14 Sep 2019
Ethereum has attracted the world’s largest decentralized applications ecosystem behind it, with an estimated number of participating developers in the range of 250,000-350,000. This growth has been nothing but spectacular from a global and diversified reach perspective, only five years after its launch.
At this footprint magnitude, it is not surprising that Ethereum encounters some attrition from developers or projects for a variety of reasons. In startup terminology, the rate of attrition is called “churn”, and it simply refers to the % of users that leave. If a company wants to continue increasing the adoption of its products, its usage growth rate must exceed its churn rate. In human resources, this is also called “employee attrition”, and for mid-to-large companies, up 5-10% attrition rate is considered generally acceptable.
With this background in mind, let’s examine Ethereum’s churn and its significance (or lack of), both from a quantitative and qualitative perspectives.
At the quantitative level, Ethereum’s churn is apparently insignificant, just because its ecosystem continues to grow at quite healthy growth rates that are much higher than the reported churn along the way. Less than two years ago, the Ethereum ecosystem was estimated at about 100,000 developers. Today’s 250,000-350,000 numbers are a clear indication of net growth. At the most recent ETH Boston hackathon event that took place Sept 6-8 2019, 13% of the attendees were new to Ethereum, and another 21% were beginners, together being indicators that a lot more developers are coming to Ethereum than those who are allegedly leaving it. To get a sense of the continued growth of the Ethereum developers ecosystem, just follow ETH Global to track the impressive attendance and participation numbers at these events around the world. As I wrote previously, Ethereum meetups, hackathons and ETH Global events are the envy of other aspiring blockchains in that space.
It is not surprising that, according to Techstars, 95% of the blockchain projects applying to their program are building on Ethereum.
From a qualitative perspective, we tend to hear comments similar to this:
“Developers are immediately seduced by ETH promises, but they leave after a few months, at least those who actually tried to build on it. The projects that work run into trouble. It’s great for prototyping but not commercial release.”
However, one must examine the real reasons and motivations behind these anecdotal occurrences. Despite a few publicized cases, such as those enumerated by this misleading article (The Unbundling of Ethereum), this is not a pattern, but rather normal attrition from those that didn’t find Ethereum to be suitable for them.
As it turns out, not only the number of projects that left Ethereum is relatively small, most of them share a common thread: not being a fit for what Ethereum is very good at.
The fact is,- many projects used Ethereum’s ERC-20 token generation feature as the lynchpin of their genesis, without thinking further too much about whether they would be able to actually develop on Ethereum. Generating an ERC-20 token is a very low-barrier type of activity, and succeeding in it doesn’t mean that you have either figured out or committed to developing on the Ethereum platform.
Some other projects wanted to treat Ethereum like a database and record every single transaction on it. If you treated Ethereum like a database, you would have run into processing issues that only databases are optimized for.
Several projects didn’t really require the underlying decentralization infrastructure features of Ethereum. Decentralization comes at a price of course, specifically as transaction processing speed is compromised in favor of maintaining the sustainable economic security of the blockchain.
Other projects didn’t know how to use Ethereum’s scalability to their advantage. Namely, you needed to become quite familiar with Layer 2 scalability options in order to grow your applications on Ethereum.
Another segment of projects ultimately wanted to be on a separate chain they owned, so they could derive their own economic benefits, so Ethereum couldn’t provide that option to them.
Even for those projects that left it (or will leave it), Ethereum has most probably given them good benefits whether it was to generate the economic token, develop a prototype, or issue non-fungible assets on it. And for developers, it allowed them to gain their first experiences in decentralized applications development based on cryptographic blockchains, whether they were able to prototype something or not.
The proverbial product-to-market fit process is inescapable for any product, technology, protocol, solution or movement. While Ethereum’s use cases and sweet spots are still evolving, and being refined as more of them get uncovered, we already know what Ethereum is really good at:
Generally, Ethereum is an excellent platform for managing relationships that can be (entirely) mediated by a universally trusted and openly decentralized blockchain that enforces the logic behind these connections.
Finally, when it comes to Ethereum’s most criticized weakness, its transaction processing speed, one needs to factor the existing Layer 2 capabilities that expand Ethereum’s transactional throughput beyond its on-chain’s limit of 15 TPS. Fast forward a few months from now, after Ethereum’s next hard fork, the Ethereum chain could reach up to 4,000 transactions/second via ZK roll-ups, shadow chains and chain witnesses. And once stateless clients are implemented, it will unlock a lot of built-in latency as clients will not be required to store large state data, instead they would just have to verify Merkle branches.
More and more, Ethereum is at the center of a re-definition of the web programming stack, as I previously noted that Ethereum’s stack is one of its biggest strengths.
Ethereum’s criticism is often associated with a criticism of the Ethereum Foundation, its more visible artifact. However, just focusing on the EF is wrong because they are only co-ordinating perhaps 10% of what is actually going on within the broader Ethereum ecosystem. The rest happens on its own. When Ethereum got started, its Foundation was the key entity that was responsible for initially developing most of its core functionality. However, gradually, the Ethereum ecosystem started to grow around its Foundation to the point where the current estimated activity ratio is 10/90 in favor of a vastly decentralized ecosystem of contributors, influencers, developers, investors, research groups, companies, projects, and a variety of diverse stakeholders.
The end result has been, as Albert Ni of the Ethereum Foundation put it aptly: “independent flourishing within a massive, diverse ecosystem, as one of Ethereum’s superpowers”.
That didn’t happen by luck.
At the August 2019 ETHBerlin Zwei conference, Albert Ni explained Ethereum’s dogmatic “subtraction mindset” (versus the addition mindset), a direct contributing factor to its sprawling ecosystem. According to Albert, a subtraction mindset creates options, and allows them to remain focused on always distributing opportunities instead of capturing them.
To better understand the full context of how the Ethereum Foundation operates today, please listen to Albert Ni’s excellent presentation.
Maybe it’s just a parable, or maybe it’s partially true, but I had heard of this analogy during my previous years at Hewlett-Packard (in the early 90’s) in the context of getting indoctrinated in Hoshin Kanri, the Japanese strategic planning process. [Today, a simpler version of Hoshin has been popularized as the OKRs process (Objectives and Key Results].
The story back then referred to a Japanese plant that had a long seed gestation period, before it showed any signs of life above the ground. You had to keep watering it for these initial years, without seeing it grow visibly above the ground. Rather, the plant was growing its roots below the ground, first. The lesson here was that planning is like growing the roots first before starting to see visible results.
More recently, this analogy metamorphosed as the “Chinese Bamboo Tree” parable, and similarly to that mysterious Japanese plant, for the first few years there is no visible sign of above the ground activity, until finally in the fifth year, the Chinese bamboo tree starts to grow exponentially. Had the plant not developed a strong (yet unseen) foundation, it would not have been able to grow so rapidly thereafter.
The analogy with Ethereum is striking. Ethereum has been in this incredible formation period for the past five years, building strong community adoption roots as the basis for a solid foundation. Although a lot of Ethereum’s success is already visible, there is much more that’s ahead of it, not currently visible, but poised to suddenly grow further soon.
To further understand the impact of Ethereum and tie together the 10/90 concept with the Chinese Bamboo Tree analogy, one needs to understand that Ethereum is not a monolithic technology. The Ethereum Foundation is not a singular entity in the same sense that one would label a given organization.
Ethereum is not “one thing” that you can point at, shoot at, swing at, or pin down. It is a set of technologies, stack tools, infrastructure, currency, belief system and at least a quarter million self-indoctrinated developers that are stubborn enough to stick with it despite its imperfections because their collective actions are making it better in all of its aspects.
Ethereum has already sprawled into most of the nooks and crannies of what this new world of decentralized constructs might look like. Many other “competing” chains that are based on single feature theoretical superiority [e.g, speed] are not realistic about the omnipresent developers network effects that Ethereum already has.
If Ethereum fails, the whole crypto sector fails. I don’t think one can presume that projects behind Ethereum’s early lead would auto-magically morph or migrate unto other blockchain infrastructures that are aspiring to be “better than Ethereum”.
Will there be other successful blockchains that serve specific needs? Yes for sure. Has there been churn coming out Ethereum projects that were abandoned or didn’t materialize? Yes, of course (though the reasons need to be inspected and not generalized).
Ethereum’s ambitions and scope on the decentralization spectrum are orders of magnitude ahead of others, with the exception of Bitcoin, and certainly Blockstack who is carving itself a different but successful path along the data ownership / decentralization paradigm.
Ethereum’s Open Finance / DeFi footprint is real and not built on a house of self-referential use cases.
Whatever imperfections Ethereum or its Foundation seem to have, they are largely compensated by the sheer variety, vigor, vibrancy and scale of its ecosystem. Of course, in theory Ethereum can be attacked but in practice, and on the grounds of its reality, Ethereum’s lead is substantial. If you want to fight Ethereum, you are fighting its decentralized army of self-indoctrinated supporters that are self-motivated and committed to make it a success.
Don’t look solely at the Ethereum Foundation to judge the status of Ethereum. But if you do, you must understand the essence of its subtraction mindset which they have been perfectly executing upon for the past five years, and that is actually what caused Ethereum’s rapid decentralized to take place, and has become its best market defense mechanism.