Outcome of the SESERV workshop on the interplay of economics and technology

posted 14 Feb 2012, 03:55 by Patrick Poullie   [ updated 6 Jun 2012, 13:45 by Costas Kalogiros ]

The main goals of the Future Internet SESERV workshop were to present and investigate economic issues related to the adoption and usage of Future Internet technologies, the targeted formation of breakout working sessions to discuss and elaborate the potential advantages and drawbacks of following the "tussle-awareness" paradigm, and to comprehensively debate topics such as "one Internet vs. multiple internets", the role of regulators, the role of economics, etc.

Focus Group Results

B. Briscoe: The “byte” is not the right charging metric – A flow should be charged on its contribution to congestion

The first keynote was given by Bob Briscoe and addressed the suffering of Internet technologies from “Market Failures”. He proposed a new metric to measure a user’s contribution to congestion, as, according to him, the total number of bytes sent by her is not representative for this purpose. He suggested weighing bytes by congestion level (which implies the marking of packets) in order to measure the congestion caused by a user.

F. von Bornstaedt: “Sending Party Network Pays” is the only way to do E2E QoS - increases accountability and trust among providers – no problem with net neutrality

Falk von Bornstaedt gave the second keynote, which was about the evolution of business models in the Internet. Supported by illustrative examples he argued that over provisioning in the Internet just as in public roads might not always solve the problem. He also drew the attention to a fundamental operator problem that is constituted by users doubling their bandwidth demand every two years, but who are not willing to pay more. To address this issue, he proposed a traffic-financing model where receivers of traffic pay.

A. Cooper: Competition among ISPs in the UK does not regulate the market: almost all ISPs employ traffic discrimination, so no consumer choice

In the framework of the third keynote, Alissa Cooper addressed the question to which extent network operators should be free to manage certain Internet applications differently from others and if operators are prevented from immoderate traffic shaping by competition. She argued that traffic management strategies can be influenced by market structure, but competition does not necessarily safeguard discriminatory traffic management. Also, she stressed that designers of future traffic management technologies should recognize primacy of incentive structures, which they want to deploy.

Tussle Analysis Sessions

After the keynotes Costas Courcoubetis and Costas Kalogiros introduced the tussle analysis methodology that is being developed by SESERV. Important concepts, such as stakeholders, tussles and spillover were explained and afterwards drawn to concrete examples (including misuse of the DNS and TCP protocol). Three parallel focus groups on understanding and dealing with tussles that may arise due to the introduction of new Internet technologies were to follow. These focus groups included presentations of representatives of different FP7 projects just as stakeholder role-plays among participants to comprehensively identify tussles that the respective projects may have to face.

R.Mason: Net neutrality should be evaluated – more analysis based on models and less ideology is needed

Robin Mason introduced two-sided markets to present a new approach to model and investigate network neutrality issues, in the last keynote. He presented and evaluated the proposed model with respect to efficiency as well as welfare and concluded, although applicable, the model needs to see further improvements due to left-out assumptions.

The role of economics in the evolution of the Internet

Next, a discussion panel on the role of economics to the Internet evolution offered interesting insights as well as thought-provoking impulses. A problem identified was that even if all raw data necessary to answer socio-economic questions regarding the Future Internet, could be collected in todays Internet, the problem would arise to extract the respective information from it, and, even if this would be possible, the interpretation of this information would constitute a conceptual problem. Further the panel addressed the interesting question, wether traffic should be charged on its purpose/application, i.e., should it be acceptable to charge differently for the same amount if traffic if it transports for example video streaming data or VoIP data.

The Workshop was closed by the rich feedback to the break-out sessions, especially valuable to the projects that presented during the sessions. Main problems identified included (i) incentive mechanisms for users to actively contribute to a networks value, (ii) operators loosing control of their networks as a trade-off for user satisfaction, (iii) decreasing revenues for interconnection providers if content centric networks are deployed, and (iv) the lack of SLA monitoring tools that allow users to ensure that the service they pay for is actually provided.

The workshop has produced valuable input for SESERV as well as for its participants. Many socio-economic issues regarding the Future Internet were identified or seen from a different angle, due to the viable discussions of experts from different domains all sharing an interest in the Future Internet. In particular, representatives of several FP7 projects told SESERV, that they received thought-provoking impulses with respect to socio-economic interests that need to be considered when further developing their technology.