The funding problem exists only for public goods in the sense they are defined in economics: This definition requires that the thing being produced to be nonrival and nonexcludable. The post conveniently avoids discussing what projects they will consider to be "public goods". Non-economists tend to think simply of either goods/services produced by the government sector (a whole lot of which are actually quite rival and excludable) or people making stuff no one wants to pay for ... Neither category requires a special funding mechanism.
Voluntary contributions mechanism (VCM) with a provision point is well known (this paper[1], for example, provides a decent summary as well as identification of a problem). This seems to be a contorted effort to take VCM-PP and retrofit it into some kind of blockchain/crypto-currency framework.
The economic definition could use an update, or at least some asterixis. Lin Ostrom's work is worth looking up, if the topic interests you.
Both rivalrousness and excludability can get gray, changing as technology or legality changes. As intangible goods become a bigger part of the economy, this matters more.
Take HN. Is it rivalrous? If I comment on HN, does this diminish others' ability to comment on HN? There isn't a totally black and white answer to this. Usually, no. Other people commenting is actually required for a forum to have value. OTOH, the premise of brigading (was once a reddit thing) is that a bunch of people flooding into a forum can diminish people's enjoyment of it. I suppose you could call that vandalism, but there are plenty of cases featuring various shades of grey. If your mom joins instagram, maybe you can no longer enjoy commenting on salacious pics.
Ostrom advocated for degrees of, rather than binary definitions.
Excludability... Also debatable and shifting. Digital music wasn't excludable for a while, then it was. Anonymous social networks (like HN) are less excludable than real name networks (like FB), but it's mostly a technology problem. It's also a legal question. The idea of making FB and such "common carriers" was floated around, which would make is legally nonexcludable.
Patents are the opposite. The exist to make nonrivalrous, non excludable goods excludable. So does DRM, copyrights. A GPL license turns an excludable good into a nonexcludable, arguably public good.
IMO, a better starting point is scarcity vs abundance.
Anyway, considering the importance of public science, software like linux & the www... I think it's worth thinking about how important the commons/public domain/classification of your choice is.
> The economic definition could use an update, or at least some asterixis.
My goal is to provide some of the asterisks. The "market failure" theorems require the economic definition. One cannot presuppose the existence of the inefficiency in circumstances where the assumptions of the theorems are not satisfied.
Take medical research. Given the existence of patents and the crappy medical papers I am seeing everywhere, I am not convinced there is an underfunding problem. Education is very much excludable and to a great extent rival (especially if you consider the issue of finding the appropriate environment and teacher to learn from).
Of course, institutions matter ... That is not the issue here. The question is whether this crypto scheme exists to eliminate a well-defined inefficiency in a coherent way, or does it just boil down to a heap of presuppositions?
>Neither category requires a special funding mechanism
Can you elaborate on this? What effective ways are they funded today? Public goods are chronically underfunded even in the western world so alternate ways of funding these projects seem beneficial to everyone.
One major problem is that the internet is global and most public goods funding happens at the nation state level which leaves a giant hole where too little funding exists for species-wide public goods such as climate change, medical research, and anything related to the internet which is all of the public goods like open source libraries, cryptography, and Ethereum projects like clients, base layer contracts, and other open source infrastructure.
Having a public goods funding mechanism at a global scale seems necessary for our increasingly interconnected world.
He said them: medical research, and anything related to the internet which is all of the public goods like open source libraries, cryptography, and Ethereum projects like clients, base layer contracts, and other open source infrastructure.
I'm a bit confused by the first paragraph. Are you saying that public good funding problems don't exist (or are solved b/c they are either funded by the government or aren't worth funding)?
If so, I strongly disagree. The OSS that makes up the backbone of all-the-computation-that-occurs-anywhere is certainly a public good; it's also way underfunded!
> Are you saying that public good funding problems don't exist
No.I am saying people tend to call a lot of things "public goods" when they are not.
If something is rival or excludable, it is not a public good in the precise sense in which economics defines them.
If something is rival and excludable, then the socially optimal level of funding is not achievable through markets (note that the socially optimal level of funding may indeed be zero).
Software is not rival (me linking with a library does not prevent you from being able to use it). However, it is excludable: A legal structure exists whereby people can be prevented from using the software.
Now, assuming something like the MIT license, you now have a nonrival/nonexcludable thing. That does fit the definition of a public good.
If you're sticking to the Paul Samuelson definition, it does. Its also the main case the author is arguing for
That said, what's the purpose of sticking to the definition? I mean, the whole concept basically exists to determine a class of things that should be publicly funded. Since then, many of the goods that were public goods were changed by technology into something else.
A classic example was free-to-air radio and TV. Public good theories were used to justify TV licences. Since then, we found both ways of excluding (cable) and ways of making money that aren't related to exclusion (advertising). Did free-to-air TV stop being a public good once the ad business got lucrative enough. Maybe it stopped being a good altogether.
The definition of a public good was never precise, but it's downright squishy these days. Quite a lot of goods these days, both capital and consumer are nonrivalrous (could also be described as non subtractable, abundant..).
Anyway, the author's usage is pretty clear. FOSS like the worldwideweb is a public good for the purposes that he's using the term.
> determine a class of things that should be publicly funded.
No. That is a misunderstanding. The concept allows the possibility that the standard competitive market price discovery mechanism is not going to lead to the optimal level of production of some goods and services. The word "optimal" here is in the usual sense of maximization _subject to constraints_.
Even if we all agree that something is a "public good" in the economics sense (therefore, the market mechanism will not be able to provide the optimal level), it is not a foregone conclusion that a government bureaucracy will be able to provide anywhere close to the optimum. For that, they would have to discover the "true social benefits" and "true social costs" created by each additional unit produced. That process is subject to usual political processes and the outcome is determined by who owns the politicians.
> Who said anybody was using the niche academic economist definition of „public good“?
Except that if something is not a "public good" in the economics sense, then the burden of proof is on the claimant that there is an under-provision problem to be solved. Is this crypto scheme geared towards solving an actual problem or is a money-making scheme for its proponents?
> Compare „string“ in computer science vs astrophysics.
That comparison is disingenuous because the scheme about "Funding Public Goods" using economic incentives and mechanisms. We are in the realm of economics here.
See also[1]:
> Co-author here. I definitely use the phrase "public goods" in the economic sense: goods that are non-rival and non-excludable.
Therefore, it seems like they are trying to build a binding scheme to help beneficiaries express the value they put on the provision of the public good. There is a rich literature in economics about mechanism design that should reveal the difficulties involved.
Separately, it is important to keep in mind that things that are nonrival and nonexcludable can also be "public bad"s. :-)
So far, I've seen "public goods" used in the blockchain world to mostly mean open source software projects concretely, but something broader aspirationally. For instance, I believe the large and well-known Gitcoin project pretty much follows this pattern. I'm not sure how well it lines up with that economics definition. Thanks for the link to that very interesting paper!
Co-author here. I definitely use the phrase "public goods" in the economic sense: goods that are non-rival and non-excludable.
This post has nothing to do with assurance contracts (which that paper seems at first glance to be equivalent to). Instead, it's tackling the different problem of what if there is some established mechanism that is already collecting fee revenue, and it wants to do use that revenue to support its wider ecosystem.
The article start with a very pertinent question about how to offer some exit mechanism for FOSS project similar to what a for profit startup can count on.
It then try to convince you that blockchain is the way to solve this question.
Therefore it end up being yet another article focused on defining a problem for the blockchain to solve.
Is the centralization of an oracle actually useful?
I wonder if people curated lists of "public goods donation addresses" and then hosted something like a leaderboard of people who donated most to any/all public goods addresses, so people or companies could be recognized with social clout (if they want) for giving back. Automatically make a page for me that shows who I've donated to a when. Would that accomplish the same goals as much or better?
(I wonder if there is an opportunity for a tax exempt NGO middleman org(s) to validate 'public good" status and make donations that flow through deductible?)
Like, if I could read an article about someone who did something societally beneficial that probably wont 'pay off' and it was easy to donate $20 to them, I might do it sometimes. Sending $20 to an oracle fund trusting they'll allocate it wisely feels weird to me - I'd need to see a long record of smart allocations to trust them.
The oracle is not monolithic for information-optimization reasons. Rather, it's monolithic because there's a single project that has funds that it's willing to give out in retroactive grants, and so _some_ mechanism must decide where those funds go.
(This is perhaps what's not clear to readers from outside the crypto space: all of these discussions are assuming that this is being implemented by some project that has network effects and can extract fees that are necessary anyway for congestion-control reasons, where the only remaining question is where those funds go)
I see, if these funds aren't soliciting donations from the public, my concerns are mostly irrelevant. (Though could be a cool parallel project, I'll ponder this a bit more.)
I'd like to see all people thrive, even if they aren't producing any tangible or identifiable good yet.
I believe that most people would naturally do more good if they had less stress, a stronger safety net, and more free time.
Children are often given at least a small delay before we expect them to start producing economic value. We know that there is value in this delay, do we know there is value in cutting these things off?
That's a lot of words to say that some money goes into a prize pool, and is awarded by a hand-appointed committee.
It would be interesting if the prizes were awarded based on some kind of trustless distributed process, but that's explicitly not happening here. Without that, it doesn't seem like a blockchain is adding anything important to the existing (though underused) concept of incentivizing innovation through prizes.
This sounds like a distributed way to award prizes, but it's entirely abstract. What prizes do they plan to award this way, and using what criteria? It's not going to help any projects if they can't anticipate what they might win.
> Funding by setting a price floor [on a project token]
Aka, the reward is the certainty that you'll be able to trade your tokens in for some minimum price. For the end user, at the end of the day, the prize is $$$.
This is the hidden layer/feature behind blockchains it seems most people don't realize. They are amazing instruments for transparent social coordination and funding. The things happening with DAOs now and multisigns are incredibly socially positive and inclusive.
Voluntary contributions mechanism (VCM) with a provision point is well known (this paper[1], for example, provides a decent summary as well as identification of a problem). This seems to be a contorted effort to take VCM-PP and retrofit it into some kind of blockchain/crypto-currency framework.
[1]: http://www2.econ.iastate.edu/faculty/vesterlund/paper/overpr...