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!
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.
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.
Words have different meanings depending on contexts.
Compare „string“ in computer science vs astrophysics.
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. :-)