Cheat Sheet: Blockchain

Does blockchain have real value for businesses?

Blockchain? That’s the Bitcoin thing, right?

It’s one of the contributing technologies, yes. Specifically, the Bitcoin blockchain is a history of every Bitcoin ever spent, along with unique – although mostly anonymous – details of who paid it to whom. The clever part is that it’s not stored on a massive central server somewhere; the system relies on people voluntarily hosting blockchain data on their own systems, even if it doesn’t actually contain any of their own records. Because the data is massively distributed and duplicated, it’s very difficult for anyone to defraud the system, or shut it down.

Fair enough, but my business doesn’t use bitcoins – so why should we care?

Bitcoin is a good illustration of what blockchain does, but it’s not the only thing it can do. At its heart, blockchain is simply a general-purpose information store with an emphasis on verifiability. You could have a blockchain database of complex mechanical parts, or medical histories, or even radioactive nuclear power fuel rods. Any sort of data that involves transactions between multiple entities could benefit from the durability and transparency of blockchain.

So how do we get set up with one of these blockchains?

Various platforms are available that make up the mixture of servers, databases and interlinks; blockchain-oriented startups such as Chain ( chain.com) will be quite happy to put together a systems design that satisfies your need. However, just like every IT fad, the fact that it’s possible doesn’t mean that blockchain is necessarily the best solution for the problem you think you have.

But blockchain has real advantages over a regular database, right?

For some roles, yes. But block chain is a complicated thing; it’ s a database, a protocol, an encryption method and a ledger all in one, and its distributed design adds complexity to the implementation as well. It can work very well when it’s used by a large community for storing data that wants to be publicly verifiable. But if you’re looking to store a small amount of short-lived internal data, it’s almost certainly not the right choice. A little thought at the outset of your project should make it abundantly apparent whether blockchain is a good fit, or whether you should stick with MySQL.

It sounds like once we set up a blockchain it will just keep growing forever. Won’t it become unmanageable?

Blockchain does indeed demand that you maintain a complete history of old transactions. But those old transactions don’t change or grow at all – the vast majority of blockchain accesses involve very recent transactions. The old stuff can be archived against the rare eventuality that you need to retrace a series of events going back several months or years.

That said, the storage issue is real. Expect to see a furious burst of development as the blockchain concept grows in popularity: the demand will surely grow for ways to store and retrieve historical blockchain records at very low cost.

What if we want the verifiability of blockchain, but without making our data public?

You might be trying to square a circle here. Putting confidential data into a platform built on widespread public participation is a recipe for trouble. Still, you wouldn’t be the only one: there are big ambitions for blockchain 2.0.

Just keep in mind the nomenclature used by the Bitcoin architects. Fundamentally, we’re talking about a ledger of transactions, of the sort that accountants have used for centuries. That’s perfect for Bitcoin – but your level-headed, old-school database guys should realise that a blockchain need not be your sole repository of information. Part of your upfront analysis should be about the split between public and private – and for private data other database models are likely to win.

So what should we say when a consultant comes along offering to solve all our problems with blockchain technologies?

It’s tempting to embrace the current fashionable technology as a cure-all, but insist on a very careful appraisal and decision process before you do anything at all with a blockchain proposal. That’s especially so if you’re participating in the delivery – most blockchain projects will just be showing you a web or smartphone front-end to a cloud deployment. If a consultant thinks that you should have your own development team and be feeding back into the GitHub-forking whatevers, it’s probably time to walk away.

“Blockchain can work very well when it’s used by a large community for storing data that wants to be publicly verifiable”