Jimmy Song: Attacking Bitcoin Network Would Cost Billions of Dollars But Still Would be Pointless

Interview
30.10.2018

Entrepreneur, developer and educator, Jimmy Song is a well-known figure in Bitcoin community. Former principal architect at blockchain firn Paxos and VP of engineering at wallet management platform Armory Technologies, today he runs a training company called Programming Blockchain, that provides blockchain programming and development training, and also acts as a venture partner at Blockchain Capital, a San Francisco-based investment firm.

A frequent speaker at various events, where he appears wearing his trademark cowboy hat, Jimmy Song also runs a popular blog on Medium and has over 130 thousand followers on Twitter, where he shares his views on the cryptocurrency ecosystem evolution. On top of that, boasting a considerable programming experience, he also contributes to the development of Bitcoin’s underlying protocol.

In an exclusive interview with ForkLog Jimmy Song tells about how he got involved in the crypto industry and the importance of his mission of teaching new developers. Apart from that, he reveals why Bitcoin is the most secure cryptocurrency and why hard forks are meaningless by the nature, and also points to the most important solutions needed to improve the network and to make it more secure.

Photo: Paralelni Polis / Pavel Sinagl

ForkLog: Hello Jimmy, will you tell our readers more about yourself? How did it all start for you and how you came to Bitcoin?

Jimmy Song: I guess I I’ve been programming since I was like nine years old, I’ve been doing that for a really long long time. I’m 42 now, so that’s like 33 years that I’ve been doing some form of programming.

I started doing startups like back in 1998. I came out of college and that was like around the clock. Those were the early years of the Internet, and the first startup I did was called ATM Technologies, it’s not really around anymore. It was like a website translation into different languages and stuff like that. I also did a bunch of different stuff, something like a healthcare startup, a web hosting startup, a coupon savings site, and many other things. So basically I’ve been programming since I graduated from college and even remember times when there were no Javascript frameworks in place and I had to code everything on my own.

And I think I was doing like a phone startup when back in 2011 when I learned about Bitcoin. I thought that it was really interesting, but I didn’t buy unfortunately even though it was a dollar at the time. I did buy some Bitcoins later on and it was around 2013 when I started getting into it more as a programmer. It was the time when we had that first bubble in April 2013, and it was actually my first day when everything went down (laughs). And then later that year Bitcoin went up again and that’s when I started working on some open source projects.

Actually, the first open source project I contributed to was something called colored coins. The reason I got into that was that I was looking for jobs at Bitcoin Reddit and there was a post saying that Python programmers were wanted. I know Python and I replied, and it turned out the guy was actually from Ukraine. He told me that he would pay me like a fair wage per hour and he wanted me to work on this open source code. I said, okay, I get to contribute to open source and I’m gonna get paid in Bitcoin.

I was working in Austin for a coupon site at the time and I was working forty hours a day a week for that startup, and now it was like forty hours a week working for this guy in Ukraine getting paid it Bitcoin. So, at some point it was just way too much and eventually I had to quit and do more Bitcoin things.

That was my first exposure to Bitcoin, and a lot of the stuff I just didn’t know anything about. Bitcoin has a lot of these RPC calls, and this guy told me he wanted me to do these RPC calls for colored coins. I said: “Okay, I have no idea how any of that works, but I have to look at this and figure everything out.”

Then I also did very similar things, like there was this new standard called BIP-32 for HD-wallets, and basically I had to teach myself a lot of it just from having to cook for this guy. And you know, that’s often the best way to learn something when somebody’s paying you to learn it.

So I did that and eventually I found my way into Bitcoin development. I was working for Monetas, a startup that was based in Switzerland, and then I went to become VP of Engineering at Armory, and today I do my own thing teaching my class. That’s my story of how I met Bitcoin.

ForkLog: Your voice is quite loud on social media, and you are also known for running a successful blog on Medium as well as for your articles on Bitcoin Tech Talk. Was it hard to establish yourself within the Bitcoin community?

Jimmy Song: It all started from my days at Paxos [former ItBit – ForkLog], they’re based in New York and they have an exchange and other things. They had me working on this enterprise blockchain stuff, and my boss there actually asked me to write a post because they were starting a blog. So I wrote a blog post, they liked it and published it and then told me to write another one. So I wrote another long one, more on Bitcoin, but they found it too much Bitcoin-based and said they didn’t really want to publish it.

I said: “Oh, crap, I spent a long time on this very long article, so what am I gonna do with this?” So I asked around including some people at Satoshi Nakamoto Institute if they wanted to publish it, and they said “no”. So I decided to open a Medium account and to publish it there, it was February or March last year. I also had a Twitter account with very few followers, I think it was less than three hundred at the time.

So I decided to publish it on there [Why Bitcoin Will Get Scaling Without Segwit or Large Blocks – ForkLog], and it started going by. That was my first exposure to this, and people like Jameson Loppe quoted it in his article for CoinDesk, some other people were quoting it for various other stuff as well.

So, I saw that a lot of people liked that and I decided that I’m going to publish some other stuff because I had a follow on from that article. I published two or three more that a lot of people liked, and I continued publishing and started getting lots and lots more followers, after that I was invited to The World Crypto Network so I started getting into that whole YouTube game, and it’s sort of exploded from there.

I think people like that I am explaining stuff in a rational way. It’s funny, because I’ve always been as a coder, and to get recognition as a writer or a YouTube personality was something new for me. Anyway, these are new skills so why not try it. Eventually I got to the point where that was becoming more and more of my life. I ended up quitting Paxos and I wanted to do some teaching because I was convinced that that was something that people needed.

ForkLog: And at the same time you also contribute to Bitcoin Core development?

Jimmy Song: Yes, I contributed to a lot of open source projects by that point, but in 2017 I realized that want to contribute to Bitcoin Core because it’s rational for my investment. I started doing some commits, I think I had like fifteen commits to my name, it’s not that many, but I wanted to know about Bitcoin development so I could write about it.

I wrote an article for Bitcoin Tech Talk, ‘A Gentle Introduction to Bitcoin Core Development’, that was like my main motivation for doing it. I wanted to make sure that there were developers because I see that as like the major vulnerability that Bitcoin ecosystem doesn’t have enough developers.

And this is why I also teach my online course because that’s the major thing that’s missing: we need more developers. There are lots of people that are doing ICOs, but not enough people that actually understand the protocol. To do actual Bitcoin Core development you need to know a lot of fundamentals, and that’s the course I decided to teach.

What I realized is that there’s probably room for a person like me to utilize a lot of the skills that I didn’t really get to utilize for the fifteen years of my eighteen years of my career.

It’s a two day course, eight hours a day, and the way I do it I teach the material but then I also make the students code actual things. They have to actually make a Bitcoin library from scratch and all the way to network programming and figuring out how to connect to nodes and broadcast transactions. It’s a very fundamental level, I want to teach all the basics so that all of these people I’m teaching eventually get to contribute toward development. And that’s my goal, to create more developers.

ForkLog: Last month a DOS vulnerability exploitable by miners was discovered in older Bitcoin Core versions. This even lead to the security release of Bitcoin Core 0.16.3, and while many developers downplayed the level of the threat, it still leaves plenty of questions. What’s your take on the issue and what other vulnerabilities Bitcoin is exposed to today?

Jimmy Song: The big thing is that software will always have bugs, and some of them are unforeseen. Also, there are always forces that will try to centralize Bitcoin, to make it easier to deal with certain scenarios. Governments, for example, are typical reasons why people want to decentralize. However, Bitcoin is more or less immune to attempts of centralization, and I’m pretty optimistic in that regard. That’s why software bugs are just those other major vulnerabilities.

ForkLog: What causes concern is that on technical level it wouldn’t cost an attacker too much money to exploit the latest vulnerability. This leads to a suggestion that if certain forces, for example, the state, would really want to harm Bitcoin, well, basically the doors were open for them.

Jimmy Song: As for this particular vulnerability, it would take lots, lots of of mining equipment to exploit it, and that’s very hard to procure even for the state. You could have like 1 or 2 percent of hash power to try to execute that kind of attack, but all you really are going to get is like one or two blocks to disrupt the network for 30 to 60 minutes. There’s no way that would last for too long. It would have been figured out and patched very quickly.

So the only way for someone like state to do something permanently is to have a lot more hashing power. To really be able to disrupt the Bitcoin network you would need like 60 percent of hashing power, and if you look at the mining landscape and see how much it costs to create a miner, there’s only a few places where you can get those chips without spending too much money. It’s really difficult to get that much power.

I agree, it’s still possible. If you are willing to spend an infinite amount of money you can procure all that on the market, but as soon as you start doing that prices of those things will go jack up. So if you are going to exploit something like that you will have to procure the equipment and try something like selfish mining attack. But you would literally spend billions of dollars in order to make that work.

It’s already a lot of risk to set up this kind of attack, because most attackers, even at the state level, want to be assured that it’s going to work. And if you fail, that makes the Bitcoin network super robust.

ForkLog: Bitcoin is already pretty robust as it is. Are there any other cryptocurrencies that come close to it in terms of security?

Jimmy Song: No, Bitcoin is the most secure crypto, and its main property is decentralization. Many of other coins are Bitcoin hard forks, and hard fork is the evidence of centralization. If you are able to tell people to upgrade their clients, or otherwise they are out of your coin, it’s sheer centralization and not much different from a central bank.

Photo: Paralelni Polis / Pavel Sinagl

ForkLog: What future do you predict for some projects that hard forked from Bitcoin in recent months?

Jimmy Song: All of those are completely worthless, even though I managed to increase my Bitcoin holdings as a result (laughs). Anyway, my take on this is that all those coins are some stupidity. It’s crazy, many of them came out in January or February and they were trading like one percent of BTC. Many of them have gone 90-95 percent since then, and it’s insane how much they’ve lost. All this happens because they offer nothing new or interesting, or solving actual problems. In fact, this is the situation with most altcoins.

ForkLog: The scaling debate was a heated topic for many years, and many people believe that the 1Mb block size limit gives Bitcoin plenty of advantages from the security point of view. The so called ‘big blockers’ went a different path though insisting that taking this limit off would better scale the network.

Jimmy Song: It isn’t that 1Mb is optimal for me, the thing is that it doesn’t require hard fork. A hard fork is almost by definition forcing centralization, and that’s a bad trait. If you get more scale for more centralization, I never take that. You want decentralization, that’s what makes Bitcoin interesting, and as soon as you centralize things it’s not interesting any more.

And that’s why Bitcoin is different from everything else. Almost every altcoin, more over, everything else that has ever existed is either not digital or not centralized. Gold is decentralized but it’s not digital, Ethereum is digital but it’s centralized. You want decentralized digital scarcity and that’s exactly what Bitcoin brings, that’s the major innovation. And to me no other innovation has come close to that level of breakthrough.

ForkLog: There were some controversial scenes during your debate with Roger Ver on the Blockchain Cruise in September, and it didn’t seem like the opponent actually wanted to hear what you were saying about Bitcoin and Bitcoin Cash. Soon after that you wrote a blog post where you brought your thoughts together, explaining why BCH is actually fiat money. Any chance you were heard?

Jimmy Song: I don’t think so. During the debate he never responded to any of what I said. I was trying to elevate to conversation to the level of ideas instead of people, but all he wanted to talk about was credentials. Like ‘I read this,’ ‘Have you read this?’ or whatever, and I think it’s a very low level form of discourse. It’s like talking about celebrities, and that doesn’t interest me in the least.

Eleanor Roosevelt once said: “Small minds talk about people, mediocre minds talk about events, and the great minds talk about ideas.” I wanted to elevate the conversation to be about ideas, and Roger Ver didn’t want to engage at that level, he just wanted to talk about people: Craig Wright, Blockstream or Bitcoin Core people. That’s all he cares about, and I see him as someone who doesn’t really understand the technology very well, someone who just wants to sell his project, and that’s Bitcoin Cash.

So I don’t think he heard me, and pretty much everyone in the Bitcoin Cash community just wanted to talk about how rude I was to Roger, or that I left the stage. It’s completely irrelevant stuff, it’s not about ideas. The funny thing is that I don’t represent Bitcoin, because Bitcoin is decentralized, but Roger does represent Bitcoin Cash, and that’s the difference.

ForkLog: The recently announced The B Foundation has already received its portion of heavy criticism from some members of the community, who say similar things: no one can represent Bitcoin and that’s why such initiatives can’t be a good idea. What do you think of it?

Jimmy Song: As long as it is not trying to centralize Bitcoin I am all for any sort of innovation. Even if this is a bad idea, all of us will learn something. My hope is that they are not going to centralize it, but I know Alena Vranova, I know Giacomo Zucco pretty well, and I don’t think that’s the direction they are going. From what I know and from what I was explained, they have a domain knowledge about what needs to be done in Bitcoin. There are a lot of people who want to contribute money towards efforts, they just don’t know where to give it. This is like a traditional charity, when people know that this organization actually does a good work and can be verified. This is a way to make things more efficient, and if that’s their goal it could work.

ForkLog: It’s already less than two years until the next halvening. What do you expect to happen in the network by that time? What Bitcoin will be like from the security and price perspective?

Jimmy Song: I’m not a price expert, but generally over the long time Bitcoin tends to go up in price since there’s a fixed supply and increased demand. And the halvening tends to be like a Schelling point which brings down the inflation and decreases selling pressure because miners don’t have to sell as much as before. So I imagine the price will go up, but sometimes it takes a long time for the market to catch up with that.

Security-wise, that always depends on software, and that is something that is constantly improving. I’m not sure if the halvening does necessarily affect this very much, but I expect that by that time we will have more solutions.

ForkLog: What solutions you personally anticipate the most?

Jimmy Song: I think security is always the primary thing because I believe Bitcoin to be a store of value. Therefore for me a lot of innovations are around the payments, or how to make payments faster. Also a lot of things are about how you store your coins, stuff like Casa Hodl are doing, open source hardware wallets and things like that.

Jimmy Song was interviewed by Andrew Asmakov

Found a typo? Highlight text and press CTRL+ENTER

Subscribe to our Newsletter

<

Related posts

Tags: , ,