"It ain't called high risk on account of we're hopheads, pardner"
By Daniel M. Ryan
Once upon a time, in the later middle of the Great Altcoin Boom of 2014, there emerged a new contender called Crypti. Pre-announced in June and launched at the beginning of July, its first list of features had one would end up standing out: sidechains. Blockchain ledgers that were run by a subset of the entire Crypti network and secured by the entire network by continual contact with and verification by Crypti's main blockchain. Originally, its blockchain securing was done with a blend of proof-of-stake algorithms; later, it was revamped to run on a Delegated Proof-of-Stake model. The main chain and all that's in it was then secured by 101 subsidized Delegates, each of which held their position through suffrage by Crypti holders. The Delegates were also responsible for building up the blockchain by adding new blocks with the most recent transactions in each's memory pool. These transactions were broadcast over the network from regular users before the Delegate added them when its time came. Since each addition was done in round-robin style, the block time of this system is far more stable than the more approximate targetings found in Proof-of-Work.
The Crypti folks, once they realized what they had, figgered out that they had developed a platform that was a real competitor to the now-hot Ethereum. But the trouble was, not many altcoin folks were all that interested. Crypti, being a relatively late entrant in the 2013-'14 flood of altcoins, arrived not long before the Great Altcoin Bear Market of '14-15. That bear, it was a killer. Altcoin indices, such as they were, looked like the stock of a single big bank over the first three quarters of 2008.It was that bad. And as was its wont, it claimed many good altcoins along with the not-so-good. Right about the time when the Big Bear went into hibernation, as of the new year of 2016, two of the Crypti heavyweights realized that they were sitting on a fantastic technology that fewer and fewer folks were interested in. A "legacy issue," the big-city folks call it. In Crypti's case, the legacy issue was it being mauled by a powerful, ravenous and callous-hearted bear. A bear that did not discriminate in the least against a so-called ---coin and one with promise.
Then, right around the time when the frontiersmen still in altcoin land were mostly mired in calloused skepticism, two of Crypti's heavyweights had a multimillion-dollar brainflash. If Crypti's legacy issue was it being mauled by the big bear, whose bitings and pawings had nothing to do with the promise of the tech, then what better way to shake them off than by announcing a new altcoin that was a fork of Crypti? What better way than to start with a blank slate and market the tarnation out of it? Thus was born Lisk – the Decentralized Application & Sidechain Platform.
This pair of intrepid and shrewd frontiersman, they being Max Kordek and Oliver Beddows, left their Crypti kithfolk on friendly terms. Others in their kith had also realized that Crypti's wounds from the bear-maulin' were saddeningly deep. The pair showed their friendly by offering a deal to every Crypti holder: if you plunked in your Crypti for Lisk in its pre-sale, you'd get credited 1300 satoshis per Crypti. Since 1300 was the then-flaccid market price of Crypti at the time, it was a square deal.
The overall pre-sale was one of those variable-price dealies in which a fixed amount of Lisk – 85 million – was to be allocated to each buyer in proportion to the total amount of Bitcoin sent in. The more Bitcoin (and Crypti) that rolled in, the higher the implied price of each Lisk and the lower amount one Bitcoin would get you. To make things more complicated, there was a four-tiered structure of bonuses to incentivize early pre-buyers. This set-up meant that the number of Lisk per Bitcoin moved on downwards as the pre-sale gathered more funds. Consequently, the third member of the original Lisk trio – Community Manager Joel Fernández – had as part of his chores the duty to explain why everyone's Lisk balances were apparently shrinking.
I have to hand it to the trio: their presentation and campaign was not only slick but also very well-thought out. There was documentation aplenty and a lot of help on the thread for folks who were puzzled. Among the documentation was an explanatory guide to the Delegated Proof-of-Stake system, a Lisk Academy, a dedicated Lisk Forum away from the hurly-burly of the main announcement thread, and even a Lisk App Software Developer Kit embedded in the Forum. Yes, there were lots of instructions on how to use Lisk and the doc writers were thoughtful enough to make the instructions step-by-step. There was even a Lisk Chat for folks who needed to talk right away.
Yessir, it was quite the professional rollout that was going places. The marketing process contained a technique that would make Lisk the hottest pre-sale since Ethereum. As part of a publicizin' bounty program, qualified folks who signed up for a Bitcointalk signature campaign would split a pot of 100,000 Lisk for each of the four weeks the pre-sale ran. If you were intrepid enough to sign up right when the campaign began, and if you qualified and kept promoting Lisk all the way through, you'd get one share from all four pots. If you were a latecomer, you qualified for fewer. This campaign had a really potent virality, as seen in the number of qualifiers eligible for each week's divvy. For the first week's, fewer than one hundred qualified. For the last week's, the number of qualifiers was well over a thousand.
Improvement, Delays, Cautions, and…
The community was growing too, and was becoming more enthusiastic. As is the case with popular cryptocurrencies, several volunteers stepped forward entirely of their own initiative: they began shepherding the threads, answering questions and becoming the life of the party. More technically-inclined volunteers pitched in with Lisk's testnet, in which there were quite a few bugs scotched out. The bulk of them were minor, but a few were major. Along the way, the Lisk dev team decided to perform a major upgrade by switching the database from the Crypti-inherited LevelDB to the stronger PostgreSQL. They also performed some stress tests to see how many transactions per second the delegate network could shoulder. The answer they got was somewhere in the neighbourhood of a hundred transactions per second, which was well above the capacity they had inherited from Crypti. Yes, it really did look like Lisk was headed towards an unhurried orderly launch. There were some complaints when the initial end-of-April timeframe ended up as the firm May 24th, but there weren't that many. Certainly a small sample compared to the growing volume of Liskboi enthusiasm that filled a thread growing more and more rapidly. The Lisk thread grew to be larger than the entire Crypti thread – all twenty-plus months of it – in well under three months.
The reason given for the initial postponement of the launch, from late April to early May, reflected this care. If you pre-bought some Lisk during its ICO, each send of Bitcoin or Crypti generated a unique address. If you went all-in with a single spend, you had one address. If you sent some and decided to send more because of the excitement, you got two addresses. If you were like me and sent in three spends, you got three. And, so on. Lisk uses a brainwallet system, in which a randomly-generated twelve-word passphrase determines an account's private key which in turn determines the public key. As is consistent with good cryptography, this deterministic relation works only one way. (Just so you know, a cryptocurrency address is either a deterministic alias of the public key or the public key itself.) This one-way deterministic process of the brainwallet means that the passphrase determines the account. As part of the care taken before the launch, the Lisk devs repeatedly asked all pre-buyers to validate their accounts: to enter in each password for each send into a special form in their personal pre-sales pages in order to verify that their Lisk was there. The devs' goal was to have well over ninety percent of accounts validated, as they were sensibly wary about a trollstorm caused by lots of folks not validating and then very angrily wondering where the Lisk was. This caution was the public reason for the first delay. Squashing bugs and improving Lisk's network was the public reason for the final pushback to May 24th.
In a subtle but definite foreshadowing of what was to come that big day, complaints about the first postponement did not come from outside jeerers or FUDder trolls. They came from some community mainstays, who opined that there was no point in delaying the launch on account of some stakeholders being too lackadaisical to validate their stakes. If those folks got into post-launch complications – well, they had it coming ‘cause of their inattention.
Another safety precaution, which put a crimp in an otherwise normal development of an OTC market between pre-sale's end and launch, was the option of resetting your master password that gave you access to your ICO page. From a safety standpoint, this precaution made pre-launch Lisk a risky buy. A malicious ICO stakeholder could sell his or her stake(s) and then turn around and reset the master password so as to lock the OTC buyer out of the sold account. Since the reset option was a safety feature offered by the Lisk team to stakeholders to cover for any pre-buyers losing their master passwords, the Lisk team kept this feature in place and publicly warned the entire altcoin community against buying any pre-launch Lisk. This warning was spread around by several volunteers, again self-appointed in this permisssionless ecosystem.
The devs' caution was entirely understandable, in part because some of those sends were huge. There was one single pre-buy in the amount of 1,000 Bitcoins. At the final implied-ICO price of about 19,000 satoshis, leaving aside early-buyer bonuses, this mystery whale wound up with over 5 million Lisk. There were several other heavyweights who ended up being "LIskionaires" entitled to one million or more; there were more than a few who sent in double-digit amounts of Bitcoin. As we found out as we awaited the big May 24th launch, two of those big buyers were second-tier alternative-cryptocurrency exchanges looking to get the jump on the bigger exchanges.
As the hype and excitement built up after the pre-sale closed, one of those two exchanges stepped forward to announce that they had bought a large allocation from Lisk's ICO and that it was listing its IOUs for trading. The exchange was Yobit, which already had a reputation of listing the scuzzier altcoins of the frontier. (Cases in point: TrumpCoin and BernieCash.) It also had repute for attracting the high-excitement gamblin' clientele who were always ready to take their chances on a good ol' pump-and-dump. (That aforementioned TrumpCoin went through one shortly after it was listed there.) The excitement over Lisk had grown to the point where the eager punters were all-too-ready to take a nice flyer on this new sizzler.
Consequently, when Yobit opened up its IOU-Lisk market for trading, the punting and denting was so furious that Yobit racked up more than 500 Bitcoin's worth of volume from Lisk alone on that day. They stepping forward and taking a step that was shrewd if questionable, as they only backed up the "LIsk" they sold with a screenshot of their pre-sale buy, meant that they earned more than one Bitcoin's worth of trading revenue from Lisk alone on that first day alone. You can question their ethics, but it's hard to question their shrewdness. Their gate-jumping maneuver earned them several hundred dollars' worth of commission revenue on that wild first day. They shrewdly took further advantage by buying up a slew of banner ads on Bitcointalk that said, Trade Lisk on Yobit!
A much newer and much smaller exchange named Bloombit got into the same act and enjoyed a huge boost that put it in the lower-tier altcoin-exchange map. But the difference between the two was seen in the relative prices for their IOU-Lisks: all through the pre-launch, Bloombit's sold at a 70-80% discount relative to Yobit's. This discount, the increasingly excited Liskbois chalked up to Bloombit being new and shady.
But the discount reflected something wrong about each respective market: its artificiality. The supply was restricted to the amount of Lisk each exchange had bought. Each supply was only a tiny fraction of the 85 million pre-bought Lisk that would be released come May 24th. Each market started off with what was essentially a laddered series of OTC sales from each exchange as principal to its clients. The hothouse nature of each market was so artificial, these initial sales had two features in common: a) both were way above the final implied pre-sale price, guaranteeing each exchange a huge capital gain over and above the trading revenue; b) both were way below the average price in their respective prelaunch IOU-markets. Any of their customers who bought from the exchange sell-ladders-as-principal was soon sitting on a large unrealized profit simply from other punters jumping in.
Every pre-sale of a hot cryptocurrency has its boosters who project that the interval between presale's end and launch will see demand grow to the point where the newly-listed cryptocurrency jumps up to several times its implied presale price. The longer the wait between pre-sale end and launch, the more plausible these boosterings seem to be. The Lisk fans settled for a post-launch price of a dollar per LIsk, which would have made for about eleven times the resultant pre-sale price. During normal waits, these excited predictions are taken in stride. But because of those supply-restricted artificial markets, "Lisk at A Dollar" began to seem plausible.
I have to confess to you, dear reader, that I too became convinced that a dollar was feasible. I was lulled even though I recognized at the outset that the markets were artifices. Why? Because I expected that, once the initial Yobitter excitement had faded and replied with dull listless trading, that the artificiality of the market would be recognized by the punters who would sell down IOU-Lisks to something reasonable. Sure enough, the volume did lull. One day, it sank to about five Bitcoin's worth of trading. The price did sink alongside it, making for a long valley between the new-listing gambleroo and the pre-launch excitement. Oh yes, Yobit's IOU-Lisk price dropped – but it did not drop below two full dollars. The rationalization I expected to happen, once the white-hot pump-fest cooled, apparently indicated that Lisk would open up at a dollar or more. After all, it had clearly slumped. Presumably, the hotheads had taken off and left cooler, more rational head behind. So I imagined.
Lisk-opening-at-a -dollar really did look feasible in those early and mid-days of May. More than a month's worth of boostering apparently had worked up enough magic to make its launch-price more than ten times its pre-sale price. Unsurprisingly, the boosters escalated. One dollar became five dollars. Never mind that a five-dollar open would imply that Lisk's market cap, using the total supply of 100 million when including the bounty accounts and the devs' own stakes, would have given it a five hundred million-dollar market cap from the get-go.
As launch day approached, and excitement built up both volumes and prices at Yobit and Bloombit, the skeptics who tried to bring in common sense were lumped in with the trolls. As the go-time of May 24 8 PM UTC approached, Yobit's IOU price went back up above 0.01 Bitcoins. On go-day, it climbed above seven dollars.
Thank You For Flying Tornado Galactic
Just before the Big Day, the Lisk team announced the major exchange that would be listing it on launch – one that they had kept a secret because of a non-disclosure agreement. Common-sense folks like me had presumed it would be Bittrex, a big exchange that does list major alts on launch. We were right and wrong. Bittrex did announce the listing a bit before launch time, but so did the much bigger Poloniex.
This was huge news. Even for white-hot launches, Poloniex usually left it to Bittrex to be the first lister. "Polo"'s principals took their time and audited the code before adding even a biggie. In order to make sure nothing went wrong, Poloniex waited for a full three hundred and one confirmation before crediting a Lisk holder's deposit. Effectively, they were waiting for three full round-robins of Lisk's 101 delegates before making sure it had securely received the Lisk.
In normal times, with the network running normally with normal block times of around ten seconds, this 301-confirm wait would have taken about an hour.
As is customary, Poloniex opened up Lisk trading before anyone could deposit any Lisk. The buybook right on launch was a thing of beauty to any Lisker. ‘Twasn't just me to took his cue from Yobit's IOU-market. Just below two jovially silly bids of less than ten-dollar value, one at a price of a little more than a Bitcoin per Lisk and another at 447 (!), there were several serious bids all above 0.02. Right around launch time, there was more than 30 Bitcoins' worth of buys down to 0.007 BTC or three dollars. At the eve of launch, those bids were not for any hothouse IOU-Lisk. They were for real Lisk.
Naturally, we excited presale buyers believed all the hype about no-one selling below a dollar. Isn't that what the real, live bids on Poloniex were indicating? There was one skeptic. During the pre-launch buildup in mid-late May, he offered to sell his presale-bought Lisk for as low as 70 cents each. As he noted several times, anyone who believed that Lisk would open at five dollars – or even at a dollar – would snap up a nice risk-arbitrage gain with only a few days' wait. He found no takers, even though he made his pitch several times and offered to front the fee of an escrower whose reputation was golden. After getting no offers – the only one who came close was yours truly – he started predicting that the real same-day launch price would be somewhere between twenty and forty cents. Of course, he was scoffed at and no-one believed him. The at-launch buybook certainly belied his skepticism.
When the clock hit the go-time of 4 PM ET on May 24th, that long- and widely waited time for that carefully-prepared launch, LIsk looked like it was propelled up into the sky by the very wind. But as we found out right after, that wind was a tornado.
To put it bluntly, all hell broke loose.
For the launch, the Lisk team had provided Linux wallets. For Windoze users like me, the wallet came in a Docker container. For the folks like myself who didn't know Docker and weren't about to take an insta-course on how to use it, the Lisk devs had thoughtfully provided an official Web wallet. That's where Windows folks like me barged to after the genesis block got revved up and the blockchain began to build.
That is, we tried.
As someone who got a little stubborn with it, I found out that I was lucky to get a page at all. I saw "Error 502 – Bad Gateway (next line) nginx" so many times, I memorised it. And I wasn't the only one; far from it. During the chaos, Max Kordek popped in and revealed that at one point the official Website had fifty thousand open connections at one time. Since there were only a little more than 3,100 stakeholders in total, he concluded that the Web wallet was the victim of a DDOS attack. The team's otherwise-thorough preparations had not included a DDOS-protection service like Cloudflare.
To their credit, they did throw up more clusters with similar Webwallets. At the height of the chaos, they had as many as eight. The folks who stuck to the Webwallets began swapping tips on which ones were up and which ones were usable.
"Usable": that was the second point of chaos. During the whirlwind, I did manage to log in several times and fill in the "Send" form most of those times. I'm sure you've already guessed where I was sending them to.
But each and every time I tried, the form remained stuck – in one case, for more than a half an hour. Even if I was lucky enough to get the webwallet up-and-functional on the front end, the back end frizzed out.
While I was more-or-less patiently waiting for the web wallet to work, or more-and-more impatiently refreshing and re-logging in when I could, something funny was happening at Poloniex. Some lucky fellers had managed to deposit their Lisk and wait out the 301 confirms, Boy, did they dump! It wasn't long before the price sagged to below 0.002 Bitcoins: well below that mythical dollar. A snap-up to about 0.0027 proved to be a sucker rally, and the mighty Lisk got blown down to below 0.0005. That skeptic had been bang-on.
Through all this, I – and many others – were refreshing the Lisk announcement thread to try to figure out what the floop was going on. As it turns out, the bulk of the early and very fortunate dumpers were old Linux hands. Many of them had volunteered for the testnet. Naturally, this (shall we say) disparity called forth a lot of angry complaints from Windows users. The fortunate ones – all of them except a single fellow with some words of sympathy – said that we Windozers had it coming. We had had a lot of time to set up our own virtual private servers, which certainly would have gotten around the DDOS block had they been kept private, and to install the Linux brand of wallet. We had lotsa chances to do what those happy Linuxeers did.
So, bowing to the inevitably rough-hewn frontier judgment, I did just that. To the devs' great credit, the step-by-step instructions on how to install a Linux node worked as advertised. It really did only take minutes for me to set up a Linux node, fire up my own private Webwallet client at my VPS's IP, and try to fire off my Lisk to Poloniex while watching the price plummet.
You may not know this, but the underlying backbone of all cryptocurrency systems is a peer-to-peer network of nodes that coalesce around the longest blockchain – the one with the greatest number of blocks added to it - that is valid. In normal times, there's no real need to know this. But during tornado times, you figure it out quickly. My private webwallet, in a private VPS for which there was no feasible DDOS vector, did have a browser client that did what it was supposed to when I clicked "Send." But none of my txes were broadcast to the network. This, I found out by re-logging in to my old accounts only to see my full balance returned. As a sideline to my…little adventure, I had the odd experience of sometimes seeing my balance flicker between my pre-sale amount and zero.
Thus giving up, I returned to the announcement thread and found a real Godsend. At least one kind volunteer had thrown up an independent Webwallet that worked. Hastily throwing aside any qualms about sending my passwords to a heretofore unknown server, I quickly logged in to my three accounts and sent off my entire balance to my Poloniex deposit address. Since I was emptying them out anyway, why should I care if the server were logging my passphrases? Even this theoretical qualm proved to be empty as the fellow who saved my night did prove to be an honest doer of good.
And his underlying node was synched up! Mirabiile dictu, I managed to verify that two of my three account-emptying sends did enter the blockchain. The third one, neither the biggest nor the smallest, I couldn't track down. But two was a blessed relief!
While I was fumbling around, I saw after a time that Poloniex had recognized one of my sends as "pending." But what should have taken an hour to complete in normal times, took at least three hours in tornado time. When I finally gave up and went to a very late beddy-bye, only two of the three sends showed as Pending. The third, the one that I couldn't locate by using the charming Webwallet block explorer, didn't show up at all.
But – as I discovered after waking up very late on May 25th – my funds were safe. All three of the deposits – including the one that had gone missing – were safely Confirmed and credited to my Poloniex account. Even if they hadn't, they still would have been safe in my accounts. This, some folks who had more trouble than me found out. No-one lost any Lisk.
Even before that time, the bulk of the dumping had dwindled and Lisk trading was beginning to look like a semi-normal market. I even bought about 2200 more at a price below 0.00043 while my deposits were stuck: a price of less than twenty cents. But that…adventurous Night of the Launch Tornado had rattled me so much, I sold off what I'd bought for a mere scalper's profit. Never mind that my trader's sense said that I had bought near the bottom.
The next day, I sold off all my pre-bought LIsk at prices ranging from 0.0005 to 0.0006. I didn't care that the market was rebounding. Quite literally, I found out the meaning behind the old market saw "Sell to the sleeping point." I literally threw away a double over and above my post-ICO profit because otherwise I would have tossed and turned all night worrying.
The Moral Of The Story: That Ain't Boilerplate
Refined ladies and civilized gentlemen, you have just read a slice of real life in a big Boom-Town spot of the cryptocurrency frontier. Now it is true, in normal times, that you can read the cryptocurrency standard "This is experimental technology" and give it no more through than you do those "Risk Factors" in those civilized 10Ks. But in a time of a white-hot launch turned tornado launch, a time when the marketing genius of a crack team of experienced and capable devs seems to have made them the Sorcerers' Apprentices of hype, that's a time when you figure out what "high risk" really means in the cryptocurrency frontier. Suffice it to say that the real risk is not captured in your well-refined risk-tolerance metrics. In a time like the Lisk launch, it comes down to being technically-astute and resourceful enough to figure out how to do what you want to do when all around you has gone twisty. Oh yes, it is educational. You can pick up some basic knowledge on how the underlying cryptocurrency network runs during those times, just as you can pick up an idea of how a barn is built during tornado time.
"Oh, that's how they put' em together – funny how I never figured it out before."
As with all frontiers, there's not that much of a premium on refinement – but there's a huge premium on keeping one's head. Moreover, as you folk historians know, there's an even bigger premium on keeping in touch with the altcoin's community. Yes, there are some mercenary folk and some ornery folk. There's always someone to tell you that you had it coming to you because you shoulda known or were duly warned. Refined and civilized folks can make do with "innocent until proven guilty," but in the cryptocurrency frontier a refined and civilized operating principle like that one makes you a gull ripe for the juicing. In this rough part of the world the practical operative is "where there's smoke, there's fire."
But there's also something precious in these here parts: a real gut-level sense of community, where folks intuitively learn to look out for and help one another. No more so than in tornado launches like Lisk's. Twisty, turmoil times really do reveal which folks help out for the plain and true reason that they like helping.
Lisk's community did take a battering back then, but it's still strong in spirit and is now turning to the task of getting real DaPPS – now known as Blockchain Applications – into the Lisk infrastructure. Lisk really does have a shot at becoming the next Ethereum. Not for a long time, more likely than not, but the fix is in for it.
Please mark well my parting words, though. You may be bedazzled by tales of post-presale gains of two times, three times, or even more times: I know of one twenty-five bagger at its pre-launch height. These exciting Boom Town tales may entice you into this here frontier land. If you are so tempted, please note well:
It ain't called "high risk" on account of us cryptoneers being potheads.
Daniel M. Ryan, as Nxtblg, is shepherding the independently-run Open Audi Initiative Prediction Market Shadowing Project. He has stubbornly assumed all the responsibility and blame for the workings and outcome of the project.