path: root/transcripts/cryptoeconomic-systems/2019/funding.mdwn

Tadge Dryja

Funding bitcoin development

It's basically open-source funding. It's a subset of open-source funding. But these are similar problems with many open-source projects. As far as I can tell, all these digital currency cryptocurrency projects are licensed with open-source licenses. There may be some that are not, and some things are patented. I generally steer clear of the patented work. If it's decentralized and open-source, then why are you just delivering a binary to me? I don't know what's going on there.

Open-source funding is tricky. It's interesting, though, because we're literally programming money. Sometimes I joke that what I'm doing is making money. Funding is not simple at all, and it's sometimes elusive. I am going to talk about a few funding models. These are not well defined. There's a lot of blurry perspective between these things and some of them have arrows going between one and another. I don't want to say too specific things.

How does a project like bitcoin or ethereum get funded?

There's no funding. There's the .edu university funding idea; there's corporate donations, personal donations, there are for-profit companies wokring on it, there's premine or ICO ideas, or mining rewards, and non-profit companies or foundations.

# No funding

This was bitcoin's initial model. There's no funding. People just work on things they think are cool and interesting. There's no money in this, so for some people it's only part-time or their weekend hobby. This can generally have a lot of people, but they are mostly part-time. This is how a lot of people who work on bitcoin started. I just started reading the code and playing with it, with no expectation of compensation. The fact that I get paid to work on bitcoin now is really cool, but totally unexpected. If you told me 8 years ago that I would be working full-time on this stuff, I would have said no way.

One of the cool things about the no funding model is that there's an easy on-ramp. You can just open a pull request. This is not the case in other models. All the other ones, the fact that there's now compensation makes it sort of in a way a barrier to a people who are not getting paid are in many ways less likely to work on it. It doesn't necessarily make any sense, but psychologically it's true--- if someone is making a lot of money to work on something, then why should I do that for free?

# University funding

Academia... I don't quite share all the criticism that Bryan has for academia. I tried to get a PhD working on bitcoin and I only lasted a year and I figured, this just doesn't work. This whole system doesn't work, and people are jerks to me, and I don't even get to work on what I want to work on. Most of the students I was working with were not US citizens. For me, my advisor was yelling at me and it was always on the phone and we didn't meet in person once-- and one day I moved the phone away and figured, I don't want to do this any more. For me, it's easier because I am a US citizen and I wasn't making that much money anyway, so I just stopped. I had rent, but whatever. But most of my colleagues were from China and India and quitting a PhD was a huge deal for them because if they quit then they had to leave very soon. It felt like many of the professors knew that and abused it. There's a lot of problems in academia. But the idea is great; we're all in it to research and advance knowledge. The idea is good, but the implementation could be improved.

One thing that people have assumed or asked about is, they think there's .gov funding like NSF or whatever. There's actually very little of that. The DCI as far as I know here hasn't gotten funding from NSF. Not sure if we have tried that much. There's a little, if you look around. IC3 if you look around, they got $3 million in funding from NSF which is a decent amount but not much.

# Corporate donations

These corporate donations happen sometimes. Companies donate money to developers or researchers and sometimes individuals. There are people at this conference who have gotten money from an exchange called BitMEX where they said hey you have been working on bitcoin for a while, keep at it, here's some money. Bitmain was funding some people. Fidelity donates money to MIT DCI and we work together on things. There's some of that, where corporations say hey this is an important thing and maybe we're not directly benefiting from it, and we're part of this ecosystem and let's donate to it. There's also personal donations from individuals to other individuals or groups. In some cases, these individuals start VC style funds to invest in things or they-- one of the big ones in bitcoin is that some individuals started a firm with their own money and employed a lot of bitcoin developers. So that's a big thing.

# For-profit company

This one is tricky. Some cryptocurrency projects are run as companies, like a regular Delaware C-corp and there's a board. There's a CEO of the currency-- so how is this decentralized? It might be open-source, and that's useful, but it does lend itself much more to centralized development. If there's a coin run by a company and you want to change it, well you're not in that company.

# Premine and ICOs

We could have a whole talk or class on premines and ICOs. It started with premines back in 2011-2012 when altcoins started to pop up. People would say hey I'm making an altcoin, and maybe I want to give myself some of the money in the system. The idea of a premine was that there was an initial distribution of coins in the software. In bitcoin, this is not the case. In bitcoin, the genesis block has 50 coins but either intentionally or unintentionally these coins are not spendable. Everyone who wants bitcoin had to mine, and Satoshi didn't code in any for himself.

ICOs are sort of a pre-pre-mine which started years later; you want to do a premine but you haven't written the software yet and you want the money to write the software. So you give an IOU for the coins that don't yet exist because the software hasn't been written. But you want the IOU to trade and be tradeable, and then people can trade on those coins that don't yet exist. There's ERC20 and so on. This seems to raise a lot of money. It's billions of dollars. Today, this seems like an inefficient use of capital. There are projects that have received billions of dollars of funding, and they haven't delivered that much interesting work, at least proportion to the amount of money raised. Yeah I work on bitcoin and have that bias, but if you look at the proportional output, yeah... Also, there's a bunch of legality questions and maybe this mania was over and it was just an unsustainable fad.

# Mining rewards

I only know of one prominent case that has said, similar to premining or an ICO, let's have a vesting schedule with mining. What happens is that when coins are mined, some portion of that mining reward goes to the initial developers or initial investors. This feels a little bit better. In ICOs, you ask for $100 million and then you get it and then there's not a lot of motivation to actually work... there's a lot of fun things to do with $100 million. Programming is fun but I would probably want to have fun for a year first. Mining rewards is more like a vesting schedule and doesn't have the frontloading property. But why should I work on your well-funded coin? This is an interesting project, and maybe I can solve some of the problems, but you guys have $100 million and how about you pay me first? I don't want to volunteer my resources or my expertise or my time to a place that has fancy offices and treehouses in their office and stuff, without being compensated. Some part of me finds this cool, and wants to work there, and then another part says well if they don't pay me then why should I contribute work to them.

# Non-profits and foundations

There's these non-profit foundations that-- I don't really understand them that well. They are often linked to ICOs or something. After an ICO, they create a non-profit org like Ethereum Foundation or zcash foundation. But where does the foundation get the money? If you're a non-profit, but you're funded primarily from this for-profit company then how independent are you really? The people funding you might be controlling you, too. The idea is to decentralize the funding, but in practice it's not that clear to me whether it's that useful. There was a Bitcoin Foundation... you could ask some people here who worked on it, you could ask them, some of those people are here and they probably have fun stories. I have lots of fun stories. It was hard when writing these slides to not talk about specific fun stories about people and entities which might not be appropriate for onstage.

# How should we fund these things?

I don't think there's a clear answer for this. Bitcoin seems to have sublinear development costs. When bitcoin was $20, there was a bunch of people working on it. Then we had these coredev.tech meetings, and as bitcoin is worth 100s of times worth more now, there's not even 2x as many people working on it. There's more people working on it, as the price has gone up, but it's definitely sublinear. In economic terms, you want small amounts of money to go in and a lot of really useful to come out, and if it's sublinear then that's great. But is it too low or too slow not sure.

This is often a problem in other open-source projects, like GPG which is maintained by like one dude. OpenSSL, which is used by everyone, and there's bugs-- but it's someone else's problem. It's high wellfare in the economic sense, in that you're not really paying for openssl and yet you get to use it. So open-source is great and you get to download linux for free on to as many computers as possible, it's the same software some people would pay billions for and I get it for free. Maybe we could get even better cryptography and systems if we had lots of funding. But maybe it's just the opposite; maybe we get more with less funding. I am one of the people getting funding from working on open-source software.

# Worst case

Everyone tends to start their own coins and shares and coins, and then all the talent gets fractured into tons of competing coins and nobody cooperates. This is like the ICOs and look at coinmarketcap-- there's rational reasons for people to do this, so it's hard to complain. There might be some stability in advancement paths for long-term careers. I am working at DCI now, but is this a career? Maybe this whole thing disappears? This thing didn't exist 10 years ago, and there was no funding 7 years ago. Is this a good long-term career choice? The reward being sublinear in system utility for net social gains is great, but we don't don't know how to do this. Markets tend towards getting good results, but it's a hard problem.

# Conclusion

I work around here, so come bug me, ask me questions. Thank you.

# Q&A