Bitcoin Forum

I'll be sitting on a "Future of Bitcoin" panel at the 2013 conference in San Jose, and I'd like some community input about what topics you would like to see addressed.

The poll reflects my biases and interests, and is pretty much what I'm currently hoping to communicate. If you don't see an important topic in the poll, please post it as a comment.

Please keep in mind that the conference attendees are very savvy, so I don't want to bore them with stuff they probably already know.

What's about "MasterCoin"?

At the moment, I think that “price stability” would be an important issue. Large fluctuations in price are not helpful, if we wish to use Bitcoins as a carrier of value between different currency areas.
I’m a little bit jealous that you manage to go to San Jose :) I’m living in Europe and the US are still a little bit far away…

Regards, Inge

Right now THE most important issue for bitcoins is the transaction auction system, It needs to be easy to set your own transaction fee to send/receive as the current one is INSANE.

It is insane only because Bitcoin-Qt implementation is sloppy as fuck. Standard fee is actually 0.0005 BTC PER KILOBYTE, not 0.0005 BTC per transaction!

Which means that for small transaction with one input you'll be paying 0.00014 BTC, two inputs cost you about 0.00025 BTC and so on.

Auction system is already implemented, each miner sets up min fee to a value he likes, and so your tx will be picked up by a miner who agrees to accept fee you provide.

In theory you can infer min fee for each miner from block they created.

While the fee structure needs to be revisted your belief that each input/output is 1KB is wrong.  Most tx are below 1KB.  A (2 input, 2 output) is usually around 440 bytes.  It varies slightly based on signature but no 2 in, 2 is anywhere near 4KB.

http://blockchain.info/tx/e51a0e55e5ca93ddef247a8fc31df0138e3837450fc03ef4bc7db741514573f7

http://blockchain.info/tx/47447cccaefd20aaf89e0365157556c6a1bb254ebecac67d0fe6c69dcacfdc56

http://blockchain.info/tx/9299451d48e9c93703ed5acdc97058dc3a4230238f3ae1f263a25799c59c473d

I hope to have the next revision of the 2nd bitcoin whitepaper ready to go with a close-to-final MasterCoin specification by then, but I don't want to be accused of using the panel as a platform to enrich myself. I may bring it up casually if it seems appropriate and not too self-serving.

For those who don't know, MasterCoins are a special kind of colored bitcoin, NOT an alt coin, so please don't start harping.

I know. Let me guess, you mis-counted zeros?


Yes, so the fee is 0.0005*(440/1000) = 0.00022 BTC. I wrote 0.00025 BTC above, rounded it up a bit.

You can implement ripple transactions with colored coins, but that's not the same
as "Colored bitcoins can do everything Ripple can".
You cannot, for example, ripple binding orders just with colored
coins, and Ripple.com can currently do that.
maaku and I are working on an extension for freicoin that will allow
this and more. Interest bearing currencies, for example.
Of course, that extension (or a functionally equivalent one) could be
put inside Bitcoin effortlessly (since we're on the same git tree),
but I think that hard fork will be more controversial for Bitcoin than
it will be for Freicoin. We're just designing for now.

I don't see why you shouldn't talk about mastercoin there.
What do you mean by "price stability" if it's not mastercoin?

I believe that colored coins will be able to do everything Ripple can do. Maybe not every feature right away, but there is no technical reason any Ripple feature can't be implemented with colored coins.

Colored coins will bring some price stability in that two people who trust the same issuer of USD colored bitcoins can conduct trade denominated in USD if they desire. That allows them the convenience and stability (for now) of USD and the anonymity and low transaction fees of bitcoin simultaneously.

Obviously colored coins are going to be a hotbed for fraud, although some institutions may have enough trust to issue USD, gold, oil, and other commodities and have people trust them. If people want to trade these currencies and commodities without trusting an issuer, that is when you need the MasterCoin protocol.

I hasten to add that none of this requires a hard fork of bitcoin - just some additional layers on top of the bitcoin protocol. Coloring bitcoins adds a very thin new protocol layer, but much more is coming.

Derivatives can also help with price stabilization.

Here's an article on colored-coin based CFDs. There is an example at bottom:


So basically one can use Bitcoin-denominated prices, protect against exchange rate swings via CFDs.

It requires less trust in issuer than USD-coins since CFD can be bought for a fraction of its nominal value.

There is also a way to make it more secure via escrow.

I totally agree if by "colored coins" you mean more than what they currently are and the new features can make changes to the bitcoin protocol.


Oh, I see.


Ok, now I understand better what you want to say and not.
I still don't see why you shouldn't talk about it though. And I'm still quite skeptic about the feasibility of the concept.
I guess advanced options (EDIT: derivatives) could remove much of the risk, but I don't see how can remove it all. I'll read the v2 of the design, maybe I change my mind.


Just the example I gave you requires a fork, unless I'm missing something.
My desire would be for you to "sell" the idea that these features greatly justify the changes that can be designed to be minimal.
If a consensus is not reached in the core development, it will be done in an altchain first.
Freicoin already needs a fork for having colors at all, since freicoin satoshis rot. We need colored coinbases.
So we plan to introduce additional changes to allow more features at that same time.
The experiment can be exported to Bitcoin later.

Of course, this is for the future, colored coins without fork can perform tons of useful use cases already.
What we want to solve mostly is interest bearing assets (or negative interest), like demurrage local currencies or loans.
And of course make it all "rippleable", even with pre-signed offers.
I also want private parts of the transaction ala Ryan's two phase commit. But this time not just using the chain for committing the external ripple transaction like I added to his design.
I want public and private parts integrated in the same Ripple transaction.
The last step could be to add an Open Transaction layer for complex contracts meanings and legal liability.

Well, those are my dreams, but I completely believe is all doable.
By the way, input is welcomed. We just thought that we're still changing too many things fast and may be very confusing if you don't know anything about how Freicoin implements demurrage using a refHeight field on transactions.
Not precisely readable and with outdated notes to clean up all around, but if someone wants to take a look (by the way the name is not definitive if the bitcoinX community doesn't want):

https://docs.google.com/document/d/1nnul3oDO5z8sspWBKgTKKSjQ7dWoOqU4Pd8DILLmFN8