Cubbit vs The Blockchain(s): The Future of Decentralized Cloud Storage

Whether it is the new Google or the amazing solution for almost nothing, the blockchain is all over the news. Of course, this is not 2017. Back then, when the crypto market almost hit 0.8 trillion dollars, it felt like a new internet was going to materialize on a moment’s notice — a tsunami made up of hashes to wipe out the powers that be. But, word to the wise, what goes up must always come down. And this time was no different. Percentage-wise, the Great Crypto Crash of 2018 was larger than the Dot-com bust. Was it a fad then? It doesn’t seem so, as we’re still talking about it two years later — and not just in terms of digital antiques. Indeed, the blockchain is still making headlines because of its potential as a distributed protocol. Should it keep to its promise, new groundbreaking applications will be built. Think Uber, but without any board of directors holding the reins. The internet without middlemen. That’s what the story is all about.

And it’s a big one. Per the Market Watch industry report, we’re talking $25 billion by 2025. Only this time it’s not speculation. According to Deloitte’s 2020 Global Blockchain Survey, tech leaders look at Satoshi’s invention as “integral to organizational innovation”. Eleven years after Bitcoin’s inception, one could argue that this is a remarkable achievement. However, before getting down to the nitty-gritty, we must recognize that, even in cryptoverse, a schism is ongoing between those who downplay this blockchain-magic-dust thing and those who bend their knee to worship the savior. The question is whether the blockchain is just a ledger for bitcoin transactions, or rather, a world computer for all kinds of general-purpose applications. Simply put: can the blockchain truly run the world? And, by the way, isn’t it weird that a database infrastructure has come to be the hero of the tech world?

What exactly is the blockchain?

First things first, there is no such thing as “the Blockchain”. As of today, there are some hundreds (potentially thousands) of blockchains up and running. The exact number is impossible to know. Why so many? To fully grasp the nuts and bolts of these second and third-generation blockchains we need to start from the original.

Bitcoin.

The year was 2008, and the subprime crisis was in full rage. At the time, the hacker counterculture, best embodied by writings such as the cyphernomicon and the cypherpunk’s manifesto, was a coalescence of libertarians interested in privacy, anonymity, pseudonymity, and digital cash. The cypherpunk mailing list was already 16 years old. Topics ranged from mathematics and cryptography to economics and cryptoanarchy. It was on this mailing list that, on October 31st, a person or a group of people under the name Satoshi Nakamoto linked a white paper entitled “Bitcoin: A Peer-to-Peer Electronic Cash System”.

Digital cash, however, was nothing new. Contrary to popular belief, cypherpunks were trying to build it long before Bitcoin. Since 1983 at least, when David Chaum began to explore the idea of anonymous transactions to “make Big Brother obsolete”. But David Chaum’s proposal failed spectacularly. It needed a centralized, bank-like server to function, and no bank was up for signing on. Then came Bitcoin. Unlike Chaum’s Ecash, Bitcoin solved the double-spending problem without any centralized server.

What is Bitcoin then? It is a distributed, public, open-source, byzantine fault-tolerant ledger that records bitcoin transactions.

How Bitcoin works

Suppose you live with a bunch of strangers and you want to make a ledger with your housemates of who-owes-who. If anyone can add lines to the ledger, who prevents Bob from forging fraudulent transactions stating that Alice owes him money? Cryptography to the rescue: to ensure that every transaction is legit, you establish that every sender must generate a public key / private key pair and signs each personal transaction with their private key. Great! This way, you can be sure that nobody can charge anyone illegitimately. Yet this is not enough to make the ledger trustworthy. In fact, there are still a couple of problems to solve.

For one, where is this ledger? Sure, you could simply host it on a website, but that would be problematic if many different strangers were involved, or if Bob were really sketchy. That is: you can trust the website owner because you know him and little money is involved, but with strangers, things get complicated very fast. You better be sure the owner is trustworthy! And even if they are, they could always get hacked and voila! Your money is gone.

Also, what if someone can’t settle their own debts? To overcome this issue, you might establish that anyone can only sign transactions with money they actually own. The thing is, to be sure about that you need to know the entire history of transactions. To do so, an easy solution would be to broadcast transactions. This, unfortunately, is insufficient: without an order, many ledgers would be created, and there would be no way to prove that Bob is not spending your money.

Enter Bitcoin.

Bitcoin establishes a few simple rules to deal with these issues. For starters, every participant holds a copy of the ledger. As a matter of facts, the Bitcoin ledger — namely, the Bitcoin blockchain — is distributed across a network of thousands of computers worldwide. To make sure that everyone agrees on the ledger and reaches “consensus”, transactions are packed in blocks, each one starting with the hash of the previous block to create a tamper-proof chain of blocks so that, if anyone changed a previous block, the whole chain would change completely as a result. In case of conflicting ledgers, only the ledger with the most computational work is to be considered valid. Wait, what “work” are we talking about? And how are new blocks created?

Basically, it works like a lottery. Participants in the lottery — those who add new blocks to the chain — are called miners. To do so they apply SHA256, i.e. Bitcoin’s cryptographic hash function, on a list of yet-to-be-published transactions followed by a magic number called nonce. Their objective is to find an output starting with a number of zeroes that is smaller than Bitcoin’s current difficulty target, which is dynamically adjusted so as to ensure that exactly one block is added every ten minutes. In exchange for solving a block, the miner can add a very special transaction on top of it, assigning themselves a very specific number of bitcoins out of thin air — the so-called “block reward”. This, in a nutshell, is the whole secret of Bitcoin wizardry.

You might have heard that miners use powerful computers to solve difficult equations. Nonsense. There is no equation to be solved. If anything, Bitcoin is based on the fact that the only way to find the nonce is by trial and error. In a way, Bitcoin mining rigs are efficient random number generators.

This may sound simple, but it is by no means an easy task, as all miners are in a race against one another. All this computational work, let’s say it out loud, is not for nothing. On the contrary, it serves to protect Bitcoin’s blockchain from attackers, since the length of the difficulty target is directly proportional to the hash power in the blockchain. In fact, to steal money from a published transaction, you need to find another nonce that solves the block where the transaction is contained and, after that, you need to solve all blocks following that specific block — all this before anyone else adds a new block. Long story short: either you are faster than the rest of the network or you better follow the rules.

Blockchain-based storage solutions: how do they work?

For starters, they do not use a blockchain to store the users’ data. That would be a monumental waste of space! Think about it. The Bitcoin blockchain weighs a little more than 300GB. Imagine if it didn’t store just bitcoin transactions, but all kinds of data, no strings attached. How big would that be? As of right now, the capacity of so-called decentralized storage solutions amounts to a few petabytes. Unlike Bitcoin, though, these platforms do not require every node of the network to store the data. On the contrary, they encrypt the data and employ Reed-Solomon erasure coding, so that just a small number of nodes are necessary. To store every user’s data on the blockchain would require who knows how many petabytes. Basically, it would be madness.

So, how do they work? The architecture varies from blockchain to blockchain, but other than a few minor differences, they work pretty much the same as any other zero-knowledge end-to-end encrypted cloud storage platform. In short, they employ the standard combo: a distributed hash table + Reed-Solomon erasure coding + some kind of repair process to guarantee uptime in case the number of hosting peers that have gone offline goes under a certain threshold. Yet, who is in charge of triggering the repair process?

In the case of StorJ, which is likely the most popular and market-ready self-styled decentralized storage solution, the burden of the process falls on a special class of peers, called Satellites — central coordinators whose job is to act as a trust boundary for payment, audit, file repair, and metadata management. But don’t be fooled by the plural words — coordinators. All Satellites are run by StorJ Labs. Other than the one run by Stefan Benten, which is StorJ Lab’s Strategy R&D Engineer, there only is one coordinator in StorJ, i.e. StorJ Labs, a central coordinator through and through. By StorJ Labs’ own admission, “the purists in the community (and in Storj as well), will correctly point out that we’re not 100% decentralized with regard to Satellites. While Satellites are highly distributed and highly redundant, they’re not fully decentralized within the Tardigrade network.” In StorJ’s defense, the code to run a Satellite has been open-sourced by StorJ, so theoretically any third party could do it. Yet no one does.

Let’s be clear: we’ve used StorJ as an example, but the company is far from being the only one claiming to offer a decentralized storage solution when in fact it offers a distributed storage solution. So, what is the blockchain for, then? Two things, among others: for transactions, and as a proof of capacity. For all blockchain storage solutions, though, the blockchain’s primary use case was to raise money: during StorJ’s ICO in 2017, StorJ raised $30M. Which, by the way, is not even the highest amount raised by a blockchain cloud storage platform: Filecoin raised $205M without even producing a mainnet 3 years past the token sale. And the list goes on. Ultimately, one wonders if exploiting the cryptocurrency mania was not the true business model of these companies.

What’s the difference with Cubbit then?

There are two main differences between Cubbit and blockchain-based storage solutions.

First, we do not claim to be decentralized. Although our long-term vision is to ultimately distribute the central coordinator across the network and become fully decentralized, we want to be honest with our users and call a spade a spade. At this stage, Cubbit is a distributed, zero-knowledge, end-to-end encrypted cloud storage solution. No more, no less.

Wondering what the practical benefits are? Check out Cloudstorageinfo’s review of Cubbit!

Second, blockchain solutions revolve around a marketplace, while Cubbit is free from monthly fees. How is that possible? The thing is, “decentralized” storage applications divide their users into two categories: customers and storage providers. Cubbit, on the other hand, has only one type of user, as each node is both a user and a storage provider. In fact, Cubbit nodes do not pay anything to Cubbit other than a one-time payment for the Cubbit Cell. As long as they contribute storage to the network, they can use it free of charge. To put it differently, blockchain solutions use money to incentivize storage providers to stay online, while Cubbit’s built-in incentive is the use of Cubbit itself.

In actual fact, we’ve designed this network to have an intrinsic surplus of storage. Each Cubbit Cell comes with double the physical capacity of what is made available to the user. The Cubbit Cell 1TB, for example, has 2TB of physical storage inside, half of which is available to the user as cloud space. Part of the rest (25%) is used for the redundancy, that is to say, to ensure that every file uploaded on Cubbit is automatically backed up inside the network and always accessible to its owner, perfectly safe and reliable no matter what. Our business model consists in leveraging the remaining 25% of storage to provide advanced web services to businesses or professional users, so as to cover the costs and be able to maintain the network free of charge, even in the long run.

Why, you may ask? Two reasons: stability and privacy. As regards the former, the more storage providers there are, the more secure the network is. As of today, we have ~1500 nodes for a total of~ 1.4 M files, and since we’ve started shipping, uptime has always been guaranteed.

As regards the latter, though, we may sound naive but our vision is that of a new internet — one where we won’t be at the mercy of storage providers. Privacy, we believe, is a universal human right and should not be held hostage behind a paywall.

For more information about Cubbit, visit our website.