Live-blog of Governing the Commons
This post was originally published as a Mastodon thread from https://hachyderm.io/@jyasskin/113267949307091083 to https://hachyderm.io/@jyasskin/113303493113946716. I have added some supporting links in this post.
Contents
Introduction
I’m reading Elinor Ostrom’s Governing the Commons to see if I can learn anything about how to ensure that the Web stays a viable public resource. The book says its focus is limited to “common-pool resources” in nature that affect less than 15,000 people, so I’m on shaky ground extrapolating to the Web, but I will remain optimistic!
Edit: Thank you to @luis_in_brief and @greggish for pointing out that https://knowledge-commons.net/ is the modern research on what I want to study. [Original]
One amusing passage: ”… a common set of problems … coping with free riding, solving commitment problems, arranging for the supply of new institutions (!), and monitoring individual compliance” (pg 27)
One of these things is not like the others … Or more probably indicates that we don’t share an understanding of “institutions”. [Original]
Chapter 2, pg 35–39, talks about how shared cooperative norms can help individuals achieve better results, and how watching others cooperate can help individuals cooperate. Unfortunately, a lot of the actors on the Web are corporations that have been taken over by business school graduates who were taught to ignore norms in favor of a quick buck. [Original]
Pg 42 appears to endorse another theorist’s observation that “supplying a new set of rules is the equivalent of providing another public good … a second-order collective dilemma.”
But participating in defining rules isn’t a simple public good: the more you participate in defining rules, the more your interests are reflected in the content of those rules. I hope she’ll clarify the dilemma she sees. [Original]
This body of research should be read like all other bodies of research. First, it should be read cover-to-cover, multiple times. Then, it should be read backwards at least once. Then it should be read by picking random research papers and following all the cross-references.
With apologies to the HTML spec. [Original]
Chapter 3 starts to show a tension between commons management and modern ideals of free immigration, although it also says that high emigration was important to maintaining this commons. I wonder if that’ll turn out to be a theme, or if some commons also work with more open immigration. [Original]
“The Tribunal de las Aguas is a water court that has for centuries met on Thursday mornings outside the Apostles’ Door of the Cathedral of Valencia.” pg 71 [Original]
Design Principles
Pg 90 presents 8 design principles drawn from long-enduring (think 500 years) institutions that manage common-pool resources (CPRs). [Original]
- Clearly defined boundaries: Individuals or households who have rights to withdraw resource units from the CPR must be clearly defined, as must the boundaries of the CPR itself.
This is clearly a problem for the public Web, since we want the Web to be for everyone. Maybe we shouldn’t think of the public as the ones withdrawing “resource units” from the Web, but rather aggregation tools, or a similar idea? That’s also a problem as long as the tools are owned by corporations instead of people. [Original]
- Congruence between appropriation and provision rules and local conditions
This basically says that we should expect to figure out our own rules for the Web and not be able to just copy them from elsewhere. [Original]
- Collective-choice arrangements: Most individuals affected by the operational rules can participate in modifying the operational rules.
This is again a problem if we expect 3–7 billion people to participate, but maybe less so if there’s a way to only constrain the people who risk damaging our commons. [Original]
-
Monitors, who actively audit CPR conditions and appropriator behavior, are accountable to the appropriators or are the appropriators.
-
Graduated sanctions: Appropriators who violate operational rules are likely to be assessed graduated sanctions (depending on the seriousness and context of the offense) by other appropriators, by officials accountable to these appropriators, or both.
These principles are followed by a long section with several insights… [Original]
One bit reminds me of an institution adjacent to the Web: “graduated punishments ranging from insignificant fines all the way to banishment” sounds like the system by which browser root programs regulate Certificate Authorities. [Original]
Back to the Web, this section discusses how monitoring for rule violations has to be built into the way people follow the rules themselves, or has to be directly funded by the rules. We’re starting to think about this in principles like https://www.w3.org/TR/privacy-principles/#transparency, but we have a long way to go in incorporating monitoring into the ecosystem itself instead of external researchers. [Original]
- Conflict-resolution mechanisms: Appropriators and their officials have rapid access to low-cost local arenas to resolve conflicts among appropriators or between appropriators and officials.
This feels straightforward to do well once some rules are in place and not worth worrying about before some progress has been made on other parts of the puzzle. [Original]
- The rights of appropriators to devise their own institutions are not challenged by external governmental authorities.
This will be a problem. We’ve already seen this to some extent in the CA Root Program space with CAs appealing to European regulators when they didn’t want to follow the CAB Forum rules. With the Web mattering so much to so many governments, it’ll be easy for defectors to ask a government to intervene even if we get the commons management right quickly. [Original]
- Nested enterprises: Appropriation, provision, monitoring, enforcement, conflict resolution, and governance activities are organized in multiple layers of nested enterprises.
It’s not clear to me how this will apply to the Web. This might be another principle like (6) where it’s just a reminder to get it right once we’ve found some overall rules. [Original]
Greg Bloom started an interesting sub-thread at this point:
@jyasskin this describes how the web works doesnt it? IETF, W3C, DNS, national policies, eurozone, etc. another way to understand this is polycentricity— overlapping domains of decisionmaking at various scales with various subdomains within them. Which is also to say, these principles arent applicable to one thing, because resources arent usually discrete things. It’s fractal.
It’s been pointed out (see the edited OP for credits) that researchers continued to study this aspect of the commons problem at https://knowledge-commons.net/. I’m going to keep live-blogging this book, but my readers should keep in mind that I’m just a programmer who’s new to the space, so many of my observations and guesses are likely to turn out to be naïve compared to the discoveries from 34 more years of research. [Original]
Case Studies
Chapter 4 is about how Southern California water basins created systems to manage their commons. This is the setting in which Ostrom developed her theories in the first place.
These stories strike me as the most like privatization I’ve seen so far in the book. They’re different from the traditional interpretation of the Tragedy Of The Commons in that the distribution of the new private property was negotiated among most of the appropriators, but they still came out with tradeable property. [Original]
Pg 129 discusses how the water basin associations organized a state law to let each water commons create an institution to manage itself. That worked because the state had many distinct water commons with local features.
The Web, on the other hand, spans many legal jurisdictions and only exists once in each, so it doesn’t make as much sense to create a legal framework for, say, the W3C to instantiate. [Original]
Ah, on pg 136 she distinguishes between the privatized rights to the resource units made available by the water basins (acre-feet of water/year) and the communal management of the basins themselves. If they were actually just privatized, the basins themselves would also be tradeable. [Original]
Pg 140: “The origins of institutions and changes in institutions frequently are [incorrectly] considered to be fundamentally different. Endnote: This distinction characterizes my previous work.”
I like that she was willing to point out her previous mistake.
This is also a really key point: unless you have overwhelming power, as do some governments, monopolists, and monopsonists, you can’t invent a new complete set of rules from while cloth. You can only incrementally evolve what you have. [Original]
Overall, chapter 4 was too narrowly focused on one kind of commons in one US state to feel like its examples are going to be very useful in designing a future for the Web.
It’s good that it indicates that subsequent researchers probably focus on how to evolve institutions (which we already have some of for the Web) instead of staying stuck on creating them from scratch, so there’s probably good research around for me to keep reading. [Original]
Skipping a couple stories from chapter 5, pg 149-157 discusses a fishery where national officials kept overriding the participants’ collective wishes, like eIDAS threatened to do to the certificate authority system in Europe. In fisheries where this didn’t lead to overfishing, it instead led to private ownership of the whole fishery with “undesirable … distributional consequences” due to their monopsony position driving down wages. [Original]
“Irrigation engineers [in Sri Lanka] strongly identify with the civil-engineering profession, in which esteem derives largely from designing and construction public works, rather than operating and maintaining them.”, pg 164.
Software engineering has the same problem, as does the promotion system at Google, at least. [Original]
The last story in chapter 5 has an endnote discussing how technological advancements can destabilize the agreements governing a commons. The commons in that section convinced the early adopters to discard the new equipment they’d bought.
We’re definitely seeing this in the Web as LLMs overturn lots of existing conventions and shift the balance between (what this book calls) appropriation and provision. [Original]
Chapter 6, pg 190: “Success in starting small-scale initial institutions enables a group of individuals to build on the social capital thus created to solve larger programs with larger and more complex institutional arrangements.”
This will be a good reminder for W3C leadership like me, who will be tempted to try to organize the whole Web at once. [Original]
This was implied by some of the water basin stories in Chapter 4, but Chapter 6 pg 196 emphasizes that institutional change is helped if data about the state of the commons is collected and distributed to all participants.
The difficulty of this is illustrated by the development of the Privacy Sandbox over the last several years: despite lots of research, there’s still no consensus around how much online advertising revenue depends on cross-site data, or how that revenue is distributed. [Original]
Enablers for Institutional Change
The book ends with a list of characteristics that Ostrom expects to predict institutional change to protect common-pool resources. Starting with the “internal” characteristics, and adding my comments about the Web:
- Most appropriators share a common judgement that they will be harmed if they do not adopt an alternative rule.
This probably depends on the part of the Web and may be changing quickly when it comes to LLMs. [Original]
- Most appropriators will be affected in similar ways by the proposed rule changes.
We don’t yet have any proposed rule changes, but this seems very unlikely given the variety of entities using things published on the Web. [Original]
- Most appropriators highly value the continuation activities from this CPR; in other words, they have low discount rates.
I’m worried about both the startups and large companies here: VC-backed startups get high discount rates from their VCs and short runways, and several of the long-standing companies in this space seem to have recently acquired high discount rates and demonstrated them in laying off employees. The experienced individuals do value the continuation of the Web. [Original]
- Appropriators face relatively low information, transformation, and enforcement costs.
I think we have high information costs: it’s hard to get trustworthy information about the viability of production for the Web or the effects of various rule changes.
We have good venues in which to discuss and agree on rule changes, and lots of experience finding consensus in those venues.
Enforcement costs are high, requiring use of multiple national legal systems. [Original]
- Most appropriators share generalized norms of reciprocity and trust that can be used as initial social capital.
This is hard to evaluate when the actors here are companies, often huge ones, who send different kinds of individuals to negotiate for them. The technical representatives to standards bodies have these norms and social capital. I suspect the legal and business representatives don’t, but I don’t have the experience to be sure of that. [Original]
- The group appropriating from the CPR is relatively small and stable.
Ha ha ha no. [Original]
Ostrom doesn’t present a numbered list of external characteristics, so I’ll try to paraphrase here.
- A political regime that allows substantial local autonomy.
We have a long history of this on the Web, and of fighting off attempts to centrally regulate the Internet and Web. However, recent history might be showing a reversal of that trend. [Original]
- A political regime that invests in enforcement agencies.
This one’s mixed: there’s generally good enforcement for legal decisions within the countries that large multinational companies care about, some recent enforcement of privacy and competition regulation, and very little effective enforcement of anti-fraud norms. [Original]
- A political regime that provides generalized institutional-choice and conflict-resolution arenas.
I might be missing an aspect of this, but I suspect this one’s a “no”, given how many countries need to be involved in any legally-enforceable agreements that would cover the Web. That said, we have lots of standards venues that can serve as conflict-resolution arenas. [Original]
And that’s the book! It’s definitely been helpful in starting to think about how to manage the commons involved in the Web, but I’m also looking forward to learning what the next 30 years of research discovered at https://knowledge-commons.net/. [Original]
I also live-blogged reading part of Governing Knowlege Commons.