What’s It Good For? | Etherean.org

The answer is not “absolutely nothing,” nor is it “all the things.” Like most technologies, blockchain is very good at a few things and pretty bad at everything else. It’s time to figure out where it fits into the big picture.

In the beginning, Bitcoin was touted as a better kind of money: peer to peer electronic cash. Today, the narrative has narrowed to digital gold, i.e., only a store of value. Ethereum, in the beginning, had an even more ambitious vision: a decentralized world computer. Today Ethereum has one main use case: decentralized finance, better known as “DeFi.”


This is a companion discussion topic for the original entry at https://www.etherean.org/blockchain/web3/software/2020/07/25/whats-it-good-for.html
1 Like

Assertions that are limited to public permissionless cases and aren’t actually generizeable to “blockchains” broadly:

Let’s start with what blockchain is good for: distributed consensus over a single, global ledger, with permissionless, ordered transactions that update that ledger.

Blockchains are also good for coordinating networks of private ledgers (private channels in Hyperledger and Symbiont). They are also good for permissioned transactions.

I have yet to see a use case outside of money, digital assets, or identity that really belongs on blockchain.

Broadly defined, that’s practically everything on the Internet. Name me something from your list of “stuff we should do not on blockchain” that someone couldn’t shoe-horn into “money” or “digital assets” or “identity” when those three categories are allowed to be broadly defined. I’m scanning the list of blockchain application areas from “Enterprise Blockchain has Arrived,” I’m looking at the list of industries from Tapscott’s “Blockchain Revolution,” and I’m looking at the list of SIC codes, and I don’t see a single entry that couldn’t fit into your “narrow” list of three categories.

Seriously, name me an SIC Code or any realm of human endeavor that I can’t tag at least partly with one of those three. I’m not finding one.

In practice, users don’t want to spend three dollars in gas fees each time they vote on which pizza toppings to order before a meeting, and each time they disburse funds to pay for it. There’s also the fact that moving governance onto the blockchain invariably results in plutocracy.

Here again, your arguments are limited to the public permissionless space. I could set up a public permissioned or private permissioned blockchain system that didn’t have plutocracy and didn’t require expensive gas fees to run.

Of the three essential primitives required to build applications—compute, storage, and messaging—blockchain is really only designed for compute, and then only slow, expensive compute.

Not seeing this in the permissioned space. Hedera Hashgraph claims 10,000 TPS. EOS mainnet demonstrates 4,000. Symbiont is fast. Let those be mainnets that coordinate side channels, and the total throughput is massive. On-the-horizon DLT (not technically “blockchain” but definitely distributed ledgers) projects like Radix are even faster.

It gives us self-sovereign identity

Separate point - I’m not seeing this in practice on ANY blockchain/DLT. It’s a common claim but unsupported.

This is most of what I wanted to share – happy to dig in.

1 Like

Thanks for sharing your thoughts here. I’m excited to dig in :slight_smile: I’ll respond to several of your points first, then come back to the question of permissioned chains.

I think about it this way: what were the very first things that personal computers (and, later, devices like smartphones) used for? Things like messaging (chat), email, word processing, editing and sharing photos, contacts and calendars, spreadsheets, etc. Then, of course, there’s also gaming.

I’m clearly thinking bigger than blockchain here–more along the lines of Web 3, or of a new compute stack, which I wrote more about in Faster Horses, Better Software–but these are the sorts of applications that I think a mainstream audience would find appealing and useful. As one example, if you look at Urbit’s OS1, this is precisely the sort of thing it’s being built to do. Blockchain is not especially well-suited to any of these applications, and they have nothing to do with money or digital assets.

Actually, I don’t think we disagree. Blockchain is an important primitive that can be used to provide some of the infrastructure that applications like this will rely on–identity and namespace, most obviously–but my point is that claiming that blockchain is the most important component of Web 3 simply isn’t true, and isn’t borne out by the facts.

Out of curiosity, are you suggesting that such permissioned systems are or can be governed democratically? Are there any such examples in operation today? I see a very limited menu of options aside from technocracy, plutocracy, and (hypothetically) democracy.

I guess it depends what you mean by “self-sovereign identity.” One could argue that every Bitcoin/Ethereum/whatever keypair is a self-sovereign identity–and there are definitely a slew of dapps that you can log into using Metamask and such keypairs. Blockstack and Urbit are two other examples of working, useful self-sovereign identity systems. The fact that not many applications have been built around this novel infrastructure is sort of orthogonal IMHO.

Now, to return to the topic of permissioned chains:

I personally remain unconvinced that permissioned chains offer much more than centralized or federated databases–which are, of course, much faster and cheaper than public blockchains. This is probably a product of my own ignorance. I don’t mean this as a rhetorical question at all, but could you help me understand what the systems you mention offer over a private, centralized database, or a database federated among a consortium, or perhaps even more to the point, a product like QLDB?

Thanks!

Let’s not rely on my own meager knowledge – groups like World Economic Forum and McKinsey are articulating the business value of private and permissioned blockchain systems.

With QLDB (which I’ve never used) I’d have to buy in to a non-open-source product from a single vendor with a single hosting option, AWS. If I’m a CIO trying to improve the reconciliation of intercompany charges among airlines, or phone carriers, or hospitals and their providers, or anyone else dealing with sprawling data and money reconciliation issues, the benefits of a “single source of truth” that nobody (including AWS) can screw up too easily, should be obvious.

…more in next post…

1 Like

Chronicled’s Susanne Somerville spent 60 minutes explaining just two use cases in detail, which are obviously bad candidates for a public permissionless system, and she made it clear that her stakeholder companies were VERY leery of single vendor lock-in (so no QLDB for them).

Plus blockchain/DLT has good PR, and multi-party “single source of truth” systems have a strong need to be plausible and politically palatable. The relentless PR hype around blockchain the past five-plus years have made that the politically safer route than a single vendor “quantum ledger” that Amazon can screw up if/when it’s hacked.

Systems like VeChain are a hybrid of permissioned and permissionless - they are operated (as far as I can tell) as an easy onramp for vendors looking to ease the data interchanges between themselves and their key suppliers and customers.

When Grant Thornton built an intra-company EOSIO based blockchain backbone to tie together their various ERP systems, they certainly knew about other federated database options. They held a bake-off between EOSIO, Hyperledger Fabric, and Enterprise Ethereum. (Did I mention EOSIO won? Due to superior declarative security, faster block times, easier coding, replaceable keys, etc.) Ask them why. My guess is, bragging rights (we eat our own blockchain dogfood) plus better functionality plus no vendor lock-in when using Open Source software.

1 Like

Your points are well taken. Still, to me, this is more an argument in favor of avoiding vendor lockin (via strategies such as open source) than it is an argument in favor of any particular technology or platform.

It raises two interesting questions for me:

  1. What makes a blockchain a blockchain? As opposed to, say, a “database.” What’s the fundamental characteristic(s), qualities, traits, or feature set that a platform needs in order to have this moniker? I suspect there is no bright line and that it would quickly devolve into an argument over terms. (To me, personally, it involves things such as permissionlessness and censorship resistance, but I suspect you might argue otherwise.)

  2. What if there were a non-blockchain database platform that satisfied these needs–namely, no single administrator, no risk of vendor or data lockin, open source, etc.? Does such a thing exist?

Curious to hear your thoughts!