Most people live in yesterday. Only few live in tomorrow. - xh3b4sd

I was today's years old when I understood the ETH value accrual thesis that the Ethereum researchers are pursuing using blob space. And the way to explain it is to understand a statistical principle called the Law Of Large Numbers. A better way to put it would be "the law of many samples", because what the Law Of Large Numbers tells us something like the following. The average of ever more observations will reflect the true value of a system ever more precisely.

Here is a classic example for approximating the average result of dice rolls based on many observations. In plain mathematical terms we would describe the average result as (1 + 2 + 3 + 4 + 5 + 6) / 6 = 3.5. By approximation the average of the first couple observations will be far off that expected value. But with ever more dice rolls, the average result adjusts ever closer to the true value. More samples give us a better idea of the underlying intricacies of the system at hand.

We can think of many more examples in which the average of many samples gets closer to some underlying truth. A coin flip may be not so fair in a single toss, but heads will eventually come up just as often as tails does. Measuring the size of animals can provide sufficiently reliable estimations about a species. The insurance and gambling industries make heavy use of the Law Of Large Numbers, as well as everything involving signal processing, because approximation by many samples allows us to reasonably predict the output of a system due to ever decreasing error variations.

Understanding that increased scale of a system provides more predictable error rates can then help us to understand how a relatively small system today may perform as a big system tomorrow. And that brings us to the realization I just had a couple of days ago. Ethereum's focus on blob space turns out to be very much aligned with high levels of ETH's potential for value accrual.

When we talk about blob space in Ethereum land, then we talk about data availability, or short "DA". Data availability is a very technical way of saying that the data that blockchain networks are supposed to verify, is guaranteed to be there in the first place. Because, if data does not exist, it cannot be verified. And because all of this is about blockchain networks, what really matters is the quality of the underlying block space.

Not all DA is created equal, and so some forms of DA will become commoditized, but some forms of DA will not. The commoditized versions of DA will be relatively cheap and relatively weak in the sense that more or less trust assumptions will have to be made when using those cheaper versions. The superior form of DA will always be limited, because it will always be natively integrated in the underlying system. Ethereum's native way of providing data availability for rollups is now described as blob space, a separate fee market to store rollup data. Justin Drake mentions once in a while that DA will never be commoditized. And what he means by that, is that the most secure DA will never be commoditized, because Ethereum's ability to provide data availability is simply limited, even at larger scale.

Now that we have a lot of the fundamentals out of the way, let's get to the point where value is in fact accruing back to ETH the asset. The demand driver for DA is always in parts related to the underlying security context. And even though blobs provide a superior form of data availability, blob space is surprisingly free for the most part, up to a point. And that is just a very big head scratcher for a lot of people out there. How can you possibly make money by giving something away for free?

Well, Ethereum doesn't reinvent the wheel here. There are several common economical principles that we are leveraging in order to generate demand and capture value in the process. We can summarize most of those principles using the concepts of freemium margins and network effects. And the calculation is pretty simple once we understand one thing: all free blob space will always be used. That in itself creates network effects, which in turn causes block space contention, which is just a fancy way of saying that everyone wants to post data at the same time. Once everyone posts data at the same time, blobs get "full", and Ethereum experiences usage spikes that are then revenue generating. Those usage spikes appear to be meaningless today, because we only have 3 free blobs per block. So far I was also of the opinion that blobs will not generate meaningful revenue ever, because we barely ever go over that freemium range in blob usage. But I was missing an important detail in the equation: scale.

The usage patterns of blob space are in fact fascinating to visualize. When we check blob space usage on some Dune dashboard, then all we see is a more or less flat line around the free range of currently 3 blobs per block. And with that little fluctuation there is barely any meaningful revenue generated today. The trick about the blob space strategy is simply scale, which is the way in which Ethereum is going to leverage the Law Of Large Numbers in its favour. Because scale increases the absolute size of a relatively small value.

Here is a nice way to think about it. Blob usage causes contention spikes above the water line. The water line of a small pond is pretty flat. But the water line across the deep ocean can become rippled with pretty big waves. Orders of magnitude more economic activity may burn billions of USD in blob fees alone every year, even if most blobs are free, because deep waters make for big waves. By my estimation, Ethereum could eventually burn over half a million ETH tokens consistently every single year. All estimations aside, more economic activity sampled over longer periods of time will inform us ever more accurately about the potential output of the underlying system.

And just to hammer that point home, almost all economic activity on top of Ethereum L2s will always be almost free, all the while we have the potential to generate billions upon billions in revenue flowing back to ETH. The reason that this is likely to work is because all the ocean of freemium blobs below the waterline remain free. So the vast majority of consumers can be served using superior bock space, while paying almost nothing for it. But that will only work if we generate economies of scale so that the little contention we produce pays for everyone else's free lunch.

Additionally, the costs to use L2s will remain in their own equilibrium because of the amount of free blobs per block. And then, L2s too will scale even further by themselves. They will become more efficient over time, and in the process become better at serving more users. Just like Base is scaling 1 M/g per week.

It is only the contention above the free range that causes price discovery in the blob markets. And the larger the amount of used free blobs, the larger the contention in absolute terms above the waterline, while keeping the relative size of blob usage spikes relatively small.

Another way of visualizing those principles of scale would be something like this. Compare the skin of an apple with the crust of the earth. The apple is small, and so is its peel. Planet earth on the other hand is far larger and therefore its crust is also way thicker. The relative sizes of the apple skin and the earth's crust are very similar respectively. They are proportional. But the absolute size of the crust of the earth is many orders of magnitude larger.

Today we are burning blob fees to the extend of the apple skin. Tomorrow we may remove far more amounts of ETH tokens from circulation by way of burning blob fees, but only if the economic activity on top of Ethereum uses hundreds or even thousands of free blobs per block one day. This is why the security context of Ethereum's blob space is so incredibly important. Because this very security context is one of the most important drivers of block space demand in Ethereum. Most people out there do still not understand why decentralization and MEV matter so much. By my estimation, Ethereum gets it right already today. Time will tell how any of this will play out. Maybe the reason why we can see all of this here so clearly before everyone else, is because we are already living in the future. I mean, just maybe.

The number of the week is simply 1. And that is because of something that nobody is telling you about. What nobody is telling you, is that Ethereum, despite being "abandoned", is once again number 1 in daily and weekly onchain volumes. That is a reference to the stark activity decrease that Solana is now experiencing again as the latest hype chain falls out of favour again. Turns out, crime wasn't the answer after all.