Emin Gün Sirer x Haseeb Qureshi: Unscripted [Avalanche Summit NYC]

Great question. So when we started Avalanche about six years ago, that was late when compared to where crypto is. And back then we were still in the place on the timeline where people were selling dreams. It was the hopes, dreams, and vague visions time. And at that time we decided we know how to achieve these dreams. We now have the technology to build what these people are promising, what other people are striving to achieve. We know how to create today. And I think we've demonstrated to the world how to build lots and lots of chains in parallel that can absorb infinite growth. We're the only multi-chain system. We're the only ensemble of chains that can absorb the growth that is to come. We're the only system that can allow people to build different systems according to different rule sets. Now in the current paradigm, the time has shifted to one of business integration. It is time to take blockchains from being a technological curiosity to being deeply embedded in business workflows. It is time to go institutional and that's what we're all about. And that is going to be an entirely different approach. Our success from now on is not so much dependent on some fancy technology. It's not dependent on some fancy crypto mumbo jumbo. The days when you know so and so would write some kind of a blog post with like Greek letters in it and that would sway the markets, those days are gone and thank God they're gone. We don't want to be living through that stuff. That was all BS in the first place. So what we really needed were robust systems that could absorb the growth. We needed systems that could allow multiple chains to build in parallel and work together. We've got that. Now it's time to build together. And for that I think we have to work as a group, as a community, as a set of institutions that come together and take the benefits that we know to both our own institutions and to the world at large. And we all know how much more efficient blockchains can be. And we all know how much better we can make the world. So that's the next phase.

So, that's a polite way of putting things, I guess. So when you're building things, there is always the inclination, the desire, and sometimes the pressure from the market to cut corners. You can build things that go a little faster. Does your Pinto really need the liner in the gas tank? It probably doesn't, right? But then it kind of does. You know, when you get rear-ended, you don't want it exploding. Does your Boeing jet airplane really need redundant stall warning systems? You could probably do without until you can't do without. And the same is true for distributed systems. Now, maybe the best way to try to explain this is you can easily build systems that achieve speed by letting go of safety. For example, you can build robust systems that when they're out of their operating envelope, they will tend back towards getting stable again. For example, to give you a more visceral example, planes have their wings tipped at a V so that when they tilt to one side, they naturally tilt back to a regular flight level position. That's the kind of thing you want your blockchain to do. But instead to get speed, you can cut corners and you could build things so that they fly just fine, but at the smallest perturbance, they just go careening out of balance. And we saw that with some of our competitors. There was a year when one of our competitors that we love dearly, they're good people, their chain was stopping every two weeks. That can easily happen. And that's what happens when you give up on safety and security in order to get that extra little oomph. We were always on the side of sound scientific advances and sound engineering where the core tenet was reliability and self-stabilization. So for all of those reasons, I think we've had a tremendous run until now of fairly stable and very reliable network operations. And I'm very proud of our record on that front.

So yeah, I used to work on operating systems. And as some of you might know, the time from an operating system release to the next one is often about 10 years. So if you and I come up with a new OS idea, it'll take 10 maybe 20 years for it to make its way into Mac OS, into Windows, Linux, etc. So that was what I was comfortable with, just so you know who you're dealing with. So it's okay by me if it takes that long and I will be there to see it through. So that was my sort of mentality going into this. And so when I saw this new field of blockchain and I saw the hunger for science, the need for new tech and the speediness of adoption, I was actually thrilled. That's amazing. You never see this. I would write a blog post. In fact, I remember writing this one post about the DAO hack. And I remember editing it in real time. And you know when you're editing a Google doc and people are viewing it, they get like these animal names like artwork, beaver, whatever, camel, etc. And so there were like hundreds of animal names on top of it, they were literally waiting on every little new paragraph that was being written. And people are that ready to adopt what we were doing. So it's a fascinating, happening field and that's what attracted me to it. And so it's a wonderful thing to be at the forefront of developments that way. And I think we benefited from that a lot. And it created a mania as well, right? And a lot of undeserving people also benefited from it. So you could also make up some random stuff. You could take a well-known invention, put a little bit of a flare on it and say, 'Hey, I invented a new consensus protocol.' The same way you can take a hat, put a feather on that, you invented a new hat. No, it's the same damn hat. Oh, you just put a feather on it. So, it did create that mania. It allowed our field to get a little hotter than it should have been. But it did fund a lot of development and it brought us to where we are today. But moving forward, we can't count on that because there's another game in town called AI that sucked out a lot of the liquidity where a lot of other funny games are being played and that's okay. It is now incumbent on us to take the value that we've created and the value it creates to retail, to institutions, and to everybody who can benefit from it. And that's going to be a different set of challenges and it's wonderful. We can't rely on the old tricks working anymore. And if it takes a while for institutional adoption, it takes a while for institutional adoption. It is absolutely okay. And I'm super happy to report that institutional cycles are nowhere near as bad as what I started out with in operating systems. It's not going to be 10 years. So it's a lot faster.