Blockchains are the wrong solution to data security problems, says MaidSafe

'Blockchains are terrible as a mass storage containers, so data still needs to be stored somewhere else'

The hype surrounding new technology paradigms frequently runs far ahead of reality. In blockchain's case, though, this phenomenon is even more pronounced than usual.

Already inflated expectations have been pumped up even further by the meteoric rise in the value of bitcoin and related cryptocurrencies, meaning that everyone wants a piece of the action. We see blockchain and blockchain-based applications being shoehorned into ever more improbable projects and visions, many with their own money-making crypto tokens attached.

In many instances proposed uses for blockchain are simply a case of picking the wrong tool for the job, just because it's shiny. Blockchain was originally designed with just one goal in mind: to support an electronic medium of exchange as an alternative to fiat currencies issued by a banking system that failed so badly in 2008, something at which it has, all things considered, been remarkably successful.

"In my opinion the killer app of a blockchain is bitcoin," said David Irvine CEO of decentralised networking firm MaidSafe.

Irvine continued: "Blockchain is a terrific and paradigm-changing design for a public transaction ledger, and for an open digital currency that is a fantastic fit."

Ask it to perform functions that are outside of its original purpose, though, and the fit is more square peg / round hole. Irvine describes many of the propositions being put forward as "pure marketing", especially those that advocate using blockchains as the basis for securing data.

Blockchains are an extremely secure way of maintaining an immutable record of transactions. At the same time, though, they make for lousy databases, only designed for storing very small files and with read/write performance that is glacially slow. This means that they are only suitable for storing pointers to data that must be stored in traditional databases or other media.

"Blockchains are terrible as a mass storage containers, so data still needs to be stored somewhere else, and that ‘somewhere else' still needs to be secured," Irvine explained.

The danger is that rather than making data more secure, blockchains simply add an additional layer of complexity.

There are other obstacles with blockchains to be overcome before the technology can be seriously considered for commercial use. There are the well-documented scalability issues that have already led to forking of the bitcoin blockchain and a centralisation of decision-making authority.

In addition, securing the blockchain by ‘mining' requires a colossal amount of power - a single bitcoin transaction burns enough electricity to power an average US home for a day, many times more than used by a ‘standard' digital transaction.

"This design comes at a significant cost of wasted energy as it uses a proof-of-work algorithm," Irvine said. "That should not be the case in future iterations."

Current blockchain technology, then, should be seen very much as Version 1.0. Fortunately there is much work underway to rectify its shortcomings. Irvine singles out the cryptocurrency Zcash as a promising development with its ‘zk-snarks' innovation tackling some of the practical inflexibilities of blockchains.

MaidSafe itself has taken a different track to decentralised networking with its SAFE Network, eschewing blockchains entirely. Instead it's an autonomous data network designed to store and transfer real data rather than pointers. Its architecture based on XOR networking, Irvine claims, avoids many of the issues faced by blockchains, making it potentially more suitable for supporting a broad range of use cases including IoT and robotics where data security is paramount.

"An autonomous data network removes all intermediaries and the flaws that negatively impact the management of data," he said.

MaidSafe released an Alpha 2 version of the software for public usability testing and new APIs for app developers last week.