Some ideas are ahead of their time, like Bitcoin or flying cars. And though technically possible, full scale adoption lags for years or in the case of flying cars never happens. Other ideas are perfectly placed chronologically in such a precarious moment that hindsight makes them seem inevitable. Thus, the case with the concept of a blockchain operating system. It’s a chicken and egg issue really. At the core of it, for blockchain to make significant real gains with actual use cases and real adoption as opposed to simple speculation, there must be some easier way to create products that use the technology; an operating system. At the same time, to make a good operating system it’s best to understand the use cases first and design with those in mind. Ethereum and EOS and many others are all taking a stab in the dark at providing platforms that can shoulder this new generation of application. The term Turing Machine comes up a bunch in these discussions. That's because the utopian ideal is that for a system like this to work best many computer would have to operate as one functional system. In my opinion this is still a utopian concept but the term Crypto-Turing might apply when describing these self-described blockchain operating systems. Early adopters in the crypto-turing space seem to think that the argument will eventually lead to a winner takes all monopoly. This may be true, but I’m not so sure it’s sooner rather than much later.
As I understand it, the two heavyweight contenders in the capacity to provide an “blockchain operating system” are EOS and Ethereum. But they don’t take nearly the same tack in accomplishing that goal.
As I see it, these are some of the main differences:
| Function | Ethereum | EOS |
|---|---|---|
| Consensus mechanism | Proof-of-Work currently but has always stated would eventually move to Proof-of-Stake (fork has been termed Casper for google purposes). 2018 was the original date for this fork but setbacks have seemed to keep it from fully materializing. | Distributed Proof-of-Stake. A system invented by Stan and Dan Larimer (Bitshares, Steem, EOS). This system modifies PoS (Proof-of-Stake) by allowing token holders to vote on which “block producers” (Miners) will be allowed to produce blocks for a time |
| Token Paradigm | Ether is used as currency and is traded to miners as payment for blocks solved. Similar to Bitcoin. | Tokens stake as ownership, kindof like shares or land, a given number of tokens allows you access to a percentage of the total processing power of the network. “block producers” (miners) are paid through an inflation mechanism not to exceed 5% per year. |
| Supply Ideology | Set supply of ether reached after a certain number of blocks mined. Very much like bitcoin. As with many other cryptocurrencies this has a deflationary effect during growth periods. IE, if you get in early the value of your coins can appreciate rapidly and volatility is wild compared to fiat markets. | Starts with 1 billion tokens. But this inflates every year by up to 5%. The theory is that this will help prevent volatile pricing. Although anti-speculation/volatility function is unlikely to take effect until well after release. |
| Governance | Governance is established through forks. A consensus must be reached by a majority of developers and users and forks commit the change if any. Technically speaking this process isn’t codified in a very formal way. The ideology about immutability is a hotly debated topic among the Ethereum community. The DAO incident was an example of this governance policy however. | 17/21 majority vote. That’s what it takes. Voting is stated to be an integral part of the system from voting for block producers to voting for which changes to fund. A vote is earned by owning tokens. One vote per EOS. This can be compared to bitshares and steemit I assume as they’re both predecessors of EOS and products of the same minds. |
Ethereum and EOS differ more than they agree overall. In fact, it seems that one of the few ways the two protocols agree is that there ought to be an operating system for those who want to deliver a service or product via a blockchain. As compared to each other they both offer a few notable cons as of today.
Ethereum has inherited the volatile market issues from its predecessor Bitcoin. There are economic justifications for supposing that once adoption reaches Its eventual saturation that things will calm down a bit but there is a good deal of road between here and there. Even then, the price of Ethereum is likely to fluctuate a fair amount. This is a problem, as we know from Bitcoin, in deciding what to charge for a good or service. In fact, justified or not, volatility is one of the main criticisms against Bitcoin as a new monetary standard. I assume with Ethereum it will be much of the same. As the ICOs leave beta and begin offering real services it will be interesting to see how they handle ether volatility. Services which offer high profit margins can eat the fluctuations and they’ll balance out, others with strong competition may have trouble based on a network like Ethereum. In fact, after Ethereum moves to Proof-of-Stake this problem may be exacerbated. Staking requires locking ether away for a period of time. Locked ether can’t be traded so the tradable supply goes down. The logical end of this is that because there’s a fixed supply as more miners are needed to scale transaction times more and more ether gets staked up. Thus, reducing tradable supply and causing deflation and volatility. Because PoS hasn’t been fully fleshed out in Ethereum some of these points could be moot. Given the developers excitement about creating incentive I wonder if there won’t be a clean way to prevent too much Ethereum from getting staked. Time will tell, but it is a concern to someone developing today. Some models will not work well with such a volatile currency.
Another issue with Ethereum is the beaten and dead horse (as EOS fanboys proudly point to) of transaction times. As Cryptokitties proves an unintentionally (but ultimately stupid) malicious actor can clog the network and make transaction times unbearable for certain business models. However, the transaction time is a problem currently under heavy development. It’s almost certain that transaction times will decrease. Sharding, Plasma, Casper (PoS), Raiden, and other systems are all under active development to create more bandwidth for transactions. But there’s still a problem. The issue comes down to governance. What happens when an unintentionally large volume of transactions hits any system with finite scalability? It will bog down. When your business model is dependent on fast transactions this can be a problem because there’s no guarantee of bandwidth. And even if there’s a solution to the issue Ethereum may find that as this blockade is unintentionally encountered more and more it will scare off application developers on the thought of it interrupting their service. This isn’t to say that there are no conceivable business models that can use Ethereum, simply to point out that there are limits.
EOS on the other hand has some of its own bugaboos. **First and foremost, it’s not done yet. It hasn’t been released. Everything we say and do about it is purely speculative. Secondly, block.one places its headquarters in the Cayman Islands. Ok, fishy (ignore the pun) but not a deal killer. Then there’s this aggressive legal wording on the EOS.io’s FAQ that paradoxically discourage any investment at all. **
“block.one is building the EOS.IO Software but it will not configure and/or launch any public blockchain platform >adopting the open source EOS.IO Software (the “EOS Platform”). Any launch of an EOS Platform will occur by members >of the community unrelated to block.one. Third parties launching the EOS Platform may delete, modify or supplement >the EOS.IO Software prior to, during or after launching the EOS Platform.
The EOS Tokens do not have any rights, uses, purpose, attributes, functionalities or features, express or implied, >including, without limitation, any uses, purpose, attributes, functionalities or features on the EOS Platform”
Block.one is the company currently developing EOS. This briefly translates to “We won’t commit to doing anything, the tokens are meaningless and have no value.” If that doesn’t raise your scam flags I don’t know what will. To make matters worse this little gem follows:
“It was decided that U.S. citizens, residents and entities should be excluded from purchasing EOS Tokens in the token >distribution because of some of the logistical challenges associated with differing regulations in the many states of the >United States of America. block.one does not believe that the distribution of EOS Tokens or the EOS Tokens themselves >are securities, commodities, swaps on either securities or commodities, or similar financial instruments. The EOS >Tokens are not designed for investment or speculative purposes and should not be considered as a type of investment. >Nevertheless, U.S. citizens, residents and entities should not purchase or attempt to purchase EOS Tokens.”
Nothing very clear has been stated as to why block.one would choose to shoot itself in the face this way. Please, no really, correct me if I’m wrong. I’d like this not to be an issue. In fact, if it weren’t for the fact that the project is the brainchild of Dan Larimer I’d have cleanly written the whole thing off and so would many others. Even so, I must admit I clench a little every time I read this. Of all the ICOs I’ve seen this one ranks with the obvious Nigerian Prince e-mail scams of yesteryear except that metaphorically we know an actual Nigerian Prince is behind it. If it’s a scam it’s brilliantly organized, if not wtf? I could not knock anyone for ignoring EOS thinking it a scam. That said there is a little light at the end of the tunnel. As previously stated Dan Larimer has been deeply involved with two other successful platforms (including the one you're reading this on) and seems to be roundly committed to EOS. Block.one is also withholding 10% of the total number of tokens during the ICO so there is some reason to believe they want their tokens to be worth something eventually. A partial explanation might lie in the fact that it’s well known Dan is a bit of a fan of Liberty. Personally, I feel his thoughts on this may venture a good way into wild paranoia but I digress. Perhaps they're afraid the US government is going to feel threatened by EOS and try to disrupt it. There may even have good reason to believe this is the case. Still, my own research hasn’t brought up anything very conclusive on why such a strong legal middle finger was necessary and the whole project reeks because of it.
EOS’s other general downside has to do with the ideology behind it’s tokens. Staking in EOS means committing a certain percentage of the total chain storage be used to hold the state of some application for a period of time. I think the options presented were 1 year, 2 year, or 3 year stakes. As such application developers need to pay for (rent or buy) tokens to operate their service for that time period. There a good number of business models that benefit from this. If you know that you’re going to generate a certain amount of blockchain transactions and you need to assure that they happen in a timely fashion this is a model that prevents unintentionally malicious actors (cryptokiddies) from hogging all the block time. However, this also requires a significant upfront commitment. And it’s not entirely free entry. You need to pay for something to use the network at all from the perspective of an application developer. And application developers are very important for EOS. To use it more and to scale for growth the network demands you buy or rent more tokens. I imagine token holders will be willing to give away free block time to those asking for it at first, but when EOS is mature this may be harder to come by. Ethereum by comparison requires you to present a very tiny amount of ether to create smart contracts. But once so created the transactions sent to them are paid for by the users and not by the application developer so your network of users in Ethereum is boundless.
Ethereum is either actively developing or has certain plans being considered to address many of these concerns. The overall factor for its success depends on its ability to quickly develop solutions to problems as they arise. As true as this is for any cryptocurrency it seems more so for Ethereum. As an operating system the ecosystem of the applications that depend on it are a direct result of the base Ethereum layer working in a predictable and amicable way. I think the biggest unaddressed issue with it right now is formal governance. Without formalizing a process to exclude malicious or faulty code poisoning its environment, Ethereum is likely to face many more DAO-like events. The current Parity debacle is more evidence of this. A formalized process could add a degree of certainty for developers concerned about what happens when there are mistakes. And there are always mistakes.
On the other hand, EOS addresses many “problems” with the Ethereum architecture but it’s important to point out that these are problems only from certain perspectives. If it can launch properly and do even half of the things its proposing it’s likely to be a successful platform. And though transaction time factors, it’s the model for the token that makes this system so interesting. But an Ethereum killer it is not. The biggest factors in its way are its very shady legal footing which effectively excludes it from one of the most legitimizing and lucrative world markets from participating. And the fact that upon launch (if it launches) it is useless without a cadre of useful applications. Everipedia and others have announced that they plan to develop on top of EOS but the launch process is complicated so time will tell. EOS has no value without a good app or two. Although you might say that’s true for Ethereum as well I’d have to disagree. Ethereum can work just fine all on its own as a Bitcoin replacement. It is, afterall, basically just Bitcoin with smart contracts. But EOS is built specifically for applications. If it can’t lure worthwhile work for the network to do then people will lose interest in running the network at all.
In the end, the truth is that both models satisfy different needs and there will likely be business models that lend themselves more to one network over the other. Things that don’t suffer too much from volatility and depend on a blockchain that extends well into the future will find a happier home on Ethereum. Where, once a transaction is committed, it’s state will remain forever. However applications that require a guarantee of processing time and access to a certain number of transactions/s will be better placed in EOS.