The Weekend Read: April 10

tim bubble 1 There is so much to get to in this week's edition that news of a presidential candidate accepting bitcoin and the release of CoinDesk's Q1 update to their mega-deck The State of Bitcoin gets relegated to the intro aside...

1. Consensus-as-a-service by Tim Swanson

A must read report on the differences between permissioned and permissionless ledgers, of which I was a proud but minor contributor. You can also see a nice review of the article's findings in this American Banker story. I asked Tim for a "Weekend Read Exclusive" quote on his report and the key highlights:

Distributed ledgers could provide innovative solutions to the clearing and settlement space. Bitcoin and cryptocurrency are all about censorship-resistance transactions. In contrast, distributed ledgers/consensus-as-a-service are instead about fast and secure updating of property titles and securities interests. This may sound mundane but could radically change things by disintermediating certain players (such as clearing houses) while delivering to regulators the benefits of increased transparency and mitigation of systemic risks. This consensus-as-a-service could also deliver cost savings via automation, eliminating some human intervention such as those in back office processes.

I think Richard Brown summarizes the idea well in three words: "replicated, shared ledger." In a recent email he said, "it's not a "blockchain" that gives Ripple, Hyperledger, Clearmatics or the rest their value from what I can see... it's the fact that multiple mutually-untrusting parties can come to agreement about some shared state/facts/logic -- thanks to having their own copies of some state, which a "system" guarantees to keep in sync for them."

2. Banks and the Blockchain (cont.)

Following on from last week's announcement from UBS, we have another WSJ story of bank experimentation, with BNY Mellon kicking the tires on the Bitcoin blockchain:

Despite the potential risk to their businesses, banks are intrigued by bitcoin’s underlying blockchain technology [snip] Mr. Kumar says that he is curious how the software can be used to make financial transactions more efficient

Perhaps even more shockingly, JPM CEO Jamie Dimon dares utter the "b" word in his recent letter to shareholders. A review of the article can be found here, but the letter itself is an interesting read (from Page 29):

You all have read about Bitcoin, merchants building their own networks, PayPal and PayPal look-alikes. Payments are a critical business for us – and we are quite good at it. But there is much for us to learn in terms of real-time systems, better encryption techniques, and reduction of costs and “pain points” for customers.

Some payments systems, particularly the ACH system controlled by NACHA, cannot function in real time and, worse, are continuously misused by free riders on the system.

The always interesting Dave Birch has an interesting review on all the blockchain love, balancing the hype with the promise:

There is an ongoing discussion and evolution of serious thought around what the distributed ledger can be used for, whether proof-of-work or proof-of-stake scales best and whether blockchain or consensus protocols are the best decentralised option. All I will say here is that in our work with clients (large financial institutions) on the topic, currency is probably the least interesting use of the ledger that gets discussed

What is driving all this interest? The growing fear that the age of asymmetric information (or access) is coming to a close. This development is examined in detail in this excellent article by Tyler Cowen and Alex Tabarrok, which is worth a read in full, even if it is just for their comparison of Tinder to third party escrow!

3. Stellar Consensus Protocol released

Stanford professor and Stellar Chief Scientist David Mazieres released his white paper on the revamped Stellar Consensus Protocol (nice TL;DR article here). This new approach called "federated byzantine agreement" claims to overcome the flaw found in the Stellar/Ripple consensus last December while also enabling network scaling. Not many have gotten their head around this federated trust model, including your humble author, so we will leave the debates for now to the more egg headed among us.