is a good thing, as it helps data to move more efficiently and cheaply around the internet. But any given provider still has to transit through other people’s network.
So there how connected you are to other nodes and parts of the network influences your speed, price, efficiency and that’s why international cables are important.
If you can build fibre, undersea or terrestrial, and then allow communities to buy access to cable, internet will be cheaper. It may be way expensive for communities to run cable by themselves but if someone had to run the cable, invest in it and sell it to them on a community basis that can dramatically reduce the price of fibre capacity of those communities. It would require huge investment for somebody to do the intermediary job.
If you go by the ITU data, 90 percent of the world is connected by 3G network. The 2G coverage is even greater. It covers most of the world. One of our semi-finalists had created a mesh-network with the help of a local service provider. It is based on a new business model for telecommunication companies to provide free 2G to enable all the benefits of the open web to all.
Why did mobile growth explode in Sub-Saharan Africa? It’s also probably true in the Indian context. It’s because the users got free incoming SMS and calling. So that did two things. One it immediately increased the network effects of these devices and services. Two, it immediately helped people in seeing the value in paying for these services. When you get free SMS, you want to write back. When you get call you want to call back.
So what this model is trying to achieve is to give everyone free 2G connectivity bundled with phone. And again that increased the network effect and immediately helps you in seeing the value in paying for the internet.
This is just a concept and no telco is offering it yet.
As told to Pratap Vikram Singh
(The article appears in the April 16-30, 2017 issue of Governance Now)