Three Things #27: July 24, 2022

Three Web3 applications that are here to stay

Way back in February—that feels like it was ages ago, doesn’t it?—I wrote about the big three Web3 applications that are unambiguously useful and important. These are the three points I’ve been leaning on for the past couple of years when people ask me, “What’s it all good for?” I want to revisit this topic because it’s so critical, and because a few things have changed since February. It’s important to me personally because I’ve spent the last few years of my professional life building infrastructure to power these applications, and I intend to keep doing so for the foreseeable future. Also, it’s critical to revisit this question after a big market dislocation like the one we’ve just seen and the inevitable mainstream backlash.

So, what is Web3 good for? Even if the existing crop of blockchains and applications fail in the long run—and I’m not at all convinced that they won’t—I’m totally confident that these three applications are with us for good.

The first time I drove a Tesla, back in 2015, I was totally convinced it was the future of driving. It was 10x better than any car I’d ever driven before. I feel that way today when sending stablecoins or signing in using my crypto keys. Photo by Taneli Lahtinen on Unsplash

Thing #1: Dollar Stablecoins

The dollar may have its flaws, but these things are relative and, globally, the dollar is still king, now more than ever. USD is trading at or near all-time highs against a number of other major currencies. This is, quite literally, the result of enormous demand for dollars around the world, relative to demand for other currencies. For the unfortunate residents of countries with particularly bad monetary policy and economic mismanagement—take your pick as there is no small number of basketcases—the dollar is a safe haven, a place to park one’s hard-earned wealth to protect it from inflation. Yes, that may sound counterintuitive in a time when inflation is rampant everywhere, even in USD terms, but it’s far worse in other currencies.

Of course, not everyone in the world has easy access to dollars. That access is especially limited in the developing world. Folks in developing countries get clever when they need to—using the black market, opening foreign bank accounts and trading between them, trading mobile phone credits or gift cards—or, increasingly, using cryptocurrency. Bitcoin is a great alternative and, volatile as it is, serves as a lifeline to people experiencing the worst hyperinflation, but for everyone else it’s simply too volatile and is not yet mature enough to serve as a real inflation hedge. Enter dollar stable coins.

I wrote about stable coins last time, but I want to focus specifically on the least risky sort: fully-collateralized dollar stable coins.

Simply put, dollar stable coins are a “killer app” that’s better and safer than any of the alternatives, to get much-wanted dollars into the hands of people around the world. Of course, not all stablecoins are created equal—I wrote back in February that Terra and others are built on unsustainable ponzinomics—but there are safer options. USDT on Tron, a barely-decentralized scam network operated by a huckster, has taken off because, in spite of Tron’s flaws and Tether’s dodgy accounting, the combination is surprisingly functional and fees are low. Same deal for the Binance app, which has also taken off in certain markets. This shows that folks in a lot of places care a good deal more about getting “hard dollars” (on reliable networks with low fees) than they do about decentralization.

Finally, you don’t have to live in a banana republic with a hyperinflating currency to appreciate the utility of stablecoins. If you’ve ever tried to move money from one bank to another—with all the requisite paperwork, fees, delays, and arbitrary limitations—you’ll know by now that sending stablecoins is a 100x better experience. It’s bad enough trying to do a transfer within the same bank; sending fiat money between banks or, god forbid, across borders, is an absolute clusterfuck.

Stablecoins are here to stay. I don’t care what regulation comes down the line. The genie is out of the bottle and you ain’t putting it back. Anyone who’s experienced the usability of stablecoins is not going back to the bad old days. It’s one of those inevitabilities: once you’ve seen it, you cannot unsee it, and it’s just so obviously the way things were meant to be. Bitcoin, Ethereum, Metamask, smart contracts, NFTs, all of them can go away, but dollar stablecoins (and the infrastructure that undergirds them) are here to stay.

For more: Try sending some funds using a stable coin if you haven’t already. Send me a DM and we can try a roundtrip transaction.

Thing #2: Sign In With Ethereum

Identity management and authentication in a Web2 world is totally, thoroughly, completely broken. I struggle with it every day, and I’m a fairly technical user: I’m comfortable using a password manager, authentication tokens, even FIDO keys, and receiving SMS messages on multiple devices. For folks who aren’t as tech savvy the situation is much worse. I have multiple friends and relatives who still use the same password everywhere, write passwords on sticky notes and in paper notebooks, and are stymied every time their usual password fails to meet some obscure password format requirements. The state of this situation is far worse than I think most people realize. It’s a great example of technical demands outpacing human-centered design considerations.

It turns out that passwords are just a terrible technology. They were sort of okay a decade or two ago when people only needed to manage a handful of them, and when the things they secured weren’t that important anyway. Now that our lives are largely digital, and our most sensitive information is stored on third party servers and (in theory) protected by passwords, the situation has become totally untenable. Multi-factor authentication helps a little, but SMS isn’t much better than passwords. It’s clumsy and the UX is atrocious. And, of course, phone numbers get hacked all the time. Phone numbers were never intended to be used as unique identifiers, but application developers fall back on them for lack of any better form of sybil resistance. Authenticator apps, digital tokens, and FIDO keys offer better security but the former end up on centralized servers protected by, you guessed it, passwords, and the latter are expensive and won’t work for everyday users.

Well, now we have something better—much, much better. In the same way that sending a stable coin is 100x better than sending a wire transfer, logging into services using a keypair (rather than a password or a phone number) is a truly magical experience the first time you do it. It makes you wonder why everything doesn’t work this way, and it makes you realize that everything should and, eventually, will work this way. No password manager. No passwords. No silly password length or complexity requirements. No 2FA, no SMS, no authenticator app. Just pure, beautiful cryptography, wrapped in good UX.

The cypherpunks had a dream in the nineties that everyone would have a keypair, and that the public key would be widely known so that everyone could communicate and exchange information securely. It took a couple of decades longer than planned, but that is finally coming to pass. Metamask, Phantom, and other browser-based crypto wallets allow a user to manage multiple keys and identities, and to authenticate to applications without relying on any centralized, third party services. Where data are hosted on a third party server, they can be end to end encrypted to preserve privacy.

No matter what happens to cryptocurrency and blockchains, sign in with Ethereum and similar solutions are here to stay without a doubt. Over time they will become more user friendly, and we’ll have better tools to manage keys, back up private keys, restore them in case they’re lost, etc..

For more: Try using a service like Skiff.com. Seriously, try it. It’ll blow your mind, I promise.

Thing #3: DAOs (in theory not in practice)

I also wrote about DAOs back in February. Since then, my opinion has shifted a little bit. I feel about DAOs the way I feel about communism: I still believe in the theory of DAOs, but I’ve lost some faith in the practice of DAOs, at least for now.

The practice of DAOs, simply put, is messy. In my experience they tend to be disorganized, chaotic cooperatives with very little accountability. They tend to spend more time bikeshedding and debating internal organizational policies and procedures (i.e., meta work) than doing real work. They tend to be staffed by people who have little to no experience being part of high-functioning teams (which are typically centralized and hierarchical) in typical companies. They’re often full of people who are token-rich and are “checked out” from day to day business. Of course not all DAOs are created equal and I’m generalizing, but this is largely what I’ve personally observed over the past couple of years. In short, a DAO is a nice idea but it’s not usually the structure you want to get actual work done.

Having said that, in the same way that a died-in-the-wool communist will tell you “true communism has never been tried,” none of this messiness in any way discredits the idea of a DAO. It’s a powerful idea that’s still evolving and trying to find its footing, but I’m certain that it’s not going to go away, for two reasons.

Firstly, as I wrote about last time, a DAO can be a much more practical structure than the alternatives. The only existing tools we have to massively scale human collaboration are the corporation and the nation state. A DAO is easier, faster, and cheaper to spin up or tear down than a corporation (not to mention a nation state). DAOs are to companies as cloud computing and VMs were to bare metal hardware and colocated racks: in the same ways that cloud computing and virtualization made startups—and the process of launching and experimenting with digital products and services—much faster, easier, and cheaper, DAOs will have the same impact on human collaboration more generally. We’ll see (and have already begun to see) wild experimentation in designing and building new types of ventures and human institutions, from banks and insurance companies to no loss lotteries, prediction markets, quadratic funding, ponzi-casino games and much else besides. All of this is just the beginning.

Secondly, the idea of placing the ownership, governance, and economic upside of a project in the hands of a much broader group of people than you can with a corporation is very powerful idea: we need look no further than Bitcoin and Ethereum to see the first mega-decacorn success stories, and others will follow. Of course, as I alluded to above, this is complex and can go wrong in a thousand fancy ways, but I’m confident that this, too, will be solved with enough time and enough experimentation. Companies were messy in the beginning, too. The best DAOs are already learning from DAO failures, and from the best practices of companies, cooperatives, and governments to build much more efficient, effective governance structures and practices.

DAOs are not well suited to every use case. Part of the challenge is figuring out which use cases allow them to excel. They seem to be pretty good for investment funds (like the original DAO), for example, and they’re probably less good at managing complex, fast-moving, rapidly evolving technical projects. It’s unclear whether they’ll be able to offer compelling service and support to consumers—companies may always be better at that (just as they’re better at it than countries are). And it’s unclear how to scale them efficiently, or whether that’s even possible. We’re still in a phase where people are throwing DAOs at everything to see what sticks, just as they previously did with tokens. Some things will stick, and later we’ll look back and laugh at this ridiculous experimentation. While I don’t think companies are going away anytime soon, I do genuinely think DAOs have a chance of replacing nation states one day (and maybe forcing companies to get better at governance), but I guess we’ll see.

For more: Join a DAO. There are plenty to choose from. DM me if you need pointers on where to start.