Something with the math just doesn't add up for me

[Waler]

Hey everyone! How are you all doing?

I'm sure that SOMEONE SOMEWHERE has already brought that up, but I failed to find any answer.
How can bitcoins scale?
I mean... I love the idea of bitcoins, and I'm fascinated with the idea of p2p money. I think the future really is here.
But the "limited supply" part of bitcoins bothers me because I fail to see the ability to scale it.
I know, I know, people have talk about deflation and stuff like crazy around here - but I don't care about deflation. Just about running out of money.

Here's why something doesn't add up for me. Let's imagine a perfect scenario for a moment - no wallet are lost, all bitcoins have been generated and 21 million bitcoins are now the main currency of the entire world.
With 6 billion people in the world (let's ignore population growth for now), using the full bitcoin currency is impossible so people have naturally moved to using satoshis (0.00000001 bitcoin if I got it correctly) which means ratio between people and money is 350,000 satoshis per person.
Sound like plenty, right?

No quite. Given that we are using satoshis (moving down a decimal point is inflation which is against the bitcoin philosophy), the cheapest thing ANYONE could EVER by, will cost 1 satoshi. I mean less than bubblegum cheap. And given that there will still be people how are rich and people who are poor - buying a house, paying for food, having electricity and everything that we need will be insanely costly.

And that - in an ideal world.
Add population growth to the mix. Now more people need to share the same amount of bitcoins.
Add most importantly - add losing bitcoins to the mix. Bitcoins have already been lost. And we will keep losing them without any means of returning them from the abyss.
And so, bitcoins will be slowly lost and the population will slowly grow, the no-one will have any money at all.

Now, while I like bitcoins (A LOT), I think that this is a major flaw in the design. Maybe I'm missing something?

Hopefully, the next generation of bitcoins (or another really good cryptocurrency) will have build in a way to scale up. Maybe with linear growth?

Could anyone share his thought on the subject?

[CIYAM]

In all likelihood we are going to see the emergence of at least one or two other Bitcoin currencies (such as LTC) that will be "silver" compared to "gold" (and these may well have closer to infinite supply).

So such scaling is not really going to matter that much as I'm pretty sure a very large proportion of the world's population do not have any significant amounts of gold also (i.e. think of Bitcoin as being digital gold).

P.S. Welcome aboard!

[LordMord]

If we really need to we could divide Bitcoins/satoshis even more. So that is not really a problem.

[John (John K.)]

If we really need to we could divide Bitcoins/satoshis even more. So that is not really a problem.

This. There has been talks on the exact topic before this, about the future where a satoshi would be worth much more then now.

[Kazimir]

If we really need to we could divide Bitcoins/satoshis even more. So that is not really a problem.

Yep, if I'm not mistaken, even in the current specifications bitcoins can actually be divided in up to 16 decimal places (although the current implementations only use .

So essentially there is a maximum of 2.1x10²³ units. Which is quite enough for the foreseeable future. Even with losing coins and growing world pop

And if we'd really have to, we can even increase the number of decimals beyond that, by common agreement on the protocol.

[realnowhereman]

It does come up a lot; and mainly because people (perhaps not familiar with low-level protocols, and why should they be?) worry too much about an easily changed detail.

https://en.bitcoin.it/wiki/Protocol_specification

First, the developers would increment this number:

tx
tx describes a bitcoin transaction, in reply to getdata

Code:

Field Size  Description   Data type   Comments
4           version       uint32_t    Transaction data format version 

Then they'd add a new spec for that higher version of transaction by changing this field

The TxOut structure consists of the following fields:

Code:

Field Size  Description   Data type   Comments
8           value         int64_t     Transaction Value

so that "field size" was 16.  Ta da, more resolution available to subdivide some more and all the existing coins worth exactly what they always were.

In truth, even this wouldn't be necessary.  About 11 bits of the value field are currently wasted.  There can be at most 21 million bitcoins in one transaction, or  2,100,000,000,000,000 satoshis (2.1*10^15).  The biggest number you can store in a 64 bit field is 18.4*10^18 (approx).  You can see then that version 2 transactions could simply be that the scaler for the "value" field get changed and we gain a factor of 1000 ish in resolution with no change of width.

In short: don't worry about it.  It's going to be a long time before it's a worry; bitcoins would have to be worth more than a million dollars each before we even need to think about this problem... so if you're really worried buy one bitcoin now and you can fund the development when the problem occurs.

[Waler]

First off, thank you guy for reading and commenting. I appreciate the time you gave reading what I had to say.

John, kazimir, realnowhereman and lord, I actually talked about dividing up the satoshis firther. From what I can see, that's exactly the inflation the current bitcoin system is trying to avoid - creating more money. However, if bitcoins will catch on as a main currency, I guess that will be inevitable.

CIYAM, I checked out litecoins (and have a shiny wallet to prove it) - while it is the "silver to the bitcoin gold", it also faces the same problems. It doesn't come close to infinite supply (just X4 the cap of bitcoins) that a growing population of people that also slowly lose the currency will need.
The difference between bitcoin and gold is that if someone lost gold (for example a ship carrying it sunk) - someone else will eventually find it. It's not gone, just hidden and unused. There has to be a way to "find" that lost gold, or eventually it will all be gone...
I love the concept of "gold" and "silver". What about taking the idea further and adding a third coin, the will have infinite (yet VERY slow) regeneration? It could handle the lost coins without creating inflation, using it for the daily transactions and leave the expensive things to the BTC and LTC.

However, in the case of building the economy on dividing up to 16 decimal places, it might be a while before it's going to be a real problem... Hopefully we'll have a solution by than.

[LordMord]

Don't see how that is inflation. Your money is not worth less you can just send smaller amounts of it.

[realnowhereman]

First off, thank you guy for reading and commenting. I appreciate the time you gave reading what I had to say.

John, kazimir, realnowhereman and lord, I actually talked about dividing up the satoshis firther. From what I can see, that's exactly the inflation the current bitcoin system is trying to avoid - creating more money. However, if bitcoins will catch on as a main currency, I guess that will be inevitable.

No; inflation would be issuing more than 21 million coins.  Increasing the resolution with which you can transmit coins is just that, no more.  Money wouldn't be created it's just expressed in finer divisions.  A tonne of gold would still be a tonne of gold, no matter how finely you divided it up (although physics puts harder limits than the bitcoin protocol).

Imagine at present you could only send whole numbers of bitcoins.  Would it be inflation if the developers added a feature to enable you to send 0.1s of a bitcoin?

Let's say that we want to be able to deal at a resolution of $0.001 (I'm picking this as the smallest price we care about); but with bitcoins.  The current protocol allows for as little as 0.00000001 BTC to be transmitted.  Therefore the highest exchange rate the current protocol limits us to (for real-world considerations) is: 0.001/0.00000001 or $100,000.

Changing the protocol minimum resolution to 0.0000000000000001 BTC wouldn't change the value of any existing bitcoins.  Instead, at a BTC/USD price of $200,000, we would still have a way of expressing $0.001.  In fact, at that resolution, we would have to reach BTC/USD price of $10 trillion before we lost the ability to describe one thousandth of a dollar.

It's not a worry because a price of $100,000 per bitcoin is not going to just sneak up on us; (if we ever even reach something like that it would make the bitcoin economy worth 2.1 trillion).


Edited to add:

I think you're mixing up inflation (the money supply) which will always be measured in whole units and divisibility which is (mathematically) infinite.

Bitcoin's money supply is asymptotically tending towards 21 million.  That is neither deflationary or inflationary.  If the world bitcoin economy grew then we would expect to see price deflation as a bigger economy would be divided over a static supply.

People losing bitcoins isn't a problem because they simply make the rest of the holders of bitcoin a tiny bit richer.  The value isn't lost to the economy, just to the person losing the coins (and that's their problem).

[Waler]

I think it's a bit clearer now.
Thank you very much!