Joint Statement on Net-Neutrality and the TRAI Consultation Paper

The content on this website is strictly the property of Insight and the Students’ Gymkhana IIT Bombay. If you wish to reproduce any content herein, please contact us:
Chief Editors: Mihir Kulkarni, Niranjan Thakurdesai
Mail to: insight@iitb.ac.in

The TRAI has recently come out with a consultation paper, which has been the cause for much debate around net-neutrality, and for good reason. Network neutrality (or net-neutrality for short) is an operational principle of the Internet, where Internet Service Providers (ISPs) or telecom operators, i.e., anyone who manages any part of the Internet, is neutral to (a) the origin and destination of any traffic, and (b) the content of the traffic.

The consultation paper includes several arguments for why network neutrality must be compromised or weakened. This joint statement counters those arguments, focusing on four technical aspects of digital networks, and urges the TRAI to strongly support net neutrality.

Net neutrality and congestion

The first and foremost argument put forth by the consultation paper is that net neutrality must be violated to solve congestion in the network: “10% of mobile users actually consume 90% of operators’ bandwidth” (5.2, similar statements also in 5.17, 5.26, 5.28). This argument is fundamentally flawed technically, for congestion can effectively be addressed by looking only at the quantity of data, while preserving net neutrality. For instance, it would be well within the principles of net neutrality to serve the first GB of a user’s data fast, and the second GB of data slower. The second GB can also be priced higher. Such mechanisms are all too common in the physical world of mails. If the post office gets 90% postal mail from 10% of its users, it can either (a) impose limits of quantity of mail, or (b) price differentially based on amount of mail sent, or (c) simply increase its capacity. What it cannot do is to pry open every letter and price differently based on the content. To do so would be absurd. Likewise, if a road network is facing congestion, it would be absurd to charge road tax based on the identity of who is using the road, or based on whether the commuter is going to a bank or to a grocery store next to the bank.

Net neutrality and service differentiation

The second major argument put forth by the consultation paper, in different words (e.g. in 5.18, 5.21, 5.22), is that net neutrality must be violated to provide service differentiation, which is necessary for example in telemedicine applications or for specific business customers. This argument is also technically flawed. Network neutrality does not mean there is no service differentiation. It means that the choice for better service is made by the end customer, not by the network. For instance, if a remote clinic wants a certain network capacity to a city hospital’s telemedicine portal, it chooses to buy that extra capacity by paying the network operator. On the other hand, if net neutrality is violated, and the network operator made the choice, the clinic could end up with the same network capacity, not to the required hospital website, but to an e-commerce website of no use to the intended telemedicine.

Intelligence at the edge

Net neutrality is at the heart of the Internet architecture. The Internet architecture is essentially one where intelligence is pushed to the edge of the network: these are the content servers as well as the clients of these services at the other edge of the network (the smart-phones and laptops and PCs). This idea is fundamental to Internet’s success. This is the reason why any small developer or business can develop a website or smart-phone app and have it used by its customers. This contrasts the traditional telephone architecture where the end device is dumb and the network is intelligent (relatively speaking): you can do nothing but punch numbers or receive phone calls using the end device. The relative success of each of these models and the fact that even telephone companies have adopted the Internet model speaks for itself as to which architecture is better for innovation, for the economy, and for society. So when telecom network operators seek more control of how much traffic is used to which website or application (i.e., violate net neutrality by having “intelligence” in the network), the fundamental idea behind the Internet is under threat. The implications of this are huge, given that the Internet is central to today’s information age. (To its credit, the TRAI document acknowledges this in 5.6, 5.13, but not with sufficient weightage).

Misleading terminology

Finally, the consultation paper is misleading and confusing in title and terminology. The title as well the entire document uses the term “over-the-top” (OTT) services as though such services are some special services or applications. Now, any standard digital networks textbook, likely in its first chapter, would tell that any service or application using the Internet, runs “over-the-top” of the network. There is no example of an application which is not “over-the-top”. Thus the proposed regulatory framework encompasses every application and every use of the Internet, not just some special services. While the term OTT is common outside the consultation paper, it is misleading all the same. It is like a road construction company calling a school accessible by the road as an over-the-top service. Sure, one has to use the road to reach the school, but the term suggests that the road somehow has a role in the school’s education service over and above just using the road to reach the school.

In summary, there are no sound technical or economic reasons to violate net neutrality. In fact, such violation threatens the essential idea of the Internet itself. We urge the TRAI to maintain net neutrality in its strongest possible form, as this is crucial for a digitally empowered India.

Signed by:
Bhaskaran Raman, Professor, Department of CSE, IIT Bombay
Supratik Chakraborty, Professor, Department of CSE, IIT Bombay
Uday Khedker, Professor, Department of CSE, IIT Bombay
Pushpak Bhattacharyya, Professor, CSE Dept., IIT Bombay
Kameswari Chebrolu, Associate Professor, Department of CSE, IIT Bombay
Sriram Srinivasan, Adjunct Professor, Department of CSE, IIT Bombay
Om Damani, Associate Professor, Department of CSE, IIT Bombay
Parag Chaudhuri, Associate Professor, Department of CSE, IIT Bombay
Mythili Vutukuru, Assistant Professor, Department of CSE, IIT Bombay
Krithi Ramamritham, Professor, Department of CSE, IIT Bombay
Kavi Arya, Associate Professor, Department of CSE, IIT Bombay
Jayant Haritsa, Professor, Department of CSA, IISc Bangalore
Murali Krishna Ramanathan, Assistant Professor, Department of CSA, IISc Bangalore
Amitabha Bagchi, Associate Professor, Department of CS&E, IIT Delhi
Arpita Patra, Assistant Professor, Department of CSA, IISc
Gaurav Raina, Department of Electrical Engineering, IIT Madras
Niloy Ganguly, Professor, Department of Computer Science, IIT Kharagpur
Sandip Chakraborty, Assistant Professor, Department of Computer Science, IIT Kharagpur
Dheeraj Sanghi, Professor, Department of CSE, IIT Kanpur
Aaditeshwar Seth, Assistant Professor, Department of Computer Science, IIT Delhi
Animesh Mukherjee, Assistant Professor, Dept. of Computer Science and Engg., IIT Kharagpur
Somprakash Bandyopadhyay, Professor, Indian Institute of Management Calcutta
Kolin Paul, Associate Professor, Department of CSE, IIT Delhi
Ajit Kumar Chaturvedi, Professor, Department of EE, IIT Kanpur
S K Gupta, Professor, Dept of Computer Science, IIT Delhi
H.Narayanan, Professor, Dept. of Electrical Engineering, IIT Bombay
Sridhar Iyer, Professor, Department of CSE, IIT Bombay
Kumar Appaiah, Assistant Professor, Dept of EE, IIT Bombay
D Manjunath, Professor, Department of EE, IIT Bombay
Joydeep Chandra, Assistant Professor, Deptt. of CSE, IIT Patna
S. Akshay, Assistant Professor, Dept of CSE, IIT Bombay
Saurabh Lodha, Associate Professor, Department of EE, IIT Bombay
Mahesh Patil, Professor, Department of EE, IIT Bombay
Vijay Natarajan, Associate Professor, Department of CSA, IISc Bangalore
Souvik Mahapatra, Professor, Dept of EE, IIT Bombay
Vinayak Naik, Associate Professor, IIIT Delhi
Shivaram Kalyanakrishnan, Assistant Professor, Dept of CSE, IIT Bombay
K.V.S. Hari, Professor, Department of Electrical Communication Engineering, Indian Institute of Science, Bangalore
K. V. Raghavan, Assistant Professor, Dept. of Computer Science and Automation, Indian Institute of Science, Bangalore
Nikhil Karamchandani, Assistant Professor, Dept. of EE, IIT Bombay
Anil Kottantharayil, Professor, Department of Electrical Engineering, IIT Bombay
Varsha Apte, Associate Professor, Department of CSE, IIT Bombay
Debraj Chakraborty, Associate Professor, EE, IIT Bombay
Purushottam Kulkarni, Associate Professor, Dept. of CSE, IIT Bombay
Ashutosh Trivedi, Assistant Professor, Dept. of CSE, IIT Bombay
N. L. Sarda, Emeritus Fellow, Dept of CSE, IIT Bombay
Hema A Murthy, Professor, Dept. of CSE, IIT Madras
Rahul Garg, Professor, Dept. of CSE, IIT Delhi
Rogers Mathew, Asst. Professor, IIT Kharagpur
Virendra Singh, Associate Professor, EE Dept., IIT Bombay
Sundar Vishwanathan, Professor, Dept. of CSE, IIT Bombay
Uday Reddy B, Assistant Professor, Dept of CSA, IISc Bangalore