One relatively overlooked application of the blockchain is that of data storage. This is odd, given that the technology is all about data: moving it, sharing it and keeping it whole. So why the lapse of coverage?
It may just be my blinkers – data storage is not the most sexy of functions. But, having dug into it a bit over the past few days, I’m realizing how varied and important the solutions are. It’s not just the need for safe and reliable data to value just about anything we do… It’s also the increasing role that data plays in our lives. If data is the new oil, and the big data enthusiasts proclaim, then we really need to think seriously about how we handle and store it.
Here’s a brief overview of the main blockchain innovators in the data storage space:
The group that has been at it the longest is Scotland-based Maidsafe. It has been working on decentralized computing for over 10 years now, and it doesn’t rely on a blockchain for data storage (it has built a distributed alternative which, to my non-expert ears, sounds a lot less streamlined). It does, however, rely on a native cryptocurrency for incentives – MaidSafeCoin is now the 10th largest in terms of market cap. Users contribute unused computing capacity in exchange for the native cryptocurrency, which can be sold or used to pay for network services.
Instead of information being uploaded to a central server, it is broken up, encrypted and stored on random computers across the network – several times each, to ensure availability. Data is moved around the network as computers are turned on and off and as demand ebbs and flows from certain areas, to improve access, speed and security (information is difficult to hack if you don’t know where it’s going to be).
Maidsafe aims to do more than offer distributed storage – it wants to establish a “decentralized internet”, in which surveillance and data theft are impossible.
After a murky ICO in 2014, Maidsafe’s second funding round was a £1.3m equity crowdfunding October 2016 (lower than the target of £1.75m). The alpha network was released in August of 2016, and has encouraged an ecosystem of apps includes data storage, email, forums and video conferencing, with more to come. One drawback to this project is the long development time. It’s an ambitious goal, true, but over 10 years’ buildup – especially with distributed technology evolving as fast as it is – runs the risk of being obsolete before it starts. Not to mention the risk that other, more agile competitors that started later with a different tech base can overtake on the inside…
Such as Sia, for example. Rather than try to rebuild the internet from scratch, Sia is starting with data storage. Like Maidsafe, it aims to harness unused computer space to offer a low-cost, decentralized alternative to Amazon’s S3 and the like.
To that end the Boston-based company built a proprietary blockchain that uses smart contracts to handle the payment from the user to the space contributor. Payments are made in the native cryptocurrency Siacoin, currently 40th in terms of market cap.
As with Maidsafe, information is split up, encrypted and distributed, retrievable only with the user’s key.
After an initial crowdfunded round of $500,000 in early 2014, the first beta prototype was launched in 2015, with version 1.0 following in June 2016. Sia’s parent company raised a further $750,000 in September 2016, from VCs Raptor Group and Procyon Ventures.
Storj, based in the US, is another startup going after the enterprise storage market. It started out on the bitcoin blockchain (with transactions occurring off-chain), although it recently announced its intention to migrate to ethereum.
As with Sia, participants receive a native cryptocurrency (in this case, Storjcoin, currently #35 in the market capitalization rankings) in exchange for offering their storage capacity. After an initial ICO in June 2014 of almost $500,000, Storj raised a further $3m in February of this year from angel investors, making it the best-funded startup in the data storage space.
Testing of the network began in 2014, continued through 2015 and in April 2016, Storj launched in beta and was added to Microsoft Azure.
A spectacularly ambitious project comes from the BigChainDB stable: its public network IPDB, which stands for Inter-Planetary Database (next stop = the universe!). The project aims to be the database for the “emerging decentralized world computer” (possibly that of Maidsafe, but more likely to be that of Golem, a project focused on harnessing unused computing power rather than storage space).
Rather than try to turn a blockchain into a database, BigChainDB approaches the problem from the other direction – by adding blockchain functionality onto database technology (this may sound like trying to fit a round peg into a square hole, and I’m not a database technician so no expertise here, but it could lead to a more adaptable and flexible solution).
The Germany-based startup has managed to raise £5m so far, including €3m in a Series A last September from Earlybird Venture Capital, Anthemis Group, Digital Currency Group and innogy SE, among others.
The barriers to these becoming mainstream are 1) regulation, and 2) reliability.
If data is sensitive (and it often is), then someone is going to want to regulate it. “Sharing” is simply not possible with certain types (health, fiscal, etc.), and even if encryption and protection assure that only those who should access it can, the chance that others could gain access is enough for regulators and potential clients to hesitate, or even downright object.
It’s a bit like the decision between keeping your gold spread around the world (where it is more vulnerable, but on a piecemeal scale), or in a vault 600m deep in a mountainside (where it’s harder to get to, but if the bad guys do…).
The reliability issue is also going to be a concern for the important stuff (like identity, finance and government documents). Even with layers of redundancies, is it enough? Sure, distributed risk is preferable in that system failure is less likely. But what about responsibility? If the distributed system fails, no-one takes the blame, but also no-one makes up the loss.
However, the idea is interesting, and could well be where the internet is heading. Once the technology has advanced further and speed and distribution issues are ironed out, could it replace our current siloed format of storing data, with ownership, rights and other fundamental concepts in the murky area of this-is-not-what-we-meant-by-equal-access? Will the benefits and redistribution of wealth offset the humongous costs of changing the way the current system works?
Time (sorry, I mean the market) will tell.