In the future... ICOs will run on EOS!

I wanted to post a quick follow-up to my post comparing EOS and Ethereum and focus on one specific issue: Initial Coin Offerings (ICOs). It's worth spending some extra time thinking about how platforms like EOS and Ethereum handle ICOs, since such coin offerings make up a significant portion of the demand for smart contract platforms (lately at least). If you are interested in a comparison of EOS and Ethereum that focuses more on issues related to running generic Decentralized Applications (DAPPs), feel free to check out my previous post. Here, I'm going to focus specifically on applying the different features and limitations of the two platforms to ICOs. Needless to say, this post is biased, but I like to consider myself pretty open-minded. Feel free to post any comments or criticisms in the comment section below! :)

Image source. AFAIK fair use under satire/parody aspects of copyright/trademark law.

EOS has all the features to be the dominant platform for efficient, reliable, and innovative ICOs.

Specifically, the key features that set EOS apart from Ethereum, especially with regards to ICOs are:

Free for users
Protection from denial-of-service attacks
Opportunity for innovative ICO structures
No need for placeholder coins (apps can run directly on the EOS network)

In my opinion, these four issues specifically make a very good case for launching your ICO on the EOS network as opposed to the Ethereum network. Hopefully after I elaborate slightly on each issue you will see the advantages of EOS too!

EOS allows startups to design ICOs that are completely free for users!

For me, as a potential investor, this advantage just makes sense. In order to make wise investments, we already have to do our due diligence and identify potentially valuable opportunities, judge the state of the markets, etc... There's no reason that we should also have to worry about whether or not the amount of fees we attach to our investment will be enough to convince miners to accept our transaction, or if our fees will be too small that we totally miss out on the opportunity to invest! This is a no-brainer! With EOS, a startup that wants to hold an ICO can simply purchase a modest amount of EOS, which entitles them to a certain proportion of bandwidth, storage, and computing power on the EOS network. With that bandwidth in hand, that startup company will never have to worry about any fees on the EOS network at all. They can design truly innovative ICO models, and ICO participants will never have to worry about paying fees to invest. In addition, because of the robust, reliable block production thanks to the delegated proof-of-stake consensus mechanism, ICO participants also have to worry much less about the timing of their transaction and whether or not it will be included in a certain block. Compare and contrast these features to recent ICOs on the Ethereum network. There are several very well-known examples of recent ICOs where willing investors literally could not participate because of the fee structure surrounding the block production on Ethereum. When the Ethereum network becomes overloaded with transactions, only very wealthy investors can afford sufficient transaction fees to ensure that their ICO contributions are accepted. Of course, this failure is also partly due to the design of those ICO models, but as I will describe below, EOS also greatly improves the flexibility with which ICOs can be designed.

Protection from denial-of-service attacks

This issue is absolutely crucial, and it is one that can significantly sets apart EOS from Ethereum, especially in terms of ICOs. The simple fact is, the sheer network capacity of the EOS network will likely exceed that of Ethereum by several orders of magnitude. That makes it much more difficult to execute and sustain a denial-of-service attack against the network. Actually, this point requires some clarification: It is not possible to execute a denial-of-service attack against the network, per se, but only against certain applications. This will be more clear when we look at the Ethereum networks model for applications and ICOs in general. Because all applications, including ICOs, essentially have access to the same network bandwidth on Ethereum (subject to a dynamic fee market), if a denial-of-service attack is executed against a single application, it can effectively freeze the entire Ethereum network. In fact, it's not even required to have a malicious attack in order to freeze the network. Several perfectly legitimate ICOs have simple gathered so much interest, that those applications effectively bottle-necked Ethereum and froze out other applications. However, on EOS, this is not a risk factor. In fact, multiple competing ICOs can launch simultaneously on the EOS network without any risk of impacting each others network bandwidth. Individual applications/ICOs may suffer bandwidth bottlenecks if they do not hold sufficient EOS tokens, but there is no risk to the network itself. With this in mind, it would probably be wise for those wishing to hold ICOs on the EOS network to acquire sufficient tokens before launch to guarantee sufficient peak bandwidth for the ICO. This extra, temporary investment in EOS tokens can then easily be resold, unlike the ETH tokens required for participation in ICOs on Ethereum, which are simply burned as fees to the miners.

Opportunity for innovative ICO structures

EOS also opens the door for significant innovation in providing equitable, fair, and unique ICO models. This type of innovation is somewhat limited on the Ethereum network by the slower, less reliable block production and by the dynamic, unpredictable fee model that goes along with block production. For example, suppose you want to hold an innovative ICO model that rapidly auctions off individual tokens using millions of micro-transactions. Or suppose you want to dynamically distribute tokens based on some non-trivial, computationally intensive process. As far as I understand, it would be very costly, inefficient, and/or impossible to implement these types of ICOs on the Ethereum network. The requirement for gas fees makes it prohibitively expensive to innovate in the computational distribution of tokens, and micro-transactions are simply non-viable during the peak of an ICO when required fees could be orders of magnitude larger than the ICO contribution itself. However, because holding EOS tokens entitles the ICO application to a certain amount of network bandwidth and computation, ICO models on EOS can be designed with innovative, computationally intensive algorithms, and they also make possible ICO distribution models that rely on micro-transactions. Disclaimer: This is pretty much speculation on my part, but I think I'm probably pretty close to correct... haha.

No need for placeholder coins (apps can run directly on the EOS network)

Finally, one more slight speculation on my part, but I believe the EOS network can eliminate the need for placeholder ICO tokens. Many ICOs, such as the EOS ICO, have chosen to adopt a model in which the ICO is held using a placeholder token on the Ethereum network. For example, the EOS ICO has an ERC-20 placeholder EOS token on the Ethereum network, the distribution of which will be frozen at the end of the ICO period. The reason a startup might do this is that they want to ensure a fair, auditable distribution of tokens, but they don't want to actually USE the Ethereum network to run their application/business. For example, EOS is using the Ethereum network to manage the ICO distribution because they want an extended, fair, and auditable distribution, but the actual EOS network obviously cannot exist on Ethereum because the block production is far to slow and unreliable, and because of the fee structure. However, with EOS, the network is so powerful, reliable, and scalable, that blockchain startups that choose to launch their ICOs on EOS can also confidently build their entire application directly on the EOS network by simply purchasing a certain amount of bandwidth and computing power. They never have to worry about the network freezing from denial of service attacks or failed applications (i.e. the DAO), and they never have to worry about a dynamic fee structure pricing them out of business. In fact, as far as I can tell, Ethereum itself could be run on the EOS network.

Perhaps I have not stated these four points as clearly as possible, and I'm sure there is a lot more to be said on both sides of the argument, but I believe it's safe to say that EOS offers several innovative features that simply are not available on the Ethereum network. Feel free to comment below with comments or criticisms!

All the best,
Trogdor :)