library

The library analogy

This article was first published on Dr. Craig Wright’s blog, and we republished with permission from the author.

Some people think that storing data on Bitcoin is analogous to a public service. There is an error with this line of thinking. A better analogy would be to think of storage in Bitcoin as one in a public library. We can imagine access to files in Bitcoin as access to a physical book in the library. The existence of a library does nothing to stop people buying the book themselves. In fact, the inclusion of people with the ability to buy the book improves everybody’s access.

First, I shall explain the library analogy, after which I shall explain how it applies to Bitcoin and related services.

We will start by imagining a small community of 100 people who have a library with two of every book that the individuals want to read. Anybody can go to the library for free and access books for a limited amount of time. They can take notes and study using the books, even if they do not have a lot of money. Alternatively, some people might want to buy a book for themselves.

If we imagine that up to 5 people may desire any particular book at the same time, we can see that there is a queueing problem: if there are only two books, and individuals can take either book out for up to a month, it may take several months before others can get access to a book. If the book presents required reading for a university course, there could be a problem. Luckily, property rights and capitalism provide a solution.

Some individuals will be richer than others. Such individuals are likely to simply buy the book and unlikely to use the library. At times, some of the richer members of society, after they have finished with a book, may even donate books to the library. So, to look at such a scenario, we need to imagine a Pareto distribution of wealth. It is a scenario we find in every capitalist society.

Now, one of the five people will be rich enough to never need to go to the library and to buy any book they want, donating whatever they finish reading at the end. The donation increases the stock of library books. The example also shows us that the queueing problem can be minimised. The richest individual will simply buy the required books, without ever going to the library. In doing so, we now see that only four people are queueing for the book.

Next, some individuals will have free cash, but may prioritise other things if they’re able to get the library book. So, if we now change the lending strategy, allowing individuals to borrow the book for a week at a time, with requirements to return the book or be charged its value or some other late fee and that they cannot immediately take the book out again, we see that all of the individuals queueing for the book can obtain it within two weeks.

Some individuals will, still, change their preferences and buy the book. We can now say that two people have purchased the book, leaving three who go to the library. As such, we can see that we have a shared public resource that has been made viable through individuals with more available capital, although stretched.

At the end of the semester, the rich individual may have a chance of leaving the book to the library. If we assume a probability of 50%, we can say that on average, 50% of the time, there will be sufficient books from now on. Next semester, if five people want the book, there are likely to be two individuals who purchase it, with a 50% chance that there will be three copies at the library. As such, some individuals’ willingness to pay for goods and services results in an increase in the value of public goods and services, and allows more people to access them.

How does it apply to Bitcoin?

It has been falsely believed and promoted that data will be either free forever or offered as part of a public service. The inclusion of information on the Bitcoin blockchain is not part of a public service, but a commercial activity. Some nodes will always provide archive services. Thus, information that has been placed on the blockchain will always be available. The distinction is again linked to queueing theory.

An operator of a node may have 100 machines in a server farm providing access to the blockchain. Such machines will provide ready access to UTXO sets and information to ensure that people can access and spend their money. Equally, of those 100 machines, only five of the 100 machines are likely to offer archive services. Each company will have made decisions based on the levels of profitability that they expect to achieve.

Individuals seeking to access information are analogous to individuals going to the library. If they want to gain access without paying for the service, they will have to queue in the same way that individuals queue for the book at the library. Some people will not want to wait. Some service providers will want to make sure that there is ready access to the data. Rather than leaving such decisions to node operators’ whims, corporations storing data on the blockchain will form contracts with node operators, to make it more available and allow faster access.

What we will see is the creation of competitive markets. A company that wants fast access will pay each node, and potentially replicate services, to allow its clients to gain ready access to stored files. The internet right now is served by companies such as Akamai, providing edge services and cache. Such organisations allow ready access to information for a large number of providers around the world. Similar services will develop to ensure that information on the Bitcoin blockchain is readily available.

The investment in allowing access will provide more bandwidth and copies of the complete blockchain and speed up access to files. Yet, access to files that are not widely replicated will become slower, as the bandwidth required to serve such files to a large number of customers will be limited. If you require that people can access a particular file or piece of information without cost, the process of caching it will not occur, and archives will be slow to respond. They will still exist, but it could take a while to download or locate information.

Consequently, the growth of the Bitcoin blockchain will mirror other commercial services. Individuals who form corporations and want their data to be readily available will find more copies and access to be more straightforward. Over time, some services will invest more than others and mirror the information more widely. Some archive services will only mirror part of the blockchain, but allow individuals to prove existence and probity using simplified payment verification (SPV), with the ability to provably tie a transaction and hash to the blockchain’s history.

New to blockchain? Check out CoinGeek’s Blockchain for Beginners section, the ultimate resource guide to learn more about blockchain technology.