A Visit with the Bitcoin Foundation’s Top Scientist

Gavin Andresen, the lead scientist for the Bitcoin Foundation (and one of its only two staff members) sat down with a few of us at the UMass Amherst Knowledge Commons meeting on Wednesday.  Having read so much hype and misinformation about Bitcoin over the past few months, I was excited to have a chance to talk to someone directly connected with this brilliant experiment in algorithmic institution-building.  Bitcoin is, of course, the digital currency that has been in the news a lot recently because of its surging value among traders – and its dramatic crash.  

For months the dollar value of a Bitcoin fluctuated between $20 and $50….but in mid-March the conversation rate soared to around $250 before crashing last week to $140 and then $40 yesterday.  (Today it was back up to $95.)  This kind of stuff is catnip to the mainstream press, which otherwise doesn’t know much or care much about Bitcoin.

Andresen, a self-described geek in his forties with a pleasant manner and trim haircut, strolled into the small conference room in his black rugby shirt and jeans.  Six of us proceeded to have a wide-ranging, fascinating chat about the functional aspects of Bitcoin, the political and social values embedded in its design, and some of the operational challenges of making Bitcoin a new kind of universal currency. 

For those of you who want a quick primer on Bitcoin, I suggest the New Yorker profile  by Joshua Davis in the October 10, 2011, issue; a terrific recent critique by Denis Roio (aka Jaromil), a Dutch hacker who is working to code new sorts of digital money; or the Wikipedia entry on Bitcoin

Bitcoin is of special interest to me for its remarkable success at solving a serious collective action problem – how to create a digital money so secure and authenticated so that no one can steal its value and ruin it as a stable, trusted currency? 

The problem that Bitcoin solves as a matter of algorithmic and cryptographic design is the “Byzantine General’s problem,” which has been described as “the problem of reaching a consensus among distributed units if some of them give misleading answers.”  As one reference describes it, the problem has been compared to the problem of various generals deciding on a common plan of attack:  “Some traitorous generals may lie about whether they will support a particular plan and what other generals told them. Exchanging only messages, what decisionmaking algorithm should the generals use to reach a consensus?  What percentage of liars can the algorithm tolerate and still correctly determine a consensus?” 

Bitcoin solves this classic problem of achieving coordinated action without reliable communication or excessive (or any) defections.  Much of this success stems from the startlingly solid cryptography of the system.  The other safeguard, Andresen explained, has been Bitcoin’s “get big quick” strategy.  If enough Bitcoins can be put into circulation quickly, then it becomes much harder for any faction to corner the market in Bitcoins or to compromise their integrity.  This is important because the viability of any currency depends upon the ability of the issuer to prevent counterfeiting or theft -- a kind of free riding on the social trust that any community invests in its currency. 

Goldbug libertarians and conservatives have always wanted to base the value of the US dollar on gold so that politicians can’t just create more money, deflating the value of the currency and people’s wealth.  That’s one reason that a lot of people are attracted to Bitcoin:  its value is locked in by its algorithmic design.  For them, the answer to the question, Who do you trust – the government or a software algorithm? – is easy.  Code rules. (Or would you prefer to trust John Boehner?)

At present there are between 10,000 and 20,000 computers “mining” Bitcoins – i.e., minting new digital units of money.  Thousands of people around the world have set up computer systems to run Bitcoin software to try to identify the numbers that solve an equation.  Guessing the right number releases a Bitcoin to the miner, who can then hoard it, buy something with it, or trade it for dollars or other conventional money.   

About 50 Bitcoins are released into circulation every ten minutes, and there are currently 11 million Bitcoins issued, which is about half of the expected total coinage of 21 million Bitcoins, a release due to be completed in the year 2040.  The rate of new Bitcoins mined is fairly constant, thanks to a self-adjusting algorithm in the system that changes the rate of new coinages as the size of the Bitcoin network grows.  At today’s dollar conversion rate of $95.40/Bitcoin, the value of all Bitcoins in existence is about US$1.05 billion. 

A few years ago, some Bitcoin investors and geeks created the Bitcoin Foundation to help fortify the Bitcoin software infrastructure, promote the integrity of its cryptography and protocols (and thus confidence in the currency), and promote Bitcoin to the general public. 

As the chief scientist, Andresen said that he tries to maintain and improve the core code of the Bitcoin system and its usability, especially the security of the various “digital wallets” that people use to store the unique numbers for a given Bitcoin.  (Technically, this is a secondary market that is not a part of Bitcoin itself.)  Just as the diversification of web browsers created new coding challenges for websites, so the Bitcoin Foundation wants the currency to be compatible with various digital wallets that vendors sell.

One of the things that neither the Bitcoin program nor the Bitcoin Foundation can control, however, is the volatility of its trading price.  There have been at least two major bubbles since the currency's release in 2009.  Bubbles seem to flare up when interest in Bitcoin starts to surge, and then the press amplifies public interest and inflates people’s expectations.  Speculators jump into the game and before you know it, a bubble is underway. 

The latest bubble was apparently spurred, unwittingly, by the Financial Crimes Enforcement Network (FinCEN), a division of the US Treasury Department.  In mid-March, FinCEN issued some interpretive guidelines of the laws governing money-laundering and other crimes, which evidently gave speculators greater legal certainty for investing in virtual currencies.  The crash was aggravated by the Bitcoin trading infrastructure, which was not mature enough to handle the heavy volume of new traders.  (Private companies act as exchanges for buying and selling Bitcoins.)  Andresen chalked up the bubble and crash to "growing pains" for the currency.  He was not alarmed.

The idea of “mining” by computers to create new Bitcoins seems supremely odd to outsiders.  But as Jaromil points out in his article, the work of miners benefits the user community because it “strengthens the network of trust by making Bitcoins less likely to be counterfeited.”  Instead of having to guarantee the authenticity of money through “the monopoly on violence imposed by the sovereign state,” Bitcoin uses distributed participation via computers to authenticate the currency (and every subsequent transaction).  This is a significant technical and political achievement.

In our conversation on Wednesday, it was pointed out that the mining of Bitcoins is constrained only by the cost of electricity relative to the value of Bitcoins.  When Bitcoins were selling for about $250 apiece, this triggered something of a gold rush in the mining of coins because the ratio of a Bitcoin’s value relative to electricity costs was high.  Some people have proposed that Iceland might be a prime site for mining because the excess heat generated by mining computers could be put to good use.  Iceland as a Bitcoin mining hub?  Some serious miners have actually built custom-made chips that can use less electricity than standard, off-the-shelf computers. 

Andresen assured us that he is neither a miner nor a trader of Bitcoins.  He does carry his own share of them around, however.  He took out his iPhone and showed us eleven Bitcoins in his “wallet,” which can be instantly converted into a QR code (those readable digital squares on products) and transferred to another smartphone, if desired.  Digital money!  Andresen also showed us a nifty smartcard that can serve as a repository for one’s Bitcoins – “a “bitbills beta card.”  But he ruefully confessed that the company that originally issued these cards decided to get out of the business. 

You have to realize that when you use products and services to act as custodians of your money – albeit a new kind of digital money – you want to deal with folks you can trust.  Do you really know that the issuer of that smartcard hasn’t saved a card’s unique identifier codes for itself?  It’s conceivable that your money could be siphoned away in a flash – and then what would you do?  Can you really risk leaving your Bitcoins on your computer, where they might be hacked?  The Winkelvos twins, who are major investors in Bitcoins, say that they have theirs stored on memory sticks and locked in safe deposit boxes.  I wondered about the wisdom of Andresen walking around with eleven Bitcoins on his iPhone.  Those things get lost all the time!

Andresen said that he got his job as top scientist at the Bitcoin Foundation in 2010 by proposing a “Bitcoin Faucet” project that distributed US$50 worth of Bitcoins in a drip, drip, drip, back when each Bitcoin was worth half a cent apiece.  (That’s 10,000 Bitcoins, currently worth about $954,000.)  

The volatility of the value of Bitcoins got me to wondering:  Are they divisible?  The answer is yes.  Andresen said that a bitcoin can be divided into 100 million units, and the smallest unit is known as a “satoshi.”  This is in tribute to Satoshi Nakamoto, the pseudonym of the mysterious programmer (or group of coders) who invented Bitcoin.  Nothing much is known about Nakamoto, who claimed to be a 36-year-old Japanese man in 2009.  Andresen said that he never met Nakamoto himself, but he did have an email exchange with him.

One worry that Andresen has is the consolidation of ownership of Bitcoins by any single person or investor group.  The biggest protection against that outcome, he said, is to boost the transaction volume so that many, many people can become owners of Bitcoins.  This will help to assure their decentralization and reduce the risk of mischief.  It is important that ordinary people have relatively easy access to mining and trading Bitcoins, said Andresen.  Without that, the currency will not achieve its goals of distributing value (and money power) more broadly. 

For his part, Andresen said that he is involved in Bitcoin for the long haul.  No one knows how Bitcoin will fare in the years ahead, but there is little doubt that new forms of digital money will be launched soon – with social and political ramifications that we can barely begin to comprehend.

Comments

One issue that bothers me

One issue that bothers me that the bitcoin doesn't have any value for it self. when real currencies have real demand from people in order to pay their bills and taxes and countries around the world need the USD to pay for their oil,  the value of bitcoin is little vouge for me. it can be replaced in one day by better system , same as facebook replaced myspace. 

or that  one 0nline one 4 brands or by live4gambling my favourite link

Great Post! And: a possible solution to the 'hoarding' problem?

David,

Great post -- a very useful introduction to the promise of Bitcoin and similar technologies.  

One widely-cited impediment to the wider adoption of Bitcoin as a currency is the strong current incentive for owners of Bitcoin to hoard their Bitcoins.  While there are usually natural limits on hoarding of fiat currencies (if Narniacoins are the only accepted currency in the realm, then people must typically spend some of their Narniacoins in order to acquire food, shelter, etc), this may never be true of Bitcoin, and this will tend to bias its use towards speculation.  Perhaps this dynamic will be mitigated by the adoption of Bitcoin or similar cryptocurrencies by larger and larger communities, who'll simply find it convenient and secure to pay other members of their community in Bitcoin on a regular basis; I've no idea how to assess the likelihood of such a dynamic occurring. 

However:  following up on one of the links you posted (Roio, "Bitcoin, the end of the Taboo on Money"), I happened upon Freicoin:

A particularly interesting one is Freicoin, which grafts on ideas by Silvio Gesell for a monetary system with zero interest on credit: the value of currency “decays”, meaning that as time goes by it loses value. Freicoin cannot work as the storage of value, a common practice among Bitcoin users, therefore it circulates faster. By implementing this feature, referred to as “demurrage”, this is one of the most promising forks of Bitcoin today, at least in theory.

I find this 'demurrage' idea fascinating.  Just as the Bitcoin algorithm automatically adjusts the likelihood with which new Bitcoins are mined in order to yield smooth and predictable growth, one can imagine an additional algorithm that adjusts the value of Bitcoins (relative to the US dollars? or to what?) or the number of Bitcoins in any one person's wallet (as disconcerting as that might be) in such a way that hoarding is discouraged.  I know so little economics that I am sure to be missing some of the most obvious problems with such an algorithm.  One possible downside is that this sort of global fiddling with 'value' may be one of the things that Bitcoin's adherents are hoping specifically to avoid, as it seems akin to typical deflationary practices employed by states that use fiat currencies.  But perhaps a consistent, universally-applied algorithm that enforces a regular, inexorable decline in a cryptocurrency's value -- implemented, say, in Bitcoin 2.0 -- could provide all of the advantages of Bitcoin (anonymity, security, and all the rest) while also mitigating problems of unequal distribution -- quite a significant bonus! Heck: one could even design the algorithm so that everyone is provided with a basic, automatically-replenished number of Bitcoins in virtue of belonging to the network, and the algorithm would dictate that any coins acquired beyond that basic amount decay away quickly (either the coins, or their value).  This would still allow for large investments to be made; but they would typically need to be collective investments, financed by large numbers of people simultaneously, requiring all of the associated political coordination. Perhaps I just now described a crude version of crytocurrency-based socialism. Well, so be it:  sounds like a system with some neat features, and I'm not clever enough to spot the obvious flaws! Thanks again for your post -- I'll continue to follow up on the other links and readings it contained. Cheers,Don 

I too am excited about BitCoin!

As somone also promoting the commons, as well as the power of commoners, I too am excited about Bitcoin.  I see it as yet another pillar in this amazing "parallel society" being created by commoners, in which banks, and governments so deeply influenced by them, become increasingly powerless to stop the creation of a world that I'm convinced will actually be better for all.  Thanks so much for your contribution to this topic David!  I so enjoy reading your blog posts, and working my way through your course on the commons (via your podcasts). :)