The Network Effect
The early 21st century cultural/informational landscape is dominated by network effects. Large entities like corporations and governments leverage their networks to extract the quadratic increase in value that each marginal user adds.
Each new user to a network has a marginal cost of acquisition, which is linear in the number of new users. This makes sense, as the cost to onboard new users is constant, that is the same for each new user. Some marginal storage, compute, maintenance, etc will be required to maintain the new user’s network presence.
However, the value of the underlying network does not scale linearly, it scales quadradically with respect to, not the number of nodes, but the number of connections between nodes. This is a restatement of [Metcalf’s law](https://en.wikipedia.org/wiki/Metcalfe’s_law), which says that the value of a communications network scales proportionally to the square of the number of nodes in that network, which approximates the number of connections between the nodes, which is n(n-1)/2, where n is the number of nodes.
Putting it together, the cost to operate the network scales linearly in the number of users, but the value of the network scales quadratically in that same dimension. So each marginal user pushes up the profit margin, some incremental amount, higher than it was with n-1 users. Call this the “network effect dividend”.
These large organizations, by gate-keeping their networks, have a natural choke-point where they can control access to the network, and thus monetize the increasing marginal return to user growth. Rents in our present day for use of these closed networks tend to come from targeted advertisement and data collection/arbitrage, but could also be as simple as fares to access the network or even elevated privileges, a la X.
The Enshittification
If you tag words like “centralization” with a negative sentiment, as I do, that previous section sounds dark. It did not seem this way to the general populace while the darkness was growing. For over a decade, as the benefits of all these new networks were manifest, and the cost of centralization was not yet understood, Cassandras like Cory Doctorow, one of the classic cantankerous internet personalities, were as easy to ignore as it was to walk past the end of the world people in Times Square. The consensus was that these networks were finally paying out the dividend of decreasing unit cost of communication to the user, realizing the long-tail of relevancy and connection.
Doctorow recently coined the word “enshittification” to describe the life-cycle of these large centralized networks operated for profit. In his own words:
Here is how platforms die: first, they are good to their users; then they abuse their users to make things better for their business customers; finally, they abuse those business customers to claw back all the value for themselves. Then, they die. I call this enshittification, and it is a seemingly inevitable consequence arising from the combination of the ease of changing how a platform allocates value, combined with the nature of a “two sided market,” where a platform sits between buyers and sellers, hold each hostage to the other, raking off an ever-larger share of the value that passes between them.
The “platforms” in Doctorow’s definition are the previously mentioned closed networks: Twitter, Facebook, Instagram, Google, etc.
Doctorow’s first stage, making things better for their users, is where the business comprehends their value as a network and wants to grow that network relative to other networks, taking a larger share of the total addressable market. The only way to do that is offer a product for “free”, and establish the centrality of your network. This stage, as it relates to the first crop of networks began when they were created in the 2000’s and runs up into the early 2010’s.
The second stage is when the platform bosses look around, see the regularly increasing network potential, and set about to convert that potential financial energy into kinetic financial energy by “making things better for their business customers”. This involves making things worse for the users, as the closed network must then use some of its clout to direct this latent potential down paths the business customers would prefer, either in the form of advertisements slip-streamed into news feeds, selling out of user data from the back end, or acquiescing to government requests to shape the flow of ideas by censorship of various means. This stage followed on after the networks started to see their potential. By 2016 the network influence narrative flipped from the techno-optimist story to the potential of disinformation, foreign influence, and extremism, as this new narrative served the monetization of these networks and provided cover for decreasing utility to the users by blaming outside forces.
We now live in the terminal stage of Doctorow’s formulation where both users and business customers are squeezed to generate more return from the increasingly sclerotic and bureaucratic structure of the company behind the network. With rampant inflation, high interest rates, thrifty consumers, and political unrest, the platforms all closed ranks to keep the party going for their shareholders. Not all of them have or will succeed at doing this, with Musk’s takeover of Twitter as a lesson for how serious one should take public perception of a network’s future prospects, the stock price, of said companies. If it drops regularly and or suddenly, share holders will sell, kicking off a cascade of value loss for the network owners, or they will make drastic changes to shore up their position, up to and including selling the company into a new business model or a wholly new management approach and or team.
A Swerve
Even as the first signs of trouble that would portend a phase-change into the second stage of enshittification were appearing, the global financial crisis of 2008, many individualist early adopters of the internet were sounding the alarm. One nym was out there building a parallel structure to the closed network: Satoshi Nakamoto.
Bitcoin is an example of an open network, the antithesis of the closed network platforms. Accessing to create a new identity in the network, by creating a private key, a wallet, is fundamentally decentralized. Anyone can create as many new Bitcoin wallets as they’d like without any consent or action required or possible, by the operators of the network. Just like the adoption of the HTTP protocol, no one case stop an individual from creating a new Bitcoin wallet or standing up a new website.
The closed network or platform that Bitcoin challenges is the fiat money system. The monetary and financial system, at least in the United States, is an undeniable example of the network effect. When it was created the US Dollar served its users by providing a regular unit of account that allowed the coordination of matters across time easier than before. Then in 1933 and 1945, the monetary system was modified to benefit the business customers (other nations) and platform owners by blocking the exits. In 1933 the use of gold as money was outlawed in the United States (Executive Order 6102 ) and in 1945 the Bretton Woods agreement (Bretton Woods sysem) prevented unaligned interoperability.
The final stage, where the Dollar is printed widely to enable spending of wealth the government doesn’t have, was enabled by going of the gold standard completely in 1971, and further signposts to user and business customer abuse are noted during the global financial crisis, chronic zero interest rate policy, and covid-related monetary base expansion. The outcome is inflation and restriction of individual liberties justified by tinkering with the broken system.
Bitcoin offers a swerve off this path by establishing a parallel structure to the monetary system and providing an answer to “what would you do differently?”. In the world of monetary systems the network effect dividend already has a name: seigniorage. This is the income realized by the entity that controls the currency. In the gold based systems, the mint charged a premium to coin gold into the standard units of currency, ensuring their quality and quantity. In modern times, central banks create new money and spend it, thus reaping the benefit of their position astride the monetary platform. However, in Bitcoin there is no special access to the network, no insiders. Since the people running nodes ultimately constituency, their consensus is what makes Bitcoin work, and node runners tend to be rabidly opposed to enshittification.
Bitcoin incentivizes those who get in early and hold bitcoin. This incentive is doled out proportionally to the value you provide the network. The more BTC you hold, the more your purchasing power increases over time. You can see this in a comparison to mainstream monetary theory: that the money supply itself should expand and contract with the size of GDP. In that scenario seigniorage is consumed by the central banks that monkey with the supply, they spend or loan the newly created monetary units, while the holder is theoretically left with a currency that has stable purchasing power. Except that is not how things have played out. Purchasing power has declined since the establishment of the Federal Reserve system in 1913 such that \(1 in 1913 has the same purchasing power as \)31 does today. Where the opposite is true for Bitcoin: it’s purchasing power has gone up so much since 2009 (when the first block was mined), that it’s difficult to calculate.
##
Here we are, back at the present day, our culture and communications encircled and enthralled by a web of enshittified platforms. NOSTR is the analog of Bitcoin in our present scenario. Just as Bitcoin has no insiders and incentivizes those who prvide value to the network, so too does NOSTR. Any user of an app that interacts with NOSTR benefits from new users onboarding anywhere in the protocol. This is unknown to the platforms. It would be like Twitter getting new users, followers, and content every time a new use signed on to Facebook, or TikTok, or any other of the currently closed networks. When someone installs Damus on their iPhone, Amethyst gets first-class access to that same new user, their new follows, and their new content.
The network effect dividend is, through this method of incentivizing those who provide value to the network, driving the allocation of resources towards efforts that benefit the larger network. There are no insiders with special shares or air dropped tokens or preferred stock that can swerve NOSTR policy. It’s an open network in the same way the Bitcoin is. This model inherently disincentivizes enshittification by reducing switching costs of network provider to nearly zero. You can stop using Damus and switch to another client without losing your profile, follows, following, etc.
Doctorow includes in his suggested remedies the use of anti-trust laws and regulations to make it harder for companies to enshittify. I don’t share this outlook. The government is itself an enshittified platform, and any solutions it proffers should be assumed to be poisoned by present or future abuse and enshittification. The only obvious pathway out of this forest is to build our own unenshittable systems.
The Unenshittables
I’ve named two unenshittable networks, Bitcoin and NOSTR, in this post. Are you aware of others? They could be other modes of communication, business, cultural transmission, anything where network effects reign. Respond to this post or hit me up directly, I want to hear what other networks are out there, and in which areas.