Jimmy Song

Bitcoin developer, Venture Partner at Blockchain Capital, Educator and Entrepreneur

29th June 2018, 6:00 pm

This AMA is over!

Thank the host

We'd love to have in the Philippines - what would it take? Also is there a remote / self-paced version of the course?

Hi Paul,

I would love to come to the Philippines sometime! Unfortunately, my travel schedule is jam packed and I'm working on enough projects that going to a place doesn't really make sense without a substantial number of commitments.

I am working on an on-line version that I'll trial in short order. I'll probably have something out later this year.

How many committed devs would be the minimum viable commitment?

You recently visited Brazil, what you liked the most about the country and the community?

First, I love Brazil. I will be back some day. The people were what I most enjoyed. Everyone was super friendly and fun to be around.

Second, the sense that I got from a lot of Brazilians is that they really "get" the store of value property of bitcoin. They know what inflation is like and understand that having an asset that's not subject to inflation is really, really useful.

Third, Brazilians love their football! Since South Korea is out, I'll be cheering for Brazil the rest of the way.

What are some of your favorite blockchain projects?

What is the best article / book to convert people from the "Bitcoin is Bad, Blockchain is good" phase?

I've been trying to write some of those articles! I'll let you be the judge on that.

I like Saifedean Ammous's "The Bitcoin Standard", which explains everything very well.

Hi, Jimmy! Love to attend your talk in Blockchain Academy/SP.

What do you think about PoW change proposal? Do you see risk in the current mining hashrate centralization?


I don't see PoW change as necessary or desirable. The hash rate is in pools, which have lots of individual miners. The problem is that two things are currently conflated: mining reward variance reduction and block construction. The main danger is in pools controlling the latter. The former is absolutely desirable.

I think Matt Corallo's BetterHash would be a great way to separate the two concerns and decrease the main risk.

The assumptions under which people get scared about PoW 51% attacks is an economic actor which somehow gets some irretrievable good or service that costs a lot of money (many hundreds of bitcoins) through double spending. This is really hard to pull off in practice and pretty hard to get away with, so I don't really worry about it. The economic incentives generally align pretty well.

What do you think about "any amount of bitcoin" bet by Lubin? 馃榾 Why do you think you will win (seriously)?

I've been trying to get the guy to actually bet me, but he won't respond. I think I'll win because if there was a decentralized app using a blockchain that was popular, it would almost immediately be disrupted by a centralized app that would be able to do everything cheaper, faster, better.

The only exception I've found to this is money and that's because it's desirable for money to not change nor be under centralized control. For almost everything else centralization and change are desirable.

Jimmy what you think about "Private Blockchain". Projects like Hyperledger, Exonum and so on. A lot of crypto/blockchain enthusiast are very sceptical about this and it's really interesting what you think about it. I can definitely see the scenarios where this kind of Frameworks and Libraries will be suitable.

  • Supply chain
  • Public sector
  • Business network
  • Health records management and so on.

Thanks in advance.

I'm skeptical of big company "lab" projects in general. They all claim to be startups inside big companies but most of them fail and have an even worse record than startups. They are largely responsible for making blockchain and "smart contract" into buzzwords.

There are technologies like public key cryptography and timestamps that may be very useful in all those industries, but blockchain is a very specific set of technologies that really don't make sense for centralized systems like this. See my article on "Alternatives to Blockchain".

Thanks, I'll definitely read it.

P.S. I'm subscribed to your newsletter and it's really useful.

What's the fairest criticism of bitcoin as a technology right now?

Probably a lot of the bugs in the core code that are very annoying. Endianness choices, OP_CHECKMULTISIG, 4-byte timestamp. The original code wasn't that good.

What is the market potential of Store of Value?

Digital gold. Off shore accounts. Probably in the trillions at least.

Are there many people adopting your bip 176 into wallets or the bitcoin ecosystem in general since it got merged in dec. 2017? I like the idea especially since lightning network is up and running with similar amounts.

I would love for BIP176 to get more adoption. It's an informational BIP so, nobody is obligated to follow, it's just something wallets can do if they choose to do so. I personally think it's a lot easier to think in terms of bits, but opinions differ.

Lightning transactions definitely change things from a micro-payment perspective. We'll have to watch the space closely to see if that becomes enough of a use case that new vocabulary becomes standard there.

What is the biggest 3 threats to bitcoin?

Lack of developers Governance Takeover Hardware breakthrough

Thank you! What would you give as a probability that these threats would take Bitcoins store of value use case in the next 15 years? 5%-10%? 50%?

I can't quantify such things. Usually defined threats don't usually cause the big disturbances, in any case.

Ok. Thank you!

Can you elaborate on the governance take over? Do you mean that like node decentralization, and therefore a newer more censorship resistant solution will come along? Or do you mean like on chain voting?

And also the Hardware breakthrough? Do you mean like Quantum Computing or another threat?

I'm talking something like Segwit2x.

Hardware breakthrough would be like a 2nm chip fab that can produce ASICs 100x more efficient than current generation ones for much cheaper.

Hi Jimmy! What resources would you recommend a budding developer transitioning from learning Python to Bitcoin development? And how important is working with Solidity? Thank you!

Solidity is only important if you want to make dapps on Ethereum. Let's just say that most "dapps" are filling in ERC20 or ERC721 templates, so it's not exactly rocket science. My prediction is that the usefulness of ETH smart contracts will be less and less important over time.

I would recommend reading Mastering Bitcoin by Andreas or my upcoming book. You can also take my class

How can a dev transition into Blockchain/bitcoin development?

Read Mastering Bitcoin and/or take my class

can you explain what makes ethereum centralized ?

They hard fork whenever they want. With each one they can set any policy that they want. This is exactly like a centralized government. Examples:

  1. DAO bailout
  2. Changing PoW to PoS
  3. Sharding
  4. Storage Tax
  5. Changing costs of various operations.

These are activities that centralized entities do, changing the incentives this way and that to centrally design the behavior they want.

is the reason they can hard fork because one entity has over 51% of the mining power or are there other factors ?

They hard fork because it's centrally controlled not the other way around.

Vitalik and friends decide for everyone what they want for the network and they get it.

thanks, i guess my question was what is it about ethereum that makes it so that vitalik can make these decisions vs bitcoin where there is no centralized planning.

Their history? They started with an ICO which is very much a centralized entity. That gives them some authority over the entire project since they can claim they're serving all the people that gave them money at the beginning (but not as a security, apparently).

Two questions.

1) What did you think will be the most impactful result / value that Bitcoin brings to the world? (ex. financial freedom, privacy, etc.)

2) Do you envision any type of insurer in the event of self-banking displacing traditional banks? Do you think there will be insurance for losing private keys, etc.?

1) Sound money. This in turn gives people more of a long time preference which includes capital accumulation and building up of civilization.

2) You can certainly have some sort of insurance, but the risk profiles are really hard to build and in the face of uncertainty, insurers generally charge a lot of money. I'm guessing many will just demand to be a part of the multisig to make this work.

What I dont understand.... what is the difference in Tether to any other fintech except they handle crypto and we can "see" what they do. When trim it down the public ledger is even extreme transparent compared to any other company. In my opinion it is not perfect of course but whats the difference or scam compared to any other fintech?

Tether is centralized and you're right, there's no real innovation here except that we know how much they've essentially lent out. Of course a lot of it could be fake if they so chose, so the only real point is giving exchanges a convenient API.

Thanks Jimmy for answer!

What questions should startups ask themselves before considering to use blockchain?

What is the problem I need solved? Do I need something I can control? Am I okay with having everyone know what I'm doing? Am I ready to support the network for little to no revenue for a really long time?

Thanks for the AMA Jimmy. Sorry if my question is a bit long winded.

I have some background in programming and also in mathematics (having studied it in college). Unfortunately I don't have experience with the most used/useful programming languages in the Bitcoin space, and I also don't think my mathematics/cryptography skills are quite strong enough to contribute meaningfully to anything important.

Do you think it would be better to try and find a graduate program at the intersection of math, CS, and Bitcoin, or to try to find a job where mentorship and growth are possible in these skills? Do you think such jobs/companies exist?

Thanks again.

I would encourage you to learn some Python and then take my class. There are scholarships for people like you and if you want to get into this space, this is a great way to get in.

Check for details.

Hi, I received a notification that you answered my questions, but I can't see that answers.

Yea, I was looking for them, too. I think they got deleted. If you have a scam accusation, please use another forum to do so.

How can I contact you to send you that photo?

WOW Jimmy! Thanks for the AMA.

What's your take on the current Ethereum eco-system apps? Do you think it's the future?

Most of them don't do very much and don't have many users. If these were startups, I would pass on them for certain. I don't see much that's interesting there.

What past event do you think was the most detrimental to the growth / progress of the crypto space?

Without a doubt, Mastercoin back in the summer of 2013. It started the entire ICO trend which plagues our industry.

Interesting. Thanks for the response!

Hi there.

1) Apart from currency, books and art seem like entities in much need of censorship resistance from governments (to avoid Fahrenheit-451 scenarios). Is this feasible using the blockchain? How could one go about scaling such a large amount of data? Do storage coins fulfill this role?

2) When are you coming to Mexico?

1) The internet is your main method of censorship resistance. We already have bittorent and the like to make content available. I don't know why you need a blockchain on top. That's really a costly way to store something.

2) Not coming to Mexico anytime soon, though I'm open to it.

Hey Jimmy, I know you're not a fan of the blockchain wallet. Is there any Bitcoin wallet you could recommend for an Android phone?

For small amounts, most wallets are fine as long as you control the private keys.

Hi again! Do you know how to code in any programming languages well? If yes, please list them here. Thanks again :) !

I know my share of languages. You can see which on my linkedin page.

In regards to the global economy, what role do you see Bitcoin playing in 5, 10, and 50 years from now?

Hard to say exactly, but I imagine Bitcoin will become more important over time and eventually become the main store of wealth. Whether that happens 5 years from now or 30 years from now is hard to know. There are social considerations on stuff like this which isn't well studied or known.

Will lightening in Bitcoin be able to incorporate in mobile apps for payment in future? How will lightening affect the network security and transactions speed?

Transaction speed will be near instant. Probably at the order of loading a web page. Once the full network is up, mobile apps and websites should be able to incorporate them really easily.

How does the 鈥渄irty鈥 bitcoin flagging work exactly? Can a government or authority create a system that flags someone鈥檚 or most bitcoin as 鈥渄irty鈥, so exchanges and places will reject it and it becomes unusable?

You would need to do significant chain analysis to see which bitcoins are tainted by what. This is an area of forensics mostly guided by which inputs are spent with which other inputs and then linking them together.

The main place where this would hurt is if a government started relying on these. However, chain analysis is not that reliable yet, so I don't see this gaining much traction, though something more may be possible in the future.

How would you secure best an app needing some running btc client - docker & NGinx as a reverse proxy?

If the BTC client doesn't have a wallet, I don't know if you need to secure it that well. Docker and nGinx are pretty good, but you can just run without either one. There are tutorials to run it on a cheap Raspberry Pi and separate at a hardware level.

I recommend off-line wallets for actual BTC storage.

Hi Jimmy,

What do you think about ICOs in general? Which type of companies should do the ICO?

I think ICOs are a broken business model. They're like paying for a restaurant meal before the place is built, menu set or anything else.

If gold was not used on electronics, jewelry, etc, and instead only used as a store of value; will gold be a better store of value?

Gold's usage in electronics, jewelry, etc is relatively minor. Nick Szabo has a bunch of writings on the history of gold and jewelry was in links largely so people could pay for stuff directly. In a sense, most historical use of gold was always store of value. So no, I don't think usage as something else is what gives gold value. Its properties as money, in particular stock-to-flow ratio, is what gives gold value.

Same with bitcoin.

Consider 2 golds (same total supply), one only used as SOV, while other used as SOV, plus electronics, jewelry, etc. Wouldn't this more used gold become more valuable (due to same scarcity) and thus be a better SOV ?

You recently said that from the blockchain ecosystem only Bitcoin will survive and that Bitcoin is the single good use of Blockchain. Why do you think that? And how does this affect how Blockchain Capital invests its money in Blockchain related projects? Do you invest only in Bitcoin related projects?

Blockchain is a conglomeration of many technologies. Public-key cryptography, proof-of-work, peer-to-peer gossip network, etc. The combination of those things really only makes sense in the Bitcoin context because Bitcoin really needs decentralization to make things work. Everything else already has a centralized point of failure, so it really doesn't make sense to use a decentralized tech in a centralized system.

I don't make investment decisions at Blockchain Capital, I only offer opinions if asked. My partners are way more experienced at that sort of thing than I am, in any case.

Hello Jimmy, what do think about building self-sovereign identity systems on the blockchain? Would you consider it a promising potential application of blockchain tech outside of crypto?

It's a very interesting use case and Christopher Allen is doing some really cool things in that area. PGP was a good first attempt, but let's face it, it's not utilized very much and a reboot is sorely needed. I can see some form of hooking into Bitcoin's blockchain as a good way to make this work, but the details need a lot of fleshing out.

Hey Jimmy, where exactly did you learn about Cryptocurrencies and the like? Just Curious :)

Back when I learned in 2013, it was mostly whatever I could find on the internet. Some direction from people on bitcointalk and IRC as well, but mostly, it was just studying what I could study.

thanks :) !

What is the best resource or course for a hopeful programmer to get started. 0 previous skills

Python for everyone:

Ty for all you do

Do 2nd layer solutions to scaling make hard forks harder on the 1st layer. (This Q was asked to Andreas)

Regards Martin C.

Hard forks are in general very difficult and antithetical to Bitcoin's voluntary nature. Second layers are orthagonal to this reality.

In Mexico we are ready to launch the first Mexican StableCoin fixed in $1 peso What do you think about it?

I don't think stablecoins work. They violate the impossible triad of free capital flows, independent monetary policy and fixed peg.

How do you feel about the inequality within crypto? Will a protocol be able to get to a multi-trillion dollar valuation if the founders/foundations take 15% right off the bat, meaning would you be okay with people being multi-billionaires/trillionaires? Same thing can be said about bitcoin, does satoshi owning 1 million BTC raise any concern?

Nope. This assumes that the natural state of something is equal distribution. Nothing in nature is like that and most things exhibit a Paretto distribution. I don't expect Bitcoin to be any different.

So the point of these protocols are to create a more fair playing field and instead they鈥檙e even more lopsided than current playing fields? Hmm....馃

If your hope is to see perfectly distributed money, then only socialism is going to satisfy you and that way lies madness.

The point of these protocols is to take away the ability of the government to confiscate your money through inflation. You should only get into Bitcoin if you want control of your money back. It's fundamentally incompatible with something like socialism.

I wouldn鈥檛 say my goal is socialism or perfectly distributed money - that鈥檚 one extreme. But there鈥檚 research out there that shows 4% of individuals hold 96% of all bitcoin. Even more so, 1% holds 88%.

I would assume this only gets worse as the supply decreases and production power of coins is held by a few.

I think inequality is something that people should take seriously, since it鈥檚 a huge issue in our current society. And I don鈥檛 think you should label 鈥渕ore fair鈥 as socialism...

The only way to redistribute is to either:

A. make the people that don't have money to create goods and services others want B. confiscate money from richer people and redistribute.

The first is the market and generally, this results in a Paretto distribution. The second is socialism. Literally, claim what rich people possess as "societal" good and redistribute to others. There isn't some in between. You either have some central entity that redistributes wealth or you don't. So yes, a "more fair" distribution according to your standard is socialism.

So you think that this type of wealth distribution is sustainable in the lon-run and that it鈥檚 a better system than we currently have? Or do you think BTC will sort of only be a SoV/Niche asset class?

Just an FYI, I don鈥檛 mean to come off as hostile. I actually respect you a lot and love reading your medium posts and the interview you did with Laura Shin. Also thanks for taking the time to respond to these questions 馃榿

Is there a way to protect old pre 2012 p2pk addresses (early mining rewards for example) from future quantum computers which are potentially capable of accessing those funds? Assuming bitcoin is worth a lot more in the future, a single entity getting access to all "burned" bitcoin from back then can't be a good thing. Is this even a threat at all in your opinion?

I don't see this as a threat, largely because I think quantum computing is much further away from realistic implementation than the media hypes it up to be. The best known attack on elliptic curve discrete log is Shor's algorithm which still requires 2^128 operations on a quantum computer. Doing even a tiny fraction of that is beyond our engineering capability at the moment, so I don't see this as a threat.

If there is a threat to Discrete Log, it's going to be something out of left field that no one expected and not something people have been hyping for decades like quantum computing.