Fat protocols and the Naked Chef

In the late 90s, at the height of the dotcom bubble, I bought Jamie Oliver’s first book - The Naked Chef - from a startup online bookseller. I know, I was young and foolish then, however the fish finger sandwich remains a firm favourite.

You’re probably already thinking it came from Amazon, but back then there were dozens of companies in similar stages of development, raising money through Initial Public Offerings (IPOs) and competing for market share with extravagant marketing campaigns.

So when I say bought, perhaps I should say I acquired it. While The Naked Chef was being sold at its RRP of ￡20 at bricks and mortar stores, this particular e-commerce bookshop (I can't recall the name) was selling it for ￡9.99. They also freely distributed ￡10 vouchers, which could be used towards a selection of books, including The Naked Chef. They even included free P&P.

So, for no money, the little effort of completing an online form, they posted me a hardback book that would cost ￡20 in the shops. Unsurprisingly, they collapsed shortly afterwards, as did many others.

The great survivor from those days, Amazon, was already evolving from being just an online bookseller. They developed a platform capable of selling almost anything. And then this became the platform for other sellers. And with Amazon Web Services, they developed a cloud computing platform for all manner of services, including Netflix, Unilever and AirBnB.

These services have all been built off the back of internet protocols. Despite the enormous value dependent on the protocol layer, it is argued that the value of the internet has largely been captured in the application layer by companies like Amazon.

The line between platform and protocol is not always distinct, so perhaps we should also consider a platform layer between protocols and applications, or as the base of the application layer.

Fat Protocols

The Fat Protocol argument developed by Joel Monegro is that this dynamic will be reversed as blockchain becomes the foundation protocol. It will be harder for applications to capture monopoly value as identity and data is embedded in the protocol layer.

One view is that the value of the protocol will always grow faster than the combined value of the applications built upon it. This is pushing application developers to issue their own coins in Initial Coin Offerings (ICOs) to capture more of the value that would otherwise reside with the underlying protocols.

From IPO to ICO

More firms went public in the US between 1996 and 2000, than in the following 15 years combined. Dotcom companies found they could attract significant investment by selling shares through an IPO. The valuations achieved, often on little more than a business plan, raised astronomical amounts which made up for the regulatory burdens of a public listing.

The parallels between ICOs and the internet IPOs have been noted by the Financial Conduct Authority (FCA). Young companies are again using the lure of transformational technology to raise capital, with white papers substituting for business plans, and mention of blockchain for a .com suffix. And like then, there may be value in many ideas, but most will fall by the wayside.

In a piece on LinkedIn, Ray Dillinger, who reviewed the early blockchain code for Bitcoin, bemoans that "many billions of dollars of scams and failures and thefts have been perpetrated by abusing people's faith in and enthusiasm for that technology" as due diligence has often become perfunctory at best.

While competition during the internet bubble was at the application layer, the battlefield is not as clear for ICOs. The wary investor, for the ICO is (probably) an investment, would do well to reflect on whether the coin has value at the application or protocol layer.

For if the offering is an application, the coin is unlikely to be intrinsic to the underlying value. In the long run, without a monopoly over data and identity, the value will be dependent on the future utility of the service and its ability to retain customers. Much like the traditional approach, albeit beset by uncertainty, of discounting future cash flows to value shares in an IPO prior to the dotcom boom.

Peer to peer (P2P) energy trading projects, whether blockchain based or not, currently sit as applications on top of the existing market protocols, delivering little additional value or liquidity.

If, on the other hand, the value is in becoming the protocol, what is the likelihood that the organisation not only develops a protocol, but also engages the industry to replace incumbent service providers, who will not give up their position willingly?

Could P2P replace the Grid Trade Master Agreement (GTMA) as the protocol for energy trading, and the brokers and exchanges that match buyers and sellers? In a highly regulated sector like energy, this is not just a commercial endeavour, but often requires convincing regulators to support the changes.

Perhaps fearing a repeat of the dotcom crash, China's central bank has declared ICOs illegal. The SEC, FCA and CSA among others, have recently issued guidance notes, with the SEC determining that the DAO tokens were securities and should have complied with federal laws, but decided not to pursue enforcement action. William Mougayar recently published a guide for Safe ICO Practices in hope of regulatory largesse for "good" ICOs.

Take a look at BankyMoon, Power Ledger, SolarCoin and GridPlus for examples in the energy sector. Assuming they manage to stay within regulatory strictures, which of them will become the future protocols underpinning the energy sector, which are applications with a distinct value, and which will end up forgotten like so many dotcom booksellers?

Update

At about the same time this was first posted, Vitalik Buterin tweeted that "Open source infrastructure projects currently are struggling to get funding without the ICO+token route".

This makes me wonder whether investors do not accept the fat protocol theory. After all, even if it is no longer possible for the application layer to monopolise value, it does not mean that it be predominantly distributed to the protocol layer.

Perhaps they are unwilling to let go of the internet paradigm that "investing directly in protocol technologies generally produced low returns", or maybe there is more in the idea that value was disproportionately distributed to a platform layer.