Conflict of conventions ? What a social sciences view can reveal about the interconnection deal between Netflix and Comcast

27 March 2014

The news spread fast: Netflix and Comcast have sealed an interconnection agreement. The online video service will pay the internet access provider for the large amounts of video data1. to be delivered into the Comcast network, without detours. At the ends of Comcast’s network, internet users can hope for a better Netflix experience. A lot has been written about this deal despite the fact that the companies have been scarce on information. The press release barely extends over half a page.

What is going on here? For a long time, hardly anyone took notice in interconnection agreements, apart from professionals such as network engineers or exchange point operators. But recently, network-heavy companies, consumer activists and long-time internet witnesses get in motion about it. The public uproar caused by the deal between Netflix and Comcast might indicate an uncertainty behind the scenes. At stake may be a shared interpretation on who has got to pay whom for what when networks interconnect?

Conventions mitigate uncertainty

The economics of convention can help to interpret this uncertainty. French theorists, including Thévenot, Orléan and Boltanski, have introduced and elaborated this concept since the mid 1980s to explain social – and therefore also economic – behaviour. The economics of convention start from the idea that economic action is not only based upon utility maximisation, but that we also rely on conventions. Conventions can be understood as shared assumptions and practices that have consolidated customs. In interaction, conventions reduce uncertainty. According to the concept, situations appear as natural when all parties share the same frame of reference that has been established by the convention. The fact that the frame is produced socially slides to the background. (For a good summary of the concept see Jagd 2007.)

The development of the internet infrastructure probably is shaped by conventions as well. However, they may not have been reflected as such yet. It has rarely been questioned how private and public actors produce the internet by interconnecting network resources. This silence around interconnection arrangements could indicate that the actors who were originally involved shared the same frame of reference. The public attention for a singular case such as the Netflix/Comcast deal, however, likely stands for new interpretations emerging. They reinforce uncertainty and cause justifications. It would not come as a surprise if alternative views about interconnection modes such as peering or transit were starting to arise. The trend towards liberalisation in the telecommunications sector, infrastructure innovations such as content delivery networks and – in the Netflix case –, an increasingly sensitized user base keep the sector in motion.

If two forms of coordination conflict with each other, their implicit logics start surfacing. In order to overcome the conflict, the parties begin to justify their actions and thus disclose the foundations of the conventions they employ. They explain which context they hold true for their action and claim that this situational framework be shared by the other parties. As we justify our actions, we articulate the standards and interpretations that guide us in a situation. (2007, pp . 80-83 )

How do network actors justify their interconnection practices?

What are the old and new conventions that might conflict with each other in interconnection agreements? One would have to listen to the actors involved: internet access providers, content and distribution companies, internet exchange point operators and carriers. However, they traditionally remain silent about the details of their deals.

Nevertheless, we can test the idea in the network actor’s environment. Self-acclaimed “voice for the streaming and online video industry” Dan Rayburn recently published two ambitious blog posts (Rayburn 2014a, 2014b) about the topic. They unleashed a debate among professionals and an interested public. In the first article, Rayburn accuses mainstream media outlets to “get it all wrong” when they denounce either Comcast or Netflix of illegitimate behaviour. In the second article he provides some facts about the deal. Throughout the articles and their hundreds of comments, a number of possible frames of reference show that provide rationalities for the network actor’s behaviour. Some of which I will outline here. This is not about to judging the evaluations. It is rather, to exemplify that different rationales of justification are possible in economic action and that the network actors already are surrounded by conflicting frames of reference.

So how do the participants in this discussion justify or criticize the agreement between Netflix and Comcast ? An excerpt:

  • Independent economic action: Interconnection arrangements among infrastructure operators purely are a business matter. “Netflix ‘s streaming quality is based on business decisions, that’s it.” (2014a) They occur in an isolated economic field of action. “Users have no influence.” (duddits-fairuse 2014) Companies act out of self-interest. “ISPs are not charities.” (Gregory 2014) Transparency has no place in this setting, because contracts are not disclosed here, as they are not in other industries.
  • Natural synthesis of internet architecture and market: Interconnection arrangements take place in on the basis of a natural synthesis of commercial and technical logic of action. “It’s how the Internet works.” Commercial and creative peering has happened since the early days of the internet. It is a constitutive element. (Rayburn 2014b)
  • Techno-orthodox: Younger network actors such as Netflix lack legitimacy in how they use the internet. They shall recognize, correctly interpret and respect the technical standards: “The internet was not designed to be television. IPtv is a commercial product to do the work, not for the carriers to get stiffed on the bill.” (Crusader 2014) The Internet appears in this sense as an object with fixed properties. Its autonomy is to be protected against misuse – in this case allegedly by television.
    Techno-orthodox arguments are also brought forward to demarcate whose arguments even may be taken seriously – namely: those of technicians. “If you knew how the Internet worked, you would not be making the points that you are.” (Hammett 2014) A lack of technical punditry may seriously harm the internet: “OMFG, it’s obvious from many of the posts that a little bit of network knowledge is dangerous.” (Joe6Pack 2014)
  • Ethical reflection: Economic actors shall consider the social context of their activity as a reference. As some commenters see it, network actors currently do not adequately meet this expectation. Internet access providers confront content providers with traffic-balance requirements that are “morally indefensible” and “disingenuous” (Enger 2014). The allegation is: “extortion” (texrat 2014). “This is a techno rant,” (O’Neill 2014) complains one commenter  who misses consumer protection aspects and a reflection within the discourse.

In case of conflict: call upon authorities

Several commentators are disappointed that their frames of reference seemingly are not being internalised by the network actors. And they offer a glimpse at what unsolved interpretational conflicts may lead to: Those who fear to succumb with their frame of reference will try to form interpretional bonds with potentially powerful actors, namely: call for external supervision through regulators. “It is important to relate this action to the discussion (…) over new regulatory and interconnect structures”, argues one commentator. “The public needs to stay on top of this,” (Cole 2014) demands another.

In summary, just based on this third-party discussion, several possible frames of reference become appearent which network actors might rely upon to more or less acceptibly justify their interconnection practices: “rational business decisions with maximization logic”, “taking commercial peering as an evolutionary base of the internet” , “techno-orthodoxy” and “ethical reflection”.

What ‘s the object of interconnection deals?

As if harmonising the conflicting frames of reference were not difficult enough, the discussion below Rayburn’s blog posts indicate yet another possible source of uncertainty: the trade object itself.

“What exactly are they [the ISPs ] selling to customers which the[y] call an ‘internet connection’ anyway?” (noname 2014), wonders one commenter. And author Rayburn himself emphasizes in the second article, that in the case of Netflix the object of trade has been refined from Comcast through various quality guarantees. (Rayburn 2014a)

Economic activity does not only work better if the parties share the same frame of reference, it requires that they agree on the object of trade and its properties. It is a collective interpretive effort that facilitates business. (Thévenot 2002 , pp . 189-193)

But what is the good that the internet infrastructure produces? How can the object of trade between internet access providers, content-driven enterprises and content delivery networks be adequately described? “Traffic” or “connectivity” seem to be too general of an answer considering network dynamics. Data flows relate to money flows – but this relationship is hard to qualify.

Obviously, companies succeed in specifiying the object of the trade well enough so that business can happen. The agreement between Netflix and Comcast proves this. However, it would not be surprising if defining the object of trade may become increasingly difficult, the more different players emerge on the marketplace.

I believe it would be worthwile to pursue the question of the object a bit further. If the uncertainty about the object actually does not only prevail among end-users but also among economic network actors, this would imply a rather high need for coordination in interconnection relationships.

A lack of clarity about the object of trade would also have implications for research on interconnection arrangements: Any research that does not take into account the preliminary “making of the object” would appear incomplete.

Summary and Outlook

The interconnection agreement between Netflix and Comcast has offered an opportunity to reflect about the rationales that may lead network actors in their peering policies. By approaching the uncertainties with the concept of economics of convention it can be shown that the companies who are involved in interconnection agreements have a plurality of frames of reference available that they can base their economic decisions upon – more or less successfully. Since the agreements are not accessible and the network actors themselves hardly speak out in public, it is difficult to grasp the actual justifications and rationales empirically. Taking a detour via an industry professional’s articles allowed to test the idea of the conventions in this field. The vivid discussion in the comment section already indicates that there is – at least in face of the public – no common understanding of an acceptable justification, quite independently of any jurisprudence. The frames of reference put forward range from (1) supporting economic decisions that are based upon maximising self-interest over (2) assuming a natural synthesis between technological and economic logic as the motor of internet development and (3) a techno-orthodox interpretation of norms to (4) calls for ethical reflection by the actors. The frameworks allured to should not be confused with the actual frames of reference that may be acted upon in negotiations on interconnection arrangements. However, they show that there is a potential plurality, a source of uncertainty. This plurality ought to be investigated further in order to better understand why the internet evolves how it does.

1. According to Netflix more than 40 Million customers view over one billion hours of TV shows and movies streaming from Netflix a month –
2. Image: Mangee – Sketch of a peering constellation, license: CC BY-NC-SA 2.0



  • Jagd, S. (2007). Economics of convention and new economic sociology mutual inspiration and dialogue. Current Sociology, 55(1), 75-91. Retrieved from Google Scholar.
  • Rayburn, D. (2014, Februar 23). Here’s How The Comcast & Netflix Deal Is Structured, With Data & Numbers [Web log post]. Retrieved from
  • Rayburn, D. (2014, Februar 23). Inside The Netflix/Comcast Deal and What The Media Is Getting Very Wrong [Web log post]. Retrieved from
  • Thévenot, L. (2002). Conventions of co-ordination and the framing of uncertainty. In E. Fullbrook (Ed.), Intersubjectivity in economics. London and new york: Routledge (pp. 181-197). London: Routledge.


  • Cole. (2014, February 24). Re: Inside The Netflix/Comcast Deal and What The Media Is Getting Very Wrong [Web log post]. Retrieved from
  • Crusader. (2014, March 2). Re: Here’s How The Comcast & Netflix Deal Is Structured, With Data & Numbers [Web log post]. Retrieved from
  • duddits-fairuse. (2014, February 24). Re: Inside The Netflix/Comcast Deal and What The Media Is Getting Very Wrong [Web log post]. Retrieved from
  • Enger. (2014, February 24). Re: Inside The Netflix/Comcast Deal and What The Media Is Getting Very Wrong [Web log post]. Retrieved from
  • Gregory. (2014, February 26). Re: Inside The Netflix/Comcast Deal and What The Media Is Getting Very Wrong [Web log post]. Retrieved from
  • Hammett. (2014, February 25). Re: Inside The Netflix/Comcast Deal and What The Media Is Getting Very Wrong [Web log post]. Retrieved from
  • Joe6Pack. (2014, February 26). Re: Inside The Netflix/Comcast Deal and What The Media Is Getting Very Wrong [Web log post]. Retrieved from
  • noname. (2014, February 24). Re: Inside The Netflix/Comcast Deal and What The Media Is Getting Very Wrong [Web log post]. Retrieved from
  • O’Neill. (2014, February 24). Re: Inside The Netflix/Comcast Deal and What The Media Is Getting Very Wrong [Web log post]. Retrieved from
  • texrat. (2014, February 24). Re: Inside The Netflix/Comcast Deal and What The Media Is Getting Very Wrong [Web log post]. Retrieved from

This post is part of a weekly series of articles by doctoral canditates of the Alexander von Humboldt Institute for Internet and Society. It does not necessarily represent the view of the Institute itself. For more information about the topics of these articles and asssociated research projects, please contact

This post represents the view of the author and does not necessarily represent the view of the institute itself. For more information about the topics of these articles and associated research projects, please contact

Related articles


Scaling research and impacting change

How do the world’s most powerful internet, mobile, and telecommunications companies treat their users’ freedom of expression and privacy? On 2 May 2018, the Ranking Digital Rights 2018 Corporate Accountability...

Trust in me, just in me! Trusted Third Parties in data protection law

High time for a reform of European data protection legislation: The last corresponding directive dates from 1995 and is no longer up to the technical developments of Big Data, Smart...

Understanding Artificial Intelligence

Artificial intelligence is a controversial and much-discussed topic of our digital future. Aljoscha Burchardt sees a huge gap between the machines of today, which are able to act intelligently, and the...

1 Comment

  1. academia do importador e confiavel on 18 July 2015 at 6:07 am

    Thanks a bunmch for sharing this with all people you really recognise what you’re talking about!
    Bookmarked. Kindly also cohsult with my web site =).
    We could have a link change arrangement between us

Leave a Comment