Blockchain

JC pontificates about technology An occasional series. →Go uptown Technology Tell me more Sign up for our newsletter — or just get in touch: for ½ a weekly 🍺 you get to consult JC. Ask about it here.

A blockchain is a distributed record of information — transactions, contracts, whatever — stored across a network. Each “block” of information gets its own cryptographic code which is posted to every node on the network. Thus, you can’t futz with any information on the blockchain unless you can futz with the whole network, which is like boiling the ocean. Once written, the blocks are, therefore, effectively permanent.

Blockchain allows parties who don’t trust each other to transact in confidence. What’s done is done — what’s on the blockchain cannot be reversed. There’s a permanent record. No one controls it: it’s truly anarchic, like: no government. No mendacious middlemen like banks. just pure, untrammelled laissez-faire.

Thus, bitcoin — a currency without the backing of anyone. It just bootstraps itself into existence like a skyhook. This makes loads of sense to a tech guy. It makes none to a banker. (But you would say that, banker dude).

Architectural conundrums

David Rosenthal’s blog has a wealth of learning on limitations of blockchain architecture.

Generally software is designed to go as fast as fast as possible. But, problem: if such software is hacked, it performs maliciously as fast as possible, too. The Bitcoin protocol avoids this problem, ensuring resiliences through decentralised consensus and proof-of-work which makes it mathematically too much work to subvert the network.

Decentralisation (theoretical and practical — in that the network must not only be architected to be able to have many participants, but it must actually have many participants) is necessary but not sufficient for that resilience. BTC is not practically decentralised. There are only a handful of significant miners.

Permissioned blockchains have a central authority controlling which how blocks can be added to the chain. Because they don't require period is work they have a much lower carbon footprint & are much more efficient. But they are thus centralised and require confidence in the central authority. If it is hacked, you're in trouble.

Permissionless blockchains (like Bitcoin’s and Ether’s) do not have a central authority, but just still defend against “Sybil” attacks, where an attacker gains controls of a majority of ostensibly independent participants. Hence the practical need for many participants. Permissionless blockchains achieve this defence by making the reward for a successful attack uneconomical: it costs more to mount an attack than stand toyou gain by making it.

Therefore: participating in a permissionless blockchain has to be expensive. Inefficiency and slowness is a design principle.

To encourage anyone to be a participant (i.e., a miner) and so proof of work, you must reward them. There being (by design) no central authority who can raise funds from users to distribute to miners, the blockchain must generate renumeration by itself within its system. It creates its own transferable tokens of value on the blockchain. This is a cryptocurrency. Therefore, a permissionless blockchain needs its own cryptocurrency to be secure.

A cryptocurrency at its moment of creation, does not have intrinsic value, of course. It can only acquire that by people attaching value to it. This is a magical bootstrapping process. But it must acquire intrinsic value, for it to be worth the miners’ while doing the proof of work to earn it.

The miners will want to convert this crypto token for fiat currency: cold, hard, folding green stuff.

To convert crypto to fiat, someone must be prepared to buy the crypto. Only those who believe the crypto currency will hold/increase value vs fiat over time will do that.

Therefore a permissionless blockchain needs bullish speculators to work. It needs people with an unfathomable confidence in an invented currency. This is not a curious and inexplicable phenomenon: it is fundamental to the very architecture of blockchain. Cryptomania is a designed principle of permissionless blockchain.

Now you could make an argument this is true of any paradigm or power structure. All depend on “the punters” earnestly believing; while those in authority or operating infrastructure or institutions of the paradigm cannot do anything to destroy the illusion, they needn't (and often don’t) literally believe it themselves. They may do, but they don't need to.

This is true of faith, and politics and is increasingly true of modern morality. It is definitely true in markets: market makers don't care whether things are going up or down, as long as market users have these views. There a number of traditional firms who have piled into crypto because it is opaque, unregulated, full of suckers, and opportunities for free arbitrage are (therefore) immense. They don't need to believe in crypto at all, as long as someone else does.

Pace layering

Now given crypto emerged as an ultra-libertarian solution for operating in Hobbesian environments of zero trust, its need for the marks to have a cargo-cult faith in its value in order to work is kind of ironic can we say?

But there is trust and there is trust.iterally, and they operate at different layersin Stewart Brand’s pace layering schema. The layering point is important to understand the level at which the argument works.

The “trustless society” bit is a gimmick - a fad - playing to the present libertarian ultra-dystopian fashion of not trusting intermediaries, experts or institutions, even though through history our the institutions have proven imperfect, but largely trustworthy (in that they’ve got us this far) and better than any of their utopian alternatives. The requirement for “punter faith” in the value of cryptocurrency is much deeper, as it is part of the engineering, being the only way to incentivise miners to do proof of work.

So how — ?

Because some folks got a bit giddy—and some techno-Unabomber types, but they were kind of giddy in the first place. Blythe Masters — she who invented the credit default swap — thought it was so profound that she joined a startup in 2015. And folks listen to Blythe — why wouldn’t you?

The FT reports that as long ago as 2016 Gartner put blockchain near the top of its “peak inflated expectations” curve.

Is it — you know ...?

Snake oil? In a nutshell, yes.

You’ll see a lot of hyperbolic nonsense about blockchain, and this Medium article by Kai Stinchcombe is a pretty good antidote. In short – the benefits of blockchain are mainly predicated on idealistic values and dystopian paranoias:

• That you can’t trust anyone
• That all intermediaries do is rip everyone off and gum up the system (and you can’t trust them either)

Blockchain solves those problems but, by design, creates a wodge of other ones:

• It’s expensive, it’s slow and it consumes a ton of energy;
• Transactions are permanent and irrevocable
• There’s no mechanism to correct errors
• There’s no — ahem — trusted intermediary who can dab the breaks, nudge the steering wheel, pursue fraudsters or help enforce contracts (purist blockchain ideology assumes contracts can be and users will want them to be enforced “by code”). The bank of Canada put it succinctly: at “its heart, there exists a fundamental inconsistency or tension between a centralised wholesale interbank payment system, as we have now, and the decentralisation inherent in the DLT”
• There’s no value that a trusted intermediary can add to the system: it’s caveat emptor, user pays, entirely at owner’s risk. A uber-libertarian's wet dream, in other words. Great if you're holed up in a cabin in the woods with ammo and canned spam, but a fairly Spartan outlook for those of us who are prepared to live with a bit of benign intrusion so our GPS works okay. In other words, pure blockchain demands a trade-off most pragmatic people aren’t prepared to make.

Now if you adulterate blockchain to correct for its nasty, brutish shortness — if, in other words, you overlay trusted intermediaries, fail-safes, benign enforcement and monitoring — then:

• you’ve imported back in all the problems you thought you were trying to solve; and
• existing systems already do all of this stuff more quickly, reliably, cheaply and at greater volume than a blockchain could reasonably handle.

The most fun and, frankly, brazen attempt to correct these shortcomings by undermining the whole idea completely is surely blockchain as a service. No, I'm not making that up.

So — again, whither the use case, dudes?

STOP PRESS + + + STOP PRESS + + + STOP PRESS + + +

We have a new report of a valid usecase, by that super-innovative futurologist Lex Luthor at Luthor Capital [it’s Greensill isn’t it? — Ed]:

[Editor’s note: unfortunately Oracle has pulled its video “the future of working capital management?”, by that working capital supremo Lex Greensill off their site for some reason.]

The sovereign individual

It's not all bad: the underlying model, like the internet - provides for a radical new way of organising affairs where data is not centralised, but kept at the edges and in the hands of the users. If the user has her own digital wallet with credentialised data, this can be passed, encrypted, only where and when required, to intemediaries. The intermediaries wouldn't be able to keep it, or harvest it, or sell it, and it would be less susceptible to being hacked (instead of hacking VISA's website and getting personal data of 400 million people in one go, you'd need to hack 400m people's individual wallets. Not impossible to hack a wallet but, but the value is in the aggregated data, not individual files, so the incentives are wildly out of whack.

It also passes the responsibility for keeping that data up-to-date to its owner, and it only has to be done once.

Obstacles:

• Firstly, intermediaries who buy and sell your data will HATE this, since them collecting your data is extremely valuable to them. Don't expect to see Google or Amazon or Facebook promoting digital wallets any time soon. BUT for the person who does, and makes the case, a monopoly busting disruptive business model awaits.
• Secondly, which digital wallet? What if there is more than one? what if you lose your key? What will the poor meatsack do then? Does anyone even care about the meatsack any more?
• Thirdly, and what does this say about the distrinction between our online persona and our fleshly existence? Is this first step to synthesising our mortal frames into a virtual existence? Is this the trick we're missing? The robots are not going to turn into us, but we are going to turn into the machines?

Oh the irony.

Is it the same as distributed ledger technology?

Because DLT is okay, right?

Maybe. According to the excellent Deloittes report, think of DLT as the generic problem to be solved; a blockchain is a way of solving it; the blockchain is the particular blockchain underpinning bitcoin.

And bitcoin? Bitcoin is catnip for stupids.

Further reading

If you want the case for blockchain, just google blockchain or even talk to any poseur or charlatan in any industry who is trying to sell you something. They’ll talk your ear off.

If you want to hear the case against, there's a bit more work to do. So the jolly contrarian is gathering some links for you.

• The daddy: Kai Stinchcombe's Medium takedown.
• Deloitte’s report: Bitcoin, Blockchain, and Distributed Ledgers
• The atlantic: Derek Thompson in the Atlantic - Blockchain silly season: “Bitcoin might be where Pets.com was in 2000—a technological curiosity in search of an enduring business need. But blockchain is not where the internet was in 2000. Even blockchain’s biggest defenders can’t say what the technology’s most obvious consumer use-cases are going to be, because they plainly don’t exist yet. It is possible they never will.”
• Cointelegraph: Cointelegraph on how 92% of the blockchain-related projects on GitHub are now dead.
• Izzy Kaminska in the Pink ’Un: Everyone's getting skeptical about the blockchain hype.
• Stephen Norman: The emperor has no clothes.
• The Germans: Aaaaaand here's Jens Weiderman of the Bundesbank

See also

• Uniqueness