Good Marc Andreessen piece on the broader implications of Bitcoin

Marc writes:

Think about content monetization, for example. One reason media businesses such as newspapers struggle to charge for content is because they need to charge either all (pay the entire subscription fee for all the content) or nothing (which then results in all those terrible banner ads everywhere on the web). All of a sudden, with Bitcoin, there is an economically viable way to charge arbitrarily small amounts of money per article, or per section, or per hour, or per video play, or per archive access, or per news alert.

Another potential use of Bitcoin micropayments is to fight spam. Future email systems and social networks could refuse to accept incoming messages unless they were accompanied with tiny amounts of Bitcoin – tiny enough to not matter to the sender, but large enough to deter spammers, who today can send uncounted billions of spam messages for free with impunity.

Finally, a fourth interesting use case is public payments. This idea first came to my attention in a news article a few months ago. A random spectator at a televised sports event held up a placard with a QR code and the text “Send me Bitcoin!” He received $25,000 in Bitcoin in the first 24 hours, all from people he had never met. This was the first time in history that you could see someone holding up a sign, in person or on TV or in a photo, and then send them money with two clicks on your smartphone: take the photo of the QR code on the sign, and click to send the money.

There is more here, interesting throughout.


Off-topic but, From the article:

"Bitcoin is the first practical solution to a longstanding problem in computer science called the Byzantine Generals Problem. To quote from the original paper defining the B.G.P.: “[Imagine] a group of generals of the Byzantine army camped with their troops around an enemy city. Communicating only by messenger, the generals must agree upon a common battle plan. However, one or more of them may be traitors who will try to confuse the others. The problem is to find an algorithm to ensure that the loyal generals will reach agreement.”

BGP meet the board game Battlestar Galactica:

I did my masters on the Byzantine generals problem and my phd in crypto. On the technical level he does not know what he is talking about. From a practical POV BGP was solved in mid 80's (using crypto). Consensus systems are old hat and had been supplanted in the mid 90's by other (more lightweight) protocols Bitcoin is a crappy protocol but a good PR move.

[citation needed]

How many of those solutions A) checked all the boxes the blockchain does, B) are easier to implement, and C) don't have hidden gotchyas? Because if there are any, you should totally issue a white paper for a cryptocurrency that uses them.

Exactly, if this was true jd could become a billionaire overnight by replacing the crappy protocol in bitcoin and creating a better cryptocurrency to replace it.

can you give an example of another solution being used in practice?

Paxos does not solve malicious agent problem, only missing agent.

Bitcoin is changing the world..long 1 million dogecoins ..ya it sounds like joke i know but there is serious money flowing into in these currencies

So, about $1,800 in dogecoins. Why do you think that's a good investment?

And I wonder if it's worth investing in a bunch of the most likely to succeed e-coins? Or will bitcoin dominate the market for the foreseeable future and the rest are all just doomed to be me-too's?

Randomly googled list of E-Coins:

"Or will bitcoin dominate the market for the foreseeable future and the rest are all just doomed to be me-too’s?"

Answered in the linked article. Did you read it?

I was asking for other opinions.

Imagine the world of content without the advertising factor…
maybe an 'average is over' one... premium ad-nodataselling-well-written content for the haves, click-bait-ad-track for the have-nots…
And would you really read all those ' ten reasons why.. ' lists if you had to pay a fraction of a penny each time?

Economists who attack Bitcoin today might be correct, but I’m with Ben and Milton.

All of a sudden, with Bitcoin, there is an economically viable way to charge arbitrarily small amounts of money per article, or per section, or per hour, or per video play, or per archive access, or per news alert.

Big disagree.

1. We've always has micropayment systems. Deposit $5, get 9,000,000 quatloos. Then buy articles 34 quatloos at a time.

2. Between wallets, Bitcoin has per-transaction fees. In fact, if you try to send small transactions amounts, you get more transaction fees imposed.

"Micropayment" is like "PKI" or "neural nets." 9 times out of 10, whoever is talking about them is full of it.

Micropayment offerings have always been rather poor in the US and never had backing by the big payments vendors. In China, however, they were so successful they leaked out of the QQ virtual box into the world of real products, became a threat and were consequently banned by the government.

The big players in technology (like Google) the big players in mobile service, and the big players in banks (like my bank) have all worked on micropayment systems, systems for paying with your phone, and systems for sending normal people money, but have had little adoption or interest. I have friends in tech who have worked on these systems, with little success in getting people to use them. If BitCoin proves that there's a market for these systems, those players will all just whip out and tweak their existing systems. I can see micropayments being big, but I don't see BitCoin winning that war.

I agree with you -- but we might be wrong, since it's an iron law of that small improvements in the technology can have large effects in terms of profitability and adoption.

The entire piece is a combination of sound and dubious claims. I guess Andreessen only has to be right in a small but hugely profitable percentage of the cases in order to make money from his investments.

Is Bitcoin actually good for micropayments? As I understand it the per-transaction overhead is significant both in terms of time and computational power. I think that Andressen's remittances argument is much more cogent than his micropayments argument.

You can implement micropayments in ways beyond the obvious 'send $X', you know. For example, you can do probabilistic payments: with probability 1/100, send $1 for an expected-value to the recipient of $0.01. (The recipient can verify the randomness with a suitable transaction relating to hashes: ) Same total value (asymptotically) with 1/100 the transaction fee of sending 1 cent a hundred times.

Sure, but in that case how is Bitcoin per se enabling micropayments any more than preexisting technology? The argument was "thanks to Bitcoin, micropayments can become a reality" which only works if Bitcoin is better at doing micropayments than, say, credit cards (with which you could implement similar randomness schemes).

Also I notice that the article you linked says, in effect, "to make this work we might need some hub servers to verify transactions for merchants." Which, well, QED.

> if Bitcoin is better at doing micropayments than, say, credit cards (with which you could implement similar randomness schemes).

*Has* anyone...? With this Bitcoin scheme, you can do it now. All you need is buyers and sellers to cooperate. No need to seek permission from an oligopoly which may not be pleased at supporting a scheme designed to reduce their revenue by minimizing transactions.


The cost per transaction is a few US cents, which is a bit steep for micro-transactions but a protocol agree to accumulate say USD 5 on trust before using bit-coin to finalise it. Or maybe a special-purpose, thinly traded bitcoin clone could be used for the purpose, with fees kept down to near zero. Or maybe bitcoin fees will drop as volumes increase.

I would also agree. Even if there is no micropayment system in place now, there is nothing in the nature of traditional money which prevents its existence. If there is sufficient demand, financial institutions could start processing payments in amounts less than one cent, be it 0.1 of a cent or 0000.1 of a cent. Also, there is nothing in bitcoin which would make bitcoin transaction costs intrinsically lower that familiar dollar transaction costs, so this barrier would be equally important both to bitcoin and dollar micropayments.

I sort of agree. I admit to not fully understanding bit coin, but in reading the article I don't see a lot that couldn't be implemented using dollars.

It might be just as wonderful as Andreessen says, but I don't quite see it.

"2. Between wallets, Bitcoin has per-transaction fees."

Does it? According to this there is no per-transaction cost per se, but a generic pay the data recorders on a periodic basis.


I don't know what that guy is going on about, but I myself have lost bitcoins to transaction fees when moving money between two wallets.

Bitcoins are horrible for moving a few cents around.

The documentation on bit coin wallet transaction fees seems vague. So, based upon your actual real world experience, I'm guessing it's a case of Buyer beware.

This appears to be the official policy:
"Transaction fees may be included with any transfer of bitcoins from one address to another. At the moment, many transactions are typically processed in a way where no fee is expected at all, but for transactions which draw coins from many bitcoin addresses and therefore have a large data size, a small transaction fee is usually expected. "

It would be nice if they defined small.

The way it works is you can choose what transaction fee (or none) you want to include with the transaction. The miners who create new blocks can decide if they accept new transactions or not. They are incentivized to include transactions that include a fee, so if you include a fee your transaction will be confirmed quicker.

What about PayPal for micropayments? can't you send a dollar with Paypal, if you have it linked to your bank account? I've sent as small as $5. But maybe the vendor gets charged something by Paypal, now that I think about it.

Paypal extracts a fee to recipient except in cases of gifts. It's about 2.9% just like credit cards. (Or in some cases it exempts fees, like for charity site like

There's a company called dwolla (that is in some way associated with bitcoin I think) that has a system similar to paypal but much smaller transaction fees, and no fees for small transactions.

I agree with Dan_Weber: these are bad reasons to be optimistic about bitcoin, except for the last. Bitcoin is good for making small direct donations, but for everything else, Bitcoin doesn't help much.

- Media companies would be better off banding together and do an "all you can eat" model, like cable companies, rather than trying to sell each individual article, no matter how small. This is because consumers don't know whether it's worth reading until they spend the money (the choice of which is disproportionately taxing).

2) You don't need Bitcoin to impose a fee on email; in fact, bitcoin arose from the *prior* discovery of how to make people bear a cost in order to send you stuff, and you can just strip that part out and use it as the spam prevention measure. That part is the "proof of work" concept.

The way it works is (basically) that you generate a random n-bit string ("nonce"). The sender must find an input string which, when put through a cryptographic hash function (like SHA256), yields an output whose first n bits match the nonce you gave them. If it takes you X steps to verify it, it will take the sender about (2^n) * X steps to solve it.

Bitcoin uses that to verify which ledger has the most work, but you don't need the whole system to implement a work verifier.

> Media companies would be better off banding together and do an “all you can eat” model, like cable companies, rather than trying to sell each individual article, no matter how small. - See more at:

Well, we don't know if this is true or not, since the current finance system has made it impractical to buy individual articles. In the next couple of years we'll hopefully know the answer to this with certainty.

I'm with Silas and Dan on this one. The micropayments argument reminds me of an article that TC posted a while back that had an insightful observation (paraphrased as best I can remember it): Having solved a difficult technical problem, the inventors of bitcoin simply can't believe that they haven't also solved an economic problem. In this case, people don't hate micropayments because of any technological factor. They hate them because deciding how much reading an article or listening to a single song is worth is itself a cost, and it's often higher than the value of the service you're considering.

Because of the current 1Mb maximum block size for the Bitcoin protocol, the maximum number of transactions works out to about 7 transactions per second. Although there is by default a small fee, you can currently send transactions for free if you overide the fee, and they eventually will be recorded in a block (although it may take some time). However, once Bitcoinn regularly starts hitting the block size limit, senders will have to start increasing their transaction fees to make sure their transaction goes through, as miners will just collect the highest fee-per-byte transactions until they hit the block size. I'd love for an economist to figure out what will happen next!

The 1MB max block size is an artificial limit. It can be raised by simply updating the protocol.

I agree, but how do you convince people who've invested money in ASICs to go along with a protocol change? It's in their interest to have higher fees from a limited block size to recoup their costs, and increasing the block size would prevent fees from rising. And it's exactly these people who get to decide what program is running on their hardware.

The larger the block, the more transactions and thus fees are included in each block as a reward. It is in the miners best interest to allow for bigger blocks because that means that can collect more fees from a volume perspective. Remember, the bitcoin reward for solving a block is going down, so miners are going to have to rely more on fees for their reward. That means they need tons of transactions with small fees if they want to collect big fees from the blocks solved. It is in their best interest if blocks get bigger since they will make more money that way than trying to artificially keep the block size low and have people jump ship to another altcoin due to high fees on the bitcoin network.

All of these sound like ideas that can be implemented with USD (or tiny fractions thereof).

Except that about 200,000 pages of government regulations govern USD transactions.

It's good to see Tyler slowly educating himself about Bitcoin over time.

There is a micropayments protocol in the works for Bitcoin.
One tx per micropayment isn't really a practical solution.

The economics of tx fees are another matter. It currently costs an average of $50 to process each tx on Bitcoin, costs subsidised by the seigniorage of new coin. As that halves every 4yrs, either tx volume must go way up, tx fees must go up,  or network security must go down. Raises the question as to why most seigniorage is spent rewarding existing balances with a richer and richer exchange rate.

I thought the security of Bitcoin depended on the uniqueness of the transaction ledger (preserved by the merge algorithm favoring the longer chain, etc.) What happens to this when there are a trillion micropayments per day?

I wonder the same thing. Since "every transaction in the Bitcoin network is tracked and logged forever in the Bitcoin blockchain", that thing must get huge over time. It's already some 13GB in size: And it takes considerable amount of time to confirm a transaction, up to 20 minutes:

ok, I just learned that the developers have thought about this and there is a method for blockchain compression, there has just not been any algorithm implemented so far:

Bitcoin people call those microtransactions "dust" and they work hard against them.

A counterpoint article to Andreesen's should be included, too:

On the Matter of Why Bitcoin Matters
Marc Andreessen was a big part of turning the Web into a mainstream experience, but seems to misunderstand Bitcoin profoundly, writes Glenn Fleishman.

Tyler- It seems from your previous posts that you discount the value of "network effects" that Andreessen thinks will protect the value of Bitcoin against rival crypto-currencies. I happen to think you're right about that. Also, I think Andreessen's arguments gloss over lots of practical difficulties. For instance, he says that you can protect yourself against fluctuations in the Bitcoin exchange rate by only purchasing Bitcoins when you need to make a transaction. This is true enough, but it also adds a non-negligible friction in the form of transaction fees and bid-ask spread to each transaction. Probably at least enough to make the use of Bitcoin for micro-payments uneconomical. Once you factor in these frictions, it seems likely that existing wire service providers will be able to evolve their offerings in terms of convenience and cost to compete with the Bitcoins of the world.

Bitcoin doesn't solve micropayments, because the problem with micropayments isn't really the infrastructure, but the fact that every action has a price. You don't want to make paying for something too easy, due to all the fraud issues, and yet if you make it complex enough to make sure your user is aware that he is paying for something, the experience becomes just convoluted enough to be a chore.

Think, for instance, of how micropayment equivalents work in the Apple ecosystem. If you support plain one click buying, you get children spending thousands of dollars on things without the parent's consent, or any awareness of the consequences. If you add a password, then microtransactions are a chore. Apple will request your password every 15 minutes or so, but their system is only good because your average ipad is relatively secure: No unsigned apps and such. But imagine doing that on a PC: Visiting the wrong page installs a small browser add-on that, the moment it detects you have authorized your 15 minutes no password window, starts draining your account in the background. Yikes.

And remember, bitcoin transactions aren't undoable: you can't really go back on the ledger. If someone spends 1000 by mistake, or is scammed, the credit card company and/or Apple can do something about it. Same thing happens to your bitcoin wallet? It's like stealing your hard currency, and the fraudster wins.

So once we take reality into account, bitcoins are a terrible way of doing micropayments.

Isn't there a transaction cost to convert bitcoin to dollars, or is that an exchange rate without a bid/ask spread of any kind? Also, it strikes me that bitcoins suffer from considerable exchange rate risk. I may be missing something, but suppose someone told you that you could make free or nearly free electronic transactions, including microtransactions, but the catch was that the e-currency was the Icelandic Krona, and Iceland had decided to constitutionally and permanently freeze the quantity of these Krona. The in and out costs and risk of sitting on the Krona would certainly keep this from becoming a major transactional currency, wouldn't it? And if so, how does that differ from bitcoin?

Comments for this post are closed