The crypto knowledge gap — how we got here, why it matters and how to get better
If you enjoy this article please follow me @flatoutcrypto.
The past 12 months have seen a growing discussion as to whether cryptoasset valuations have outstripped current adoption. What is discussed less, however, is that market growth has outstripped the average understanding of both the technical and financial elements of the system. This juxtaposition between an increasingly mature market and a broadly immature knowledge base is impacting the whole crypto space.
For much of the last decade the average competence level was high. This was driven by many factors, but the main reason was because barriers to entry were high:
- Mining, storing or using Bitcoin or other cryptoassets was technically demanding for the average person
- Buying cryptoassets was harder due to lack of facilitating services
- Awareness was low, or allied to non-desirable elements (e.g. Silk Road)
- Bitcoin (and cryptocurrencies more generally) were conceptually hard to accept
As a result, the space was generally the preserve of those who were more technical (and by that I don’t mean purely developers or engineers, just people who understood how blockchain functioned) or who grasped conceptual tenets integral to Bitcoin (such as the importance of decentralisation).
The explosion of cryptoasset prices in 2017 brought an end to this, as a swell of parties (which had grown by this stage to include a mass of developers as well as a diverse range of actively involved industry bodies, companies, regulators, mainstream media and other subsets) brought public attention. This interest was met by more user-friendly services (such as Coinbase) which had steadily been improving and growing in prior years. And with it the demographic changed accordingly, lowering the average knowledge dramatically.
An industry geared to marketing
The rush to speculate meant that services sprung up to meet this demand. But this demand was not for an understanding of DLT, it was desirous of trading. And solely trading. Information sharing reacted accordingly, with trade discussion dominating traditional media, social media, message boards and cryptoassets websites.
The overwhelming question changed from “How do DLTs work and why will that change the world (which will bring me, an early investor, outsized returns)” to “What is going to go 5x in the next two days”.
This led to the rapid expansion of crypto marketing. I do not wish to portray the pre-2017 years as some utopia where we all sat around singing Kumbaya whilst working to better humanity. There was, of course, a great deal of speculation. The quest for profits has driven much of the interest in the space. But the smaller nature of the industry, the relatively high degree of understanding and the limited mainstream attention worked to limit the scale of ‘shilling’. Simply put, there were not enough projects, not enough people to sell to and not enough money to make it as worthwhile as it is now. Again, the change was swift.
- January 1st, 2017. Only 24 projects are valued at over $10m, and only Bitcoin is valued at over $1bn. Daily trading volumes sit at a healthy c. $100m. ICOs raised c. $100m in 2016
- January 1st, 2018: 507 projects are valued at over $10m. 42 projects are valued at over $1bn. 3 projects sit at over $100bn. Daily trading volumes for Bitcoin alone are $17bn. ICOs raised c. $5.5bn in 2017
- July 1st: Prices may have fallen but ICOs have raised $6bn+ in the first half of 2018
That makes for an average of $2bn a month raised for ICOs this year. It is hard to conceptualise what $2bn buys you, but be under no illusions — it buys a lot of marketing power. And this is what many ICOs use it for. Exaggerated ICO raises were used to transform teams with no need for that much money into hype machines. Digital adverts were bought. Publications were co-opted. Events and conferences sponsored. Numerous projects bought advertising space in prominent locations worldwide, including TV spots and prime real estate in locations such as Times Square. And people were also bought, retained as ‘advisors’ or ‘investors’.
Unfortunately, precious little of that money has been used on engendering people to actually use the networks or DApps. Too many teams (although there should be no doubts that there are a lot of sensible teams efficiently using funds) are generally more preoccupied with maintaining or boosting the price of their token rather than concentrating on wider adoption of their product. So many millions, billions, have been wasted on enticing the crypto community to buy in at crowdsale or propping up their token price with ridiculous schemes such as buyback initiatives, unnecessary exchange listings or buying supposed ‘partnerships’
Why it matters
But who cares? What’s the harm? If projects want to waste their money propping up their price and endangering hype then who am I to criticise?
But it does matter. It matters for a whole host of reasons because it:
- Weakens the ecosystem
- Increases volatility
- Increases exploitation of prospective investors
Weakening of the sector
Resources flow inefficiently across the industry. In any industry, resources (capital and talent) should flow towards the strongest whilst the weak die out. In crypto, however, capital currently moves to the strong and the weak. If you can market, you can raise millions — regardless of actual competency or experience in building anything. This has three effects:
- It slows down development in the space
- Weak teams are too rich to die out, given their huge ETH reserves. They can function as zombie companies, too rich to die, too inept to thrive
- Capital and resources are funneled to teams that do not have the best interests of the ecosystem in mind
The first point should be self-evident, although a slower rate of progress is probably acceptable if it’s the worst outcome. Zombie companies, meanwhile, are not ideal and will contribute to (1) but will not do lasting damage to the ecosystem beyond sharp shocks such as liquidating their reserves and winding the company down suddenly or launching a new company with their previous ICO raise and starting all over. However, this can be contained.
The third point is probably the most dangerous in the long term.
What happens if a project raises sufficient capital that it can bulldoze the competition regardless of its qualities as a decentralised network? What if the decentralised dream is simply co-opted, just like the Internet was before it? Having billions of dollars behind means you can achieve quite a lot. Decentralisation doesn’t have to win and there are plenty of competing visions of what it could look like (and if I have to read another whitepaper promising 100,000 or 1,000,000 or 10,000,000 tps while trying to hide the obvious centralisation needed for it I will scream).
A final point; many have made mention of the importance of governance. Users are increasingly expected to be an active participant in governance decisions. DPoS networks, amongst others, rely on this participation for the security of the network. How are users meant to vote when they lack the technical knowledge? Most don’t even know they are meant to vote. What impact does that have on the security of the network and the risk of centralisation on a network?
Divorce of prices from reality
The crypto markets are volatile. This is to be expected — we are still building models of how to value them, there are competing thoughts as to how/when/if different cryptoassets can be monetised, there is little adoption. In short, it is a new and developing market where we are in price discovery — volatility is to be expected.
However, I would conjecture that a lack of knowledge has exacerbated a divorcing of prices from any feasible reality. Instead of speculating on potential, an industry has sprung up to support the weight of money speculating on speculation. The alt run of late 2017/early 2018 saw many cryptoassets divorced from any rationality, a period during which it was a waste of time to read any whitepaper or think too hard about a cryptoassets because only one factor mattered — the ability to garner hype.
I am not merely restating the common observation that valuations as a whole exceeded any realistic potential valuation, but rather that a subset of cryptoassets exceeded the trends of the wider market. They became (and many still remain) divorced from any realism and were instead traded by a majority of people who did not understand them, did not understand the market and did not understand the components needed to even begin to value them. As such, they will have ended up owing assets which lost 95%+ of their value:
Exploring this in detail is beyond the scope of this article, but I would argue that the most poorly understood assets (also often the ones most hyped) were hit the hardest when the crash came, whilst also proving surprisingly resilient to genuine bad news (such as hacking, plagiarism, delays and other such small inconveniences) at various times. Why? Because bad news can’t impact the price of something not based on reality.
I could touch upon some of the unregulated aspects of trading here (washtrading, market manipulation etc) but I think these will be phased out as the market develops, bigger players launch exchanges and regulation is enforced.
However, the current practices of many ICOs are frequently appalling and aren’t discussed enough because there are too many parties profiting off an all too often opaque process. ICOs were initially conceived as essentially a means of crowdfunding and most of the early ICOs were bought up by retail investors in a single phase. As 2017 progressed this shifted, with more and more starting to do presales as a means to reward early members of the community or those who had contributed to the project in some way. This has now been even more stretched, with these presales now preceded by private sales and seed rounds.
Most ICOs now sell the majority of tokens in these early rounds, making them inaccessible to retail investors (citing regulatory reasons). They then either sell a tiny amount (5–10% of total supply) in the crowdsale or cancel it altogether. This creates both a pent-up demand and a concentration of supply.
I do not wish to downplay the regulatory risks of selling to retail investor, as these are valid. However, this shift to prohibition has been accompanied by a lack of transparency and keeping the rest of the token distribution model the same — to the detriment of the average person.
Traditional startups sell at a steep discount to early investors because they take on the substantial risk of project failure. This is why the traditional model sees the valuation of a successful startup increase over a series of funding rounds — because many die out along the way.
Too many projects have adopted this model and condensed it to make it meaningless. A seed round may take place mere weeks before the crowdsale starts. Many presale rounds end literally the second before the crowdsale starts. Many even run private sales with large discounts whilst also running the main sale (as at least one project is doing right now!).
So what is that funding for? Why are these heavy discounts given out? The way it works is thus:
- T-8 weeks: Firm does raise for say 2.5% of total expected raise (e.g. $500k on a $20m hard cap) with VC or other reputable company in the space. They will get a 50–90% discount (or a 100–1000% bonus, which sounds slightly more egregious)
- T-4 weeks: They take that money and go out to raise more money, buying some positive coverage on leading sites, maybe getting an ‘advisor’ to shill for them to get hype from lesser quality VC funds or — more likely — pools (raising $1–2m). This process could continue for a while, with some pools getting better discounts than others. Pools (or sole accredited investors) will probably get 25–70% discounts in most cases. Oh, and some VC companies will sell some of their allocation during this period to pools. Project teams will be made aware and any questions deleted or ignored
- T-0: Firm then takes that money and starts marketing in earnest so they can reach their $20m hard cap in the crowdsale. VC/reputable company backer logo will be plastered everywhere, as proof of how ‘solid’ the project is
- T+2 weeks: Token launches. If it opens at a 3–5x price then this means the team’s tokens (say 20–30% of total supply) are probably now worth $12–20m in addition to the $20m of raised ETH. If the token dumps on launch — who cares? It can drop 50–90% and the VC fund and pool will still have made money.
The funds being raised aren’t being used to allow the team to continue development before a subsequent funding round. Most teams only do one token generation event and sell off the token supply early on. No, they give steep discounts as a shortcut to the hard work of building a community. They do this in order to raise as much money as they possibly can in a crowdsale.
But who cares? Well, for one investors get screwed because they are both unable to partake and are often not privy to the details of the huge bonuses and discounts being given out. These are usually associated by pathetically small (no, a six-week vest period doesn’t count as a lockup!) or no lockup periods. These investors then get dumped on by VC funds and pools (again, to clarify this is obviously not all VC funds and pools — there are many good actors). There is no good reason not to provide the details of prior raises to all investors. Transparency is a good thing, not bad.
Trends such as these (and there are more) will reduce trust in the space, lead to increased regulatory action and ultimately stifle adoption because teams aren’t building up a community of interested users, they are simply involving a small set of people who wish to flip their tokens quickly. It is in the interests of everyone to do more to counter it and highlight flagrant abuses
So how do we get better?
There are some aspects which will naturally improve over time. For example, volatility will decrease as larger investors move into the market, more transparent exchanges emerge and as we develop better and more widely accepted models for valuing cryptoassets.
As such, I’m less concerned about trading and more interested in measures to both help bridge the technical gap and prevent the exploitation of many investors. Some of these steps are small ones, but ones I believe are both practical and will improve our industry.
A simple change. Teams don’t need to disclose individual investors, but they should make clear the effective token price and the number of tokens sold during each round of investment. ICO listing/review sites should begin to highlight companies which don’t provide this transparency and make bonuses given one of their most important factors for consideration. It goes without saying, of course, that tokens subject to vesting periods should be held in smart contracts and only released at the appropriate time (it is staggering how often a project relying on smart contracts doesn’t use smart contracts for a very obvious purpose).
Extending this slightly, teams should be expected to provide comprehensive monthly or quarterly updates including where funds are being spent and if this is in line with the crowdsale promises. All too often the last we hear of a team’s plan and burn rate is in the whitepaper.
Given regulatory unknowns, project teams may be tempted to reduce their transparency as opposed to increase it. We need to fight this and encourage transparency as much as possible. Self-regulatory industry bodies such as the ICO Governance Foundation are necessary developments in this.
More consideration given to token sales/distribution
Distributing the bulk (or entirety) of supply at the start is sometimes appropriate, but not in all cases. However, it is currently the default setting. More teams should consider doing smaller raises and retaining supply until a later point. It would probably better align interests between team and project whilst giving more flexibility down the line.
Retaining supply can also be used to incentivise users (not traders) or to drive adoption at product launch. Teams should also consider longer lockup periods. Yes, it will likely put off some investors — but there are many examples of projects with decent length lockups raising more than sufficient funds.
Furthermore, teams should stop distributing unjustifiable bonuses which are often given out just weeks or days ahead of the main crowdsale with no vesting period. There are numerous teams which have shown that it is more than possible to succeed without this behaviour. Ally the size of the bonus to the risk they are taking and the import of the money they are providing.
Pay more attention to user participation
Again, I will try not to pick on particular projects, but launching a network which relies on voter participation and not making it easy for users to actually vote in electors is unacceptable. Imagine if a government asked people to vote, but relied on the people themselves to build the voting stations and print the voting cards. How effective do you think that would be?
Decentralisation relies partially on encouraging as many users as possible to participate and as such the same applies if you do not provide clear instructions on how to run nodes, stake your tokens or whatever else is needed to facilitate as many people securing the network as possible. These are simple steps to take and firmly in the remit of the project team. It is not as simple as ‘build it and they will come’. It needs to be ‘how do we build it so that they come’.
Crypto journalism and curation of content
The issue with journalism is certainly not confined to the crypto space. Most (not all) crypto news sites are essentially paid for advertising or filled with low-quality pieces that read as if a bot has spewed 500 words onto a webpage (and sadly this is probably what many are).
Furthermore, no-one cares enough about negative reporting on a project they aren’t invested in (there are thousands of other projects — why bother wasting any of your time on the bad ones?) but people really care if they are invested. This leads to pressure and threats, as armies of shills cry ‘FUD’ at every opportunity even when the accusation is true. It’s not FUD if it’s a real issue — it’s just news. The result? As a writer, your work gets read more if you write nice things and you don’t have to deal with harassment. You can see which aspect is likely to get written about more.
As I do not hold much hope of this changing in the near future, the onus for quality is therefore largely on the curation of user-created (mostly free) content.
The amount of information resources concerning cryptoassets has increased significantly but a lot of the best quality content is now hidden from sight, either confined to private groups seeking to keep out the ‘wen Binance’ brigade, on a harder to search medium (e.g. Telegram) or simply lost in a maelstrom of voices competing to promote their usually useless favourite holding
For example, Medium contains better quality content than any crypto news site but much of it is destined to never be read as it is hard to find. Unless you are one of a surprisingly small number of people who can rely on retweets from those with the biggest following, your content will stay largely hidden. Sites such as Reddit also struggle with this problem, as bot armies upvote poor content whilst good content goes ignored. Regular users tend to share and upvote content that portrays their project in a positive light, downvoting everything else. Twitter is better, but of course largely relies on the curation of your follower list. That said, I rely on Twitter to find the best content hosted on other sites — and the quality of materials linked is heads and shoulders above other platforms.
A growing trend here which is helping a lot are daily and weekly newsletters such as TokenDaily and TokenEconomy as well as curated Twitter reads by the likes of Nathaniel Whittimore. These are of great importance and of real benefit.
Another approach as to how we might improve can be seen in projects such as Messari. Messari aims to build “a public data library that hosts standardized token issuer disclosures [that] could serve as the financial backbone of the token economy, much like the SEC’s EDGAR database in modern financial markets”. This ties in with the need for transparency but Token Curated Registries (TCRs) could also help with promoting and sourcing good quality content.
This use of TCRs, in which project information is free, open-sourced and incentivised economically to ensure accuracy and quality, could prove to be integral in improving the quality of information. They aren’t without their problems, but they can certainly help.
Project provided content
This is a tricky one but I believe that companies have a role to play in better informing the public about what DLT is and the components that make up their project. Depending on the company, the information could vary, for example:
- Projects: What are DLTs, what is a blockchain, how does consensus algorithm X work and why have we chosen it, what are the potential attack vectors we are working to eliminate, what are the problems inherent to our solution (CAP Theorem, scalability trilemma etc)
- Exchanges: What are the risks of leaving your assets on an exchange, how do you store your own tokens in private wallets, basic information about each cryptoassets (supply, current market cap, issue price, important news events, links to information about number of investors/crowdsale details etc, explanation/breakdown of whitepaper etc)
This can range from the fairly brief (such as the introductory warning that MyEtherWallet/MyCrypto bring up the first time you visit) to more detailed resources such as BitMex’s research output. It would help if there was more content in non-English languages too.
Projects should also concentrate on making their whitepapers more accessible. I took a sample of some 500 whitepapers and analysed them by word count and readability (for ease I took the Flesch-Kincaid grade, I’m sure there are flaws and issues with it but it was the only realistic way to go through so many whitepapers). I found that:
- The length of them dramatically increased, from an average of 5,869 words in 2016 to 9,381 in 2018
- The Flesch-Kincaid grade score improved from c. 18 to 13, meaning they became easier to read
These results will be somewhat arbitrary owing to the random sample of papers I took, but the increased length tallies with my own experience. Unfortunately, I suspect most of the additional content is the same regurgitated focus on what blockchain is and Bitcoin’s lack of scalability yadda yadda yadda — we get it. Bitcoin can’t do as many TPS as Visa, we’ve heard it hundreds of times. Focus on explaining your own project, how it works and how it will improve users lives.
I was surprised by the drop in Flesch-Kincaid score but it may be because the average ones are targeting the average investors and so are deliberately making it simpler. Some of the more ‘serious’ whitepapers, on the other hand, are borderline incomprehensible. Keep at least your abstract accessible to all (sidenote: It may be nearly a decade old but the Bitcoin whitepaper remains one of the clearest and best examples of what a whitepaper should be like, and is well worth reading even now).
One of the biggest drivers for change will be projects beginning to fail, driving token values to 0. The correlation of cryptoassets thus far is high, meaning that the weak projects increase or decrease in a similar fashion to the strong. As the market matures and we begin to see projects fail, investors will realise the importance of due diligence and money should stop flowing to bad teams.
The space is a new one and as such technical knowledge will proliferate over time to a wider audience, just as it did with computers, the Internet, app development and essentially every technical development.
To speed this process up, I would advise those new to the space (or those who want to understand the technical side more) to concentrate less on individual projects and more on the building blocks necessary to understand them such as:
- What the different types of DLT e.g. blockchain actually are
- What decentralisation means in the context of blockchain or Bitcoin
- What consensus algorithms are and how they work to solve the Byzantine Generals Problem
- How are networks attacked? How do we stop them from being attacked?
- Why do DLTs make the design choices they do e.g. the Scalability Trilemma, CAP Theorem
Understanding just these five starter topics in moderate detail will make you far more capable of distilling information thrown at you and being able to objectively assess which ones are taking unacceptable shortcuts or putting forward an undesirable system.
In time some problems will resolve themselves naturally. We will see more attention paid to valuation models and the entrance of institutions will bring increased levels of professionalism and research resources. Bad actors will be uncovered.
Other problems are not confined to crypto, such as the growing trend that people believe the truth they want to believe, the emergence of fake news and the sheer amount of information now spread across our global and efficient networks.
However, as outlined, there are clear and easy changes that can be made to make the situation fairer and more accessible to all. If adopted, they would make the ecosystem a stronger and more equitable space.
If you enjoy this article please follow me @flatoutcrypto.
Disclaimer: I have contributed to pools in the past and will likely do so in the future. I do not think there is anything wrong with pools from the investor side and believe the onus for transparency and to end outsized bonuses is on the project team side (although I will generally avoid contributing to projects with these sort of bonuses).