An anonymous cryptographer, who refers to himself as Harry Potter's mortal enemy Voldemort, may have come up with a solution to address two of the biggest issues holding bitcoin back—and the community is beginning to take it seriously.
Earlier this month, bitcoin once again made headlines when the Bitfinex exchange announced that it had been hacked and that almost 120,000 bitcoins (worth $70 million at today's price) had been stolen. At the same time as this news was breaking, someone was anonymously posting a white paper online which described a new cryptographic protocol with the potential to revolutionize the way bitcoin works, making it more scalable and private.
The proposal was called Mimblewimble.
"I call my creation Mimblewimble because it is used to prevent the blockchain from talking about all user's information," the author states in the white paper. The title references a spell from J K Rowling's books that "binds the target's tongue to keep him or her from talking about a specific subject," according to the Harry Potter wiki. News of the white paper was reported by Bitcoin Magazine and CoinDesk.
One of the most enduring stories surrounding bitcoin is the mystery surrounding its anonymous creator (or creators), known only by the pseudonym Satoshi Nakamoto—and the author of the latest bitcoin-related paper appears to want to continue that tradition of anonymity.
The author refers to himself as "Tom Elvis Jedusor," which is Voldemort's real name in the French edition of the Harry Potter novels (equivalent to the English "Tom Marvolo Riddle"). The author posted the white paper using a Tor hidden service website which has since been taken down, making it virtually impossible to find out where the site was hosted or by whom.
"I think that's it, the trail is cold," Andrew Poelstra told Motherboard.
Poelstra, a mathematician with Blockstream, a company focused on expanding the bitcoin code base for commercial use, was among the first people within the bitcoin community to see past the Harry Potter references and find potential in the Mimblewimble proposal.
Poelstra says that the identity of the author is irrelevant and the focus now among researchers and developers is on the proposals contained in the Mimblewimble white paper and how they could impact the development of bitcoin.
One issue the white paper takes on is scaling. In many ways bitcoin has become a victim of its own success, with the volume of transactions taking place too high for the current system to handle. This has led to internal struggles within the community and an inability to find a cohesive roadmap which most can agree on for how to move bitcoin out of the highly technical and complex world it occupies today and into mainstream usage.
Because of the one megabyte size limit of the blocks on the bitcoin blockchain, at best the platform can currently process just seven transactions per second. To put this in context, Visa says its payment system processes 2,000 transactions per second on average and can handle up to 56,000 transactions per second if needed.
To reach Visa-levels of payments processing—and make bitcoin a more viable alternative currency—the way bitcoin works needs to change. To date, the community has focused on several options to improve scalability, including scaling proposals called Segregated Witness and the Lighting Network, or a possible hard fork to bump the block size up to 2MB.
Mimblewimble offers another possible solution.
Mimblewimble could potentially be a solution to handling a huge amount of simple bitcoin transactions
"Mimblewimble describes an alternate blockchain, something that could be a sidechain or altcoin, that supports a different transaction type from bitcoin, and this is a very simple, not very expressive transaction type that basically allows one person or a collection of people to send money to another person or a collection of people," Poelstra said.
Because so much of the traffic on the bitcoin blockchain sees one person sending bitcoin from one wallet to another, without the need for additional features, Mimblewimble could potentially be a solution to handling a huge amount of simple bitcoin transactions.
Today, if you want to verify a bitcoin payment, you need to download the entire bitcoin blockchain (currently around 80GB of data) and then verify that everything checks out. With Mimblewimble, the size of the blockchain would be dramatically reduced because, thanks to clever mathematics, you could shrink multiple transactions into a single transaction, with the intervening steps compressed to almost nothing.
As well as proposing to solve scalability, Mimblewimble promises to make bitcoin transactions private.
One of the features of bitcoin which is regularly espoused by proponents of the cryptocurrency is the fact that it is anonymous, because bitcoins are transferred from one wallet to another without any other identifying information needed. However, because the bitcoin blockchain is a public ledger, every single transaction can be seen by everyone, meaning that purchases can be tracked.
Mimblewimble proposes to solve this problem by bundling multiple transactions together so that individual purchases or transfers cannot be so easily tracked.
"I'm fairly confident that it works, but I haven't taken the time to write this down that would convince anyone else, so there is still a lot of work on the academic side"
While this all sounds great, there is one big stumbling block to incorporating Mimblewimble with bitcoin: It is incompatible with the current bitcoin code base.
"It would be very difficult to integrate into bitcoin, or at least in a bitcoin as we know it," Pieter Wuille, a bitcoin developer and co-founder of Blockstream, told Motherboard. "There are no addresses, no balances, no scripts, so I expect we certainly can't integrate it in any way except as a separate experimental 'area.'"
This is a view backed by Chris Burniske, blockchain products lead at Ark Invest. "While the approaches of Mimblewimble are novel, there are other solutions to both privacy and scalability problems already, and Mimblewimble appears too much a deviation from the Bitcoin protocol to be implemented in the core software," he said.
So how could this improve bitcoin?
"The way that this would help bitcoin is if it were implemented as a sidechain so people could move bitcoin into Mimblewimble, use them, and when they need to do something that Mimblewimble doesn't support, they move it back to bitcoin or to another chain that would support those features," Poelstra said.
In this scenario, Mimblewimble would work alongside bitcoin, as a sort of low-functionality, high-scalability, high-privacy system for simpler transactions which would be appealing to many business who want to use bitcoin but resist due to the length of time it currently takes to verify transactions.
Polestra is currently working on Mimblewimble and will present a talk about it at the Scaling Bitcoin conference in Milan next month. He expects that, if it works the way it says it will, there will be significant interest in it from developers, and Polestra is confident Mimblewimble does what it promises.
"I'm fairly confident that it works, but I haven't taken the time to write this down that would convince anyone else, so there is still a lot of work on the academic side," he said. Polestra estimated that it would take between six to 12 months to write the code for Mimblewimble if a couple of developers took on the project.
As for the identity of the Voldemort, Poelstra said that the number of people who had the technical knowledge to write this paper would number at least 1,000, making it difficult to even make a shortlist of potential authors.
One of those who has been suggested as the author of Mimblewimble is Greg Maxwell, a colleague of Poelstra at Blockstream. This was based on the fact that Maxwell's work is referenced in the white paper and that Maxwell has previously been known to comment about Harry Potter. However, Maxwell responded to a Reddit thread suggesting him as the mysterious author, simply saying: "Though it may come to a surprise to [this subreddit], I am not Lord Voldemort."
Whoever its author is, Mimblewimble is a smart proposition, but just like all the other proposals for fixing bitcoin, it will need consensus before becoming really useful.