The Internet of Trusted Things
“What a piece of work is a man!” cries Hamlet as he laments Rosencrantz and Guildenstern’s savage double-cross. They morph from Hamlet’s childhood friends into sleazy middlemen, spying on behalf of King Claudius. The divide between man’s potential for trust and life’s double-dealing, rotten reality lays Hamlet low in depression.
Life is riddled with middle-men like Rosencrantz and Guildenstern, fronting a smiling, back-slapping facade while secretly plotting your demise, or at least angling for a payday. The middle-man is everywhere, all the time. He is your cashier at the grocery store, swapping money for produce; he is your broker when you rent an apartment, your salesman at the used car dealership.
The most energizing idea to blockchain entrepreneurs is eliminating middle-men, bridging trust gaps between transacting parties with cryptography, accelerating transactions, and slashing fees: like eliminating the $5 billion Western Union charges its customers to move $80 billion around every year.
Blockchain, as a trust machine, is revolutionary. It gives humanity a fresh start: undergirding a new, trustless world cleansed of our most depraved tendencies by enforcing contracts with mathematics.
But there is great irony in how blockchain-based assets trade-hands today; an irony whose mitigation is the basis for the dramatic growth in decentralized exchanges. Let me explain. Most cryptocurrencies trade hands today on centralized exchanges. These exchanges are true middle-men, hence the irony. They are black boxes into which trusting users send their coins, receiving electronic IOUs in return backed by nothing but the issuer’s reputation and name. These electronic IOUs are digital numbers in a database, traded cheaply, and at wicked-fast speeds. The cryptocurrency itself remains frozen in the exchange coffers, while the fictional account balances whizz back and forth instantly according to the user’s trading whims.
The annals of centralized exchanges are gruesome; their founders run the gamut from the merely inept to criminal masterminds. Two exchanges have particularly colorful pasts: Mt. GOX and QuadricaCX. Jed McCaleb, the founder of both Ripple and Stellar, originally founded Mt. GOX in 2010. But in 2011, before Mt. GOX made it big, McCaleb sold it off to Mark Karpeles, “Mr. Punchable” in blockchain parlance. Karpeles managed Mt. GOX out of Japan, but despite the exchange’s phenomenal success handling nearly 70 percent of all Bitcoin trades, Karpeles spent the majority of his time running a coffee shop. Between 2011 and 2013, 400,000 Bitcoin (about $450 million at the time—$5 billion today) disappeared, Karpeles blamed hackers. In early 2014, Mt. GOX filed for bankruptcy.
In the case of QuadricaCX, a Canadian exchange, every single cryptocurrency held by the exchange was controlled by one person: Its founder and CEO Gerald Cotten. Cotten’s autocratic rule began in 2016, after his co-founder left because he was implicated in… a money-laundering scandal. Birds of a feather flock together… What happened at QuadricaCX? In 2019, Cotten died. And it was discovered that prior to his death, Cotten had siphoned off millions of dollars while filling withdrawals with other client deposits (turning the exchange into a Ponzi scheme). In all, $215 million owed to 76,000 individual users disappeared into the depths of Cotten’s psychopathic pockets. But the story gets weirder.
Cotten “died” while on vacation in a part of India famous for its production of fake death certificates, and he wrote his will four days prior to departing on his trip. Cotten’s death certificate was written so sloppily it misspelled his name. I don’t know about you, but every time I get a tickle for a fake death certificate, I ask myself: “Did I take a wrong turn at some point in my life?” Naturally, many have speculated that Cotten did indeed “die” as opposed to die (no scare quotes), making off with the two hundred mil and living life in lavish anonymity.
In light of these atrocities, why do digital assets continue to trade chiefly on centralized exchanges? Because despite their philosophical elegance, early decentralized exchanges ask more questions than they answer. Etherdelta scaled after its 2016 launch by quickly listing the Ethereum tokens enjoying frothing public demand during the peak ICO period in 2017 and early 2018.
But users of Etherdelta wait on the Ethereum blockchain to confirm every trade, limit-order, deposit, and withdrawal. This blockchain-time-constraint introduces an unthinkable bottleneck for algorithmic traders and a serious nuisance even for unserious traders. IDEX came to prominence in 2018 as a hybrid exchange, maintaining two databases, taking orders centrally, and then submitting them to the Ethereum blockchain for execution. IDEX eliminated improved transaction speeds but introduced a centralized database that dirtied the decentralized purity test with a dreaded middle-man.
Uniswap launched in 2018, ushering in a new paradigm for decentralized exchanges growing to handle hundreds of millions of USD-equivalent daily transaction volume by 2020. Uniswap even outpaced Coinbase in transaction volume for a short period in September.
Uniswap eliminates the traditional order book that lists buy/sell offers in ascending and descending order and replaces it with a single liquidity pool that automatically incentivizes market-making. Traders can trade against a single price instead of a clunky order book, and everyone can become a market maker. On centralized exchanges, professionals with in-depth technical experience dominate market-making, running automated logic against exchange APIs. Uniswap democratized market-making, putting anyone with digital assets on a level playing field.
But Uniswap falls short in terms of transaction speeds. Mimo, IoTeX’s new decentralized exchange, finalizes transactions in 5-seconds while instantiating the features that made Uniswap such a smashing success. If Hamlet were alive today, he would use Mimo, and have avoided the backstabbing, Rosencrantz and Guildenstern ways of centralized exchanges.