Governing the Commons, Ostrom (1990) and Incentives Build Robustness in BitTorrent, Bram Cohen (2003)
Is Bittorrent solving a tragedy of the commons problems?
Tragedy of the commons is a concept that was popularized by Hardin in 1968. It has been observed for thousands of years before that, e.g. by Aristotle in his Politics. The idea is that a collection of individuals acting in their best interest results in a suboptimal outcome for the collective, or the commons. As a prototypical example: herding animals. Each herder wants to maximize profit, so they increase the amount of animals they have. This leads to to too many animals in a scarce field. A better solution would be to divide the area up somehow so that each farmer has enough animals to not overgraze the area. That this doesn't happen is precisely the tragedy of the common. It applies whenever a scrace resource is used in common by many individuals.
A bottom up solution to this is Common Pool Resources (CPR), that Ostrom studied in her Governing the commons. More on this later.
In Bittorrent, a file isn't scarce on its own. But in terms of downloading it is. Without Bittorrent, a file is usually uploaded to a single server, and then each additionally downloader causes things to slow down. Collectively, people would be better of if a few people downloaded the file first, then they uploaded it to the rest after, and so on. This would be the most effective way of solving the scarce bandwidth issue of distributing said resource as quickly as possible.
However, that is difficult in practice. Individuals have very little incentive to upload files to other people, especially strangers. Bittorrent largely solves this problem through game theory (tit-for-tat) and sharing of rare chunks. Each downloader automatically becomes an uploader, and if they don't provide to the system they are punished. It is thus a form of self-regulating, trust-minimized solution to a tragedy of commons problem.
Ostrom studies CPR problems and how they are solved through collective rules. Specifically, she looks at small-scale CPRs (50-15k people) where the individuals are heavily dependent on the resource for economic return. That doesn't appear to be the case for Bittorrent at all.
Ostrom defines a resource system and resource units as being the things extracted from said system. E.g. fishing grounds as resource system with fish as the resurce units. You have stock for resources and flow for harvest. This is interesting in its parallels with Bitcoin and stock to flow in such a system. In that case, scarcity is an asset! See stock-to-flow ratio, and so on.
Ostrom further studies these CPR institutions, all of which have existed for 100 years and some of which have existed for 1000 years.
Does it make sense to look at a file as a resource system, where download bandwidth is the resource unit? A provider provides this unit by uploading a specific file. Since we are talking about information, the file itself is "renewable" or infinite. But, over a given time horizon only so much of the file can be uploaded at a given time. This is what makes it scarce. One thing that is interesting about Bittorrent is that it doesn't rely on long-term games of trust. Nodes can come and go. While the benefit is not economic, there is a benefit. Additionally, since files are chunked, the rarest piece is uploaded first, as it is the most valuable (least likely to be replicated elsewhere).
A big issue is still in terms of the game being limited to a single file. Once a file is downloaded, there is no reason to stick around. Hence the use of private trackers which track a global seed ratio, albeit centrally. This strikes me as a form of Levithan approach, top down and heavy (censorship and control, etc). In that sense, it is is similar to the state.
Example of CPR: Swiss alps. For 500 year, they have an association to deal with common property. Rule simply "no citizen could send more cows to the alp than he could feed during the winter". Then enforcement and some form of distribution.
Design principles for long enduring CPR institutions. Clear boundaries/membership, collective choice arrangement, monitoring, conflict-resolution mechanism, etc.
It'd be interesting to use the design principles outlined and analyze various P2P networks in more detail. Bittorrent, Bitcoin, Ethereum etc. E.g. forking could be seen as a collective choice arena.
Another aspect: tragedy of the commons are about over utilization. What, if any, is the correspondence for under utilization?
- Free rider problem
- Stock to flow ratio