The Paradox of Decentralization

The bear market is upon us and the mood in Web3 is subdued, but far from defeated. While these market troughs bring with them some nervousness and anxiety, they are also excellent opportunities to invest in the future. It’s in these cycles of market-culling that the innovations of the future emerge, and so it only stands to reason that a reflection of what worked, and what didn’t, is worthwhile to say the very least.

What have we learned from this most recent market rise and fall?

Perhaps the biggest takeaway that I’ve seen is that Web3, in its current nascent state, has an infrastructure problem. Or, to be more exact, it has a lack-of-infrastructure problem. When we look at centralized models and Web2, it’s easy to forget that those systems built themselves up over decades. They engaged in trial and error repeatedly, and through many market cycles. They found work processes and infrastructure arrays that work, but only after many hard-learned failures.

And this is where Web3 has something to glean from the Web2 world.

Why reinvent the wheel at every turn?

Web3 is different, and it’s fair to say that we’re pretty much all aligned with how great it is, it’s power to change the world, and the utility of the blockchain. Importantly, though, we cannot let this zealotry lead us to trying to reinvent the wheel at every turn. Businesses have existed for eons and Web2 has existed for decades. We should not emulate what they do, but we should find shared problems and draw on their successes, where reasonable.

After all, so many of the stumbling blocks of Web3 — inefficiency, coordination misalignment, cumbersome economic hurdles (ahem, gas fees) — are not here to stay, but they’re definitely here now. How can we tackle these as human problems, rather than treating them like new and never-before-seen?

Chase Chapman summed it up perfectly in a recent tweet:

https://twitter.com/chaserchapman/status/1533829349836828673?s=20&t=9EKWIeyFOwtbus1QHbt8pA

“The hardest problems in DAOs are still human problems

Which is why I’m much less interested in the “autonomous” focus

It’s not that it isn’t important or exciting

I just don’t believe we have enough data about what actually works to automate things”

It’s really an excellent point, and one that most Web3 acolytes may tend to shy away from confronting. The point can’t be belabored enough, however: we need to have an open and honest discussion about governance.

Governance in Web3 is a major work-in-progress

Governance models and their tokens currently represent an excellent case study in coordination misalignment. Tokens in many cases are far too speculative, and their users are not always on the same page regarding their utility. In the same organization you will have some treating them as securities, and some treating them as paths toward governance, and this is happening oftentimes regardless of what DAO mission has stated or intended.

Furthermore, in many cases these tokens are easily gamed. Whether this is because of massive arbitrage opportunities between DAO token exchanges and AMMs/outside LP exchanges, or because some members are exercising disproportionate voting power, something needs to change. By the same token (pun so intended), giving massive voting power to members who happened to be at a DAO early — voting power that in many cases is totally unattainable by new members just a month or two later — calls into question the nature and utility of engaging in voting in the first place. At that point the system tends toward centralization automatically, which flies against the very purpose of a DAO in the first place. After all, how can we call it open governance if some members wield unmatched voting power, while others have little to none?

As much beauty and elegance as there is in the ideals of a DAO, there is also certainly a lot of room to grow. Whether that means changing how token issuance happens so that early contributors are not overly powerful in the proposal process, or in changing how teams work so that efficiency of work can increase, there are many tweaks that we are likely to see as the DAOsphere forges onward.

On the non-token side of things, DAOs have a very inefficient onboarding and project-assignment workflow. The question of who is running projects for a DAO is not easily answered, and certainly does not follow any Web2 or IRL schema. This is ostensibly a good thing, but it also has its drawbacks. The fact that Web3’s hard skew toward anonymity lets just about anybody turn up and try to run a project means that a lot of overlooked people will get a great shot at success. But it also means a lot of people wholly unqualified for leadership roles will find themselves in one, trying to lead a team of people who in many cases are far more qualified than they are. Not knowing when which of these cases is happening, and particularly when dealing with highly technical domains such as software engineering, means that traditional oversight cannot happen in any enduring way. And that’s damning for a DAO’s growth and spirit, and has undeniably contributed to the undoing of so many great DAOs that have come and gone.

So, what can we learn? Is Web2 unfairly maligned in the Web3 world?

I think a central tenet we have to be comfortable with is this: we can learn from Web2. It’s okay. It’s not a tenable position to reject everything from Web2 as if it never amounted to much. Let’s be honest: it’s amounted to a very successful behemoth. While we should be careful not to repeat the same mistakes, we also can’t throw out the baby with the bathwater.

I think Web2 processes such as assigning leadership roles, tracking member skill sets before putting them in a project manager role, and aligning around a mission statement that is fixed, rather than floating vote to vote, are all Web2 constructs that can, and will, help us. It can be argued that they are trends toward centralization, and quite frankly I think that argument is probably true, but as Web3 founders we must be honest with ourselves that centralization is not a dirty word. Don’t get me wrong, there are absolutely problems with the socio-economical centralization trends we’ve seen dominate Web2, banking, and a variety of other fields, but arguably these overreaches are just as much, if not more, problems related to human greed and vice than to centralization specifically. Web3 should never resemble Web2 in terms of centralization-as-a-business-model, but that doesn’t mean it can’t learn from some of its strategies for success.

So are we back where we started? Not quite.

This is why I’m calling this the Paradox of Decentralization. I believe that Web3 has a great outlook toward the future, and that we will all go there. But I also think it will take time, and many iterations. It will be slow approximations toward efficiency, not one swift leap. However, I do think we can speed up the process, particularly if we look toward infrastructure and work process models that are tried and true, albeit not native to Web3 currently. The only certain thing I can say is that, if we continue to try to reinvent the wheel at every turn, we’re always going to feel a step behind where we could be.