Crypto Means Business

Ask Me Anything with David

Feature Image
Ripple is hosting a live Q&A with our Chief Cryptographer, David Schwartz


Livestream transcript below:

Asheesh: Welcome, everyone. I’m Asheesh Birla, I’m the head of product at Ripple. And today we have David Schwartz, also known as “Joel Katz,” chief cryptographer at Ripple. And we’re coming to you live from San Francisco at Ripple HQ. So, one, happy anniversary, five-year anniversary for the XRP Ledger. Also, big news last night, XRP hits one dollar. Big milestone for Ripple and the entire community. Also, we just want to say that we’re committed to doing more of these community events going forward. We did one just a few days ago with CEO Brad Garlinghouse, and we’ll be doing these live streams and other community events going forward. Before we get started, we have a special guest, Chris Larsen, Founder of Ripple, that wants to say a few words about David.

Chris: I’m sorry to interrupt, everyone, but on this five-year anniversary, I just wanted to express our awe and our incredible appreciation to David Schwartz. I think we all know the obvious. There would be no Ripple. There would be no XRP. There wouldn’t be this amazing company that we have today if it wasn’t for all the early work that David gave us. And I think we all know how incredibly smart he is. On a personal note, you know, I have to say I’ve never met anybody more kind, more generous. Anybody who is just the opposite of prima donna. And that is David. And for that, we just want to say thank you and happy anniversary. And with that, I’ll leave it to you guys.

Asheesh: Thanks, Chris. Alright, so our first question, let’s get going. From cryptfx Twitter Handle: “Why Joel Katz instead of David Schwartz? I’m curious.”

David: Well that dates back to my high school days. My friend John, I hope he’s listening, had a bulletin board system back in the days before the internet was accessible to everyone. People would actually call other people’s computers to leave messages and share gossip. And he asked me if I wanted an account and I said yes. And he asked me if I wanted to use my real name and I said no. I didn’t have a real reason why. I had no clear plan. But I thought, I don’t want this tied back to me, who knows what I’m going to say. So I told him to just make something up because he kind of put me on the spot, like what name should I use. So, he was a Ren and Stimpy fan. Most of you probably have no idea what that is. But Stimpy, who’s a cat, his real name is Stimpson J. Cat. And he thought, well, he’ll take that and kind of humanize it into Stimpson J. Katz, K-A-T-Z. And I used that name for awhile, but then I started to do more serious stuff, and the internet became more of a tool than something that you played with. And I needed a more serious name, but I didn’t want just to reveal my identity, so I went through a sort of gradual rebranding. So, I thought, well the J could stand for something, why not Joel. So for a while, I was Stimpson Joel Katz. And then I said, well, people who don’t like their first names sometimes turn it into an initial, so I was S. Joel Katz. And then I finally completed my rebranding to Joel Katz, which was a nice human-sounding name. Unfortunately, it’s not my real name, so I obviously had to reveal my real identity as I became more serious. But now I have two names, I guess.

Asheesh: We call him David at Ripple if anyone’s wondering. Now, so, next question, from Elliott Smith: “I’ve been following Ripple since 2011.” Ripple was founded in 2012. “I have watched some ups and downs for the company. What’s your best moment and your most difficult moment?”

David: Wow, so I started working on what became Ripple in November of 2011. And it’s been a crazy ride. I have to say my best moment is right now. Just the recognition that we’ve been getting from the community and the success that we have and the amazing team that we’ve built. It’s just really mind-blowing. Um, most difficult moment: So, we did go through some difficult times. As you probably know, we had an issue with FinCEN. We’ve had some lawsuits. It’s been, you know, tough times, but I think we’ve got all that behind us now and I think we’re positioned just so well for the future. I’m just so extremely excited.

Asheesh: Alright, we’ve got one from Paul Wong: “What do you want to see at 10 years of XRP?”

David: So, it’s hard to predict timing. We’re at five years now, and obviously five years ago there was basically nothing, so it’s hard to figure where we would get in five years. Obviously, I would love to see Interledger — which is Ripple’s protocol for atomic cross-ledger payments — being used with XRP as sort of the heart of it, as the way that payments settle. I think in five years we could certainly see that happening on a grand scale. As far as with XRP and the ledger itself, we’ve built in a lot of features like payment channels and escrow. We have order books and all these additional features that we’ve built and I would love to see more projects using that, more development on the XRP Ledger. I think also we built in a sort of, like a peer-to-peer lending system at the heart of the ledger, which I think could be extremely interesting for the future. It’s hard to say how far we’ll get in five years, though. But those kinds of things could be very transformative.

Asheesh: Alright, so that was it for the warm-up. Now, we’re going to go deep in decentralization. So, we’ve got one from XRPedia: “What type of companies will host lists of recommended trusted nodes? Bonus question, if you’re up for it: How many lists do you expect there to be in a mature, decentralized network?”

David: So, I think anyone who uses XRP and the XRP Ledger has a stake in the network. And I would love to see that stake reflected as more participation, whether that’s running a validator, whether that’s running lists of validators. I think it’s going to be … it could be exchanges. It could be financial institutions that are using XRP for settlement. It could be companies like Ripple. But we do need to see that sort of broadening. We need to see a broadening of stakeholders. And I’m hopeful that with the recent, increased attention that Ripple’s getting, largely due to the price in XRP, will be easier to attract stakeholders. Exchanges have limited what they’ve done to just their exchange functionality, but I would like to see them take more of a stakeholder position in the networks. Exchanges have interest. People who use cryptocurrency have interest. And I would like to see more active participation. I think we’ve seen recently that governance can be a big issue and it is important to have good governance, and that requires stakeholders to act, to decide what their interests are and to make those things clear. So, I would love to see a broadening of the stakeholders.

Asheesh: Excellent. Oh, one from one of my favorites, TplusZero, one of the only guys that retweets me. He’s got a good question here: “Could you share any updates on the XRPL gateway strategy. Saw Dash partnership recently. Which projects you’ll spend your day to day on in 2018. Thank you.”

David: We’ve kind of rotated away from the gateway strategy and focused more on liquidity from XRP coming from the other native ledgers. What’s the second part of the question? Sorry.

Asheesh: Oh, it’s a comment on the…

David: Oh … 2018. My big focus is going to be improving the decentralization. It kind of gets back to what I was saying before about the stakeholders. You need a network that’s not vulnerable to external threats, that doesn’t have centralized control because as you start to get large, as you start to move billions of dollars, there’s an incentive for people to attack that infrastructure and to compromise it. There are people who may find their interests threatened by it. There may be people who just don’t like it. And that broadening of the stakeholders is just extremely important and as I’ve said before that is going to be my main focus for 2018. We are very committed as a company to decentralizing the ledger and building things like compliant-payment networks and institutional systems on top of that decentralized ledger.

Asheesh: Ok, I love this next question. I think, David, you’re going to love this one, too. It’s from XRPtips. “Are there any plans to improve support for developers who may be interested in building on top of XRP?”

David: So, that’s been a big challenge for us. We haven’t really reached out to developers as much as we could. I think we have good documentation. But I don’t think documentation is enough. I think you need tooling. I think you need to reach out. You need to let people know you exist. And we have had a hard time doing that, but we have definitely refocused on it recently, and we’re looking for new ways to improve on that. So, I guess I can kind of throw that question back and say, what do you think we should be doing? If you want to develop on the XRP Ledger, what do you need from us? Do you need tools? Do you need encouragement? Do you need investment? All of those things are open we just need to figure out what’s going to be effective. There is definitely a shortage of Ripple developers, and I think the greater attention on Ripple and XRP will help to increase that. The attention on Interledger [Protocol] is helping because Interledger is a good sort of way to get into Ripple without having to interact with the XRP Ledger directly. But we definitely need to do more, and we need suggestions on what we should do because we have not paid as much attention to that as I think we need to. I think we definitely realize that now. So, give us suggestions, please.

Asheesh: Yeah, absolutely. And right before this, David and I were actually just jotting down some ideas on some developer ideas, so, yeah, shoot them our way. Alright, so the next one. TuneAtlas: “In the wake of Lee, Charlie Lee, leaving LTC, what’s your opinion on founders departing their projects and the impact this has on development? Are operations more vulnerable to leadership change when you are more centralized or are there contingency plans on hand for these events?” That is a deep one.

David: That is a deep one. So, I think we can unpack two issues there. So, one is founders that have a direct stake in cryptocurrencies. And I think the other one is whether having a sort of Jesus at the center of your project, a sort of messiah, a nice dictator, whether that’s a good thing and what kind of threats that that carries. So, on the first question of people involved in projects holding the project’s asset, I think that’s been the norm in business. It’s been the norm for employees to hold stock in companies, and particularly key employees and executives of all kinds. So, I don’t think that’s a bad thing. I think you want a person to have some skin in the game. I think also it’s important that people know that because that can easily be abused. A person can manipulate the news for their own advantage. Incentives are really well aligned at Ripple because XRP is the major asset that Ripple holds and a lot of Ripple’s value comes from its XRP, so Ripple’s XRP incentives are nicely aligned. But I don’t see much of a benefit from the head of a project sort of taking all of his skin out of the game. It’s kind of hard to understand what his motivation is. That’s not to say that there aren’t other motivations, just wanting to see your project succeed. But you do want an alignment of incentives between the people who are running a project and the people who are interested in the project, using it, and who put their money in as well. As far as this idea of sort of having a benevolent dictator at the center, I think it’s to some extent inevitable. I think projects do become a bit chaotic when they don’t have one. BitCoin doesn’t have any sort of central leader and I think that’s been a negative. It’s certainly not fatal, and I certainly think that we could see governance improve. But I think that as long as you have a person who represents the views of the stakeholders, that’s a good thing, as long as those people are free to stop following that person. One of the first things that I did at Ripple, and as we started a scale and as we started to get to be a larger company, is I decentralized myself. I took all of my responsibilities and I handed them off to teams. Which is great because it means I don’t get a call at 3 am on a Saturday, but it’s also important because something could happen to me. I mean, who knows, someone, some external force could put pressure on me, or I could become a bad person. I mean, you don’t know. You don’t want a project this big to be reliant on a single human being. And it’s not good for that person because it just puts a target right on them It makes it difficult for them to talk. Everything they say they have to kind of think about how it might be viewed. So, I think that that diversification of the humans is a good idea, too.

Asheesh: And to that note, I mean, the early days of Ripple we would send a satellite phone with David [when I was on vacation] so we could call him 24/7. So David never really got a full vacation, which is great that we decentralized David. So, the next one’s from DaddyCrypto, awesome Twitter handle name. “What are you most proud of amongst the past five years of achievements? And what is the next technical problem for XRP?” DaddyCrypto.

David: I think the proudest achievement has to be the team that we’ve built here. I mean, I can’t take full credit for that, but I think I can take some credit for that, particularly on the technical side and particularly on the XRP Ledger side. We’ve just acquired a group of people who are just phenomenally talented and who are so aligned with what we are trying to accomplish. I think that’s just unbelievable. This is really a great place to work. I guess I’ll make a little plug for our recruiting. We’re always recruiting. No one asked me to do that, but I thought it kind of came up naturally so, please, if you’re interested in working at a great company in the crypto space, you know, we’re definitely one.

Technical challenges for XRP. There really aren’t that many technical challenges left for XRP. Decentralization is mostly an execution challenge and in terms of just sort of gradual improvements. But we do have a lot of technical challenges in the ecosystem that we are building around XRP with how will liquidity be provided for payments, integrating Interledger with XRP. Those kinds of areas we still do have a lot of technical execution to complete, particularly with what we’re doing with liquidity, bridging live payments and those areas are the areas where there is a lot of technical focus. XRP is mostly just going to be a matter of incremental improvements, things like resource consumption of the software and transaction throughput. Not a lot of major features. I think we have the features set for most of what we’re trying to do but there’s certainly the ability to add features for other use cases. It’s mostly going to be building the ecosystem around it.

Asheesh: Alright, we got another deep one for you David, from Ruben. “Hi, David. Can you explain how the Ripple validator system works? If everyone can run a validator, how does the system know which ones to trust, and, for instance, prevent a 51 percent attack from malicious parties. All in just a few seconds.”

David: That’s a great question. There’s a lot of… the way consensus works is somewhat different from the way proof of work works obviously. Consensus works by validators. Validators are used really to do only one fundamental thing and that solves the double spend problem. I think it’s important to point out that validators do not have control over how funds flow or what transactions are valid or not. They just put them in order. The only reason you need that is if I have to say 1,000 XRP, I could try to send that same thousand XRP to two different people. And if we can all agree on which transaction comes first, then we should all be able to agree that the other transaction is invalid because there are no funds for it. So, that’s the role that validators play. Validators are very inexpensive to run, conceptually, if you’re running Ripple D, which is just like running a Bitcoin full node. Running a validator is almost literally just flipping a switch, you just literally tell your server I want to be a validator. And, obviously, as the question says, you could then have millions of validators, all of which could be the same person for all that you could tell. And then you could have a malicious attack where that person sort of refuses to order transactions. And he can’t steal funds that way, but he can halt the network. He can perform a denial of service attack by simply refusing to reach an agreement. So, the way that we handle that is validators have a cryptographic identity, just like accountants do. And you have a list of cryptographic identities that you believe belong to distinct people who are unlikely to collude against you. And you ignore any validator that’s not on your list. And one of the interesting quirks about the way that consensus works is that it’s not always obvious to you whether other people are listening to you or not. If you imagine a room full of people trying to agree on something, let’s say, what time they’ll have lunch. If nobody were listening to you, it would still seem like they would, because if everybody wanted the same time and you wanted a different time, you would switch to the time that everybody else agreed on. And if you wanted the same time as a whole bunch of other people you would see them switch to that time, whether they’re listening to you or not, because they’re just switching to the most popular view, the most popular ordering of transactions. So, the network works just fine whether you’re a validator or not. Validators don’t get to choose which transactions are valid or not. They’re just a way of determining unique nodes. Nodes that we believe are unlikely to collude. So, that’s just done by listing them. You just choose the ones that you think are independent. And because that’s not something that each person could easily do individually. You’d imagine that would be somewhat of a challenging thing to do. What we envision a publisher is that essentially publishes lists of validators that they believe are operated by distinctive entities who are unlikely to collude.

Now, you could imagine that someone went to the effort of infiltrating one of those systems and got himself listed as, say 500 different validators. And in that case, if the set of validators was small enough, he could launch a denial of services to keep the network from making forward progress. But the important thing is anyone who did that attack has to cryptographically sign every single message they send. That’s how other people know to listen to them. All people have to do in response is simply remove them from the list of validators that they’re willing to listen to. So, there’s a lot of effort to build up the ability to make that attack and that ability is immediately lost the moment you try to use it. That attack accomplishes so little, it’s just a short-term denial of service attack on the ledger as quickly as people can respond, there’s little incentive to go through the effort of creating that number of validators and infiltrating those lists just to be able to perhaps launch a one-time denial of service attack on the network.

Asheesh: Excellent. Next one is from Shazad. “My question: As impressive as RippleNet is, is there room for improvement in regards to speed, scalability, and the company in general. As Steve Jobs would say, ‘Can we make it better?'”

David: RippleNet scales horizontally. It’s a server to server kind of protocol, kind of like TCP is and so you don’t have a centralized ledger that limits scalability. So, you don’t have scaling issues with RippleNet itself. There’s room for improvement in the company, though. Obviously, we’re trying to target more payments. We’re trying to create more XRP liquidity. We’re trying to expand all over the place.

Asheesh: OK, so the next one’s from Evan Timmons. “In his Q&A…” So this is the last live stream we did with Brad. “Brad Garlinghouse mentioned exploring new use cases for XRP in the future since Ripple has managed to gain solid ground doing one thing well. Do you foresee XRP becoming liquid enough to be an intermediary currency between other coins, the way that BTC Bitcoin is now?

David: As Brad said, we have focused on doing one thing and doing it really well, and that is settling payments. That’s building out a payment network that’s complaint that financial institutions and corporations can use. We do see other use cases for XRP. I think a lot of people, particularly with Bitcoin having some paralysis in its transaction fees and its transaction rates, I think a lot of people have looked at XRP for some of the same things that they would have used Bitcoin for, say a year ago. We’re pretty excited about that. We’re looking for ways that we can encourage that.

Asheesh: Alright, so the last question, from Zager. “Help me understand the escrow account deposit of 62 billion ripples XRP by creators.”

David: So, Ripple received its stash of XRP when it was founded. The original creators of the XRP Ledger gifted their XRP to the company. And the company holds something like on the order of 65 billion XRP. To provide some predictability in the supply of XRP, Ripple has taken 55 billion of those XRP and locked them in an on-ledger escrow. And that essentially behaves kind of like a smart contract. It releases the XRP at a predictable rate that’s greater than the rate at which Ripple has been using XRP in the past. We don’t anticipate that that’s going to correlate to any increase in the rate at which we use XRP just from the fact that the release rate is higher. Before we had no restrictions. So, adding restrictions certainly wouldn’t be expected to increase that release rate. And the plan is that when there’s leftover XRP at the end of each month, we’ll just lock it back into an escrow, releasing the first month in which there is no escrow. Escrow doesn’t really have a counterparty to the extent that there is a counterparty it is the XRP Ledger, it behaves very much like a smart contract. The ledger feature that it uses is called escrow, and it’s also useful for a sort of secure payment where you lock XRP up to be released to someone when they meet some condition. So, our escrows just don’t have any condition except a certain time passing. And after that time, the XRP releases. And that provides some level of predictability in the release of the XRP supply. But it also ensures that Ripple has an incentive to promote the stability of the network and the utility over the long term.

Asheesh: Alright, so I want to thank everyone for tuning in and celebrating the five-year anniversary of the XRP Ledger with us. Also, I want to thank you, David, for all that you’ve taught me about decentralization and cryptography. It’s been awesome working with you. Hopefully, awesome working with you for the next five years as well. And now we’re going to eat some cupcakes.

David: It’s been a pleasure working with you as well, and it’s been great to be here.

Asheesh: Happy holidays!

Related stories

Subscribe to the Ripple Newsletter