xOn 23/06/13 05:38 PM, Jim MacKenzie wrote:
Does this mean your able to answer the full question "How do bit coins
work, and why should I care about them?" It seems everyone is able to
answer part one, no one can give a decent answer to the 2nd half
I'll try the 2nd half.
Short version:
The world has inefficient financial systems that information
technologies are now giving us the potential to overhaul and make better.
These two technologies can be introduced to provide new efficiencies and
open new opportunities:
1) Crypto (cash-like) currency
2) Crypto+p2p tracking and flow of credit (debt relationships)
I'm going to give a talk about #2 in September.
The inefficiencies in the financial system are not always transparent,
but once you recognize them, realize how they effect you, people close
to you, and your fellow human beings, and come to recognize that there
are now technologies if widely adopted that could make things better,
you might choose to help out with their adoption.
So, there's an element of altruism feeding them, but the further
developed and adopted these things are getting, the more people are
attracted to them for immediate efficiencies / direct benefits.
How much you're motivated by one or the other, how much your own uses
match up with that of other users, and how much you speculate others
will feel the same way all determines how much you should care. It's
understandable that different people care to different extents at
different times and this isn't something that needs to be uniform.
-------------
More thoughts:
1) Crypto currencies with cash like qualities. I'm going to decalre two
sub-types of these:
* Decentralized ones like Bitcoin and Litecoin that have a predictable
supply growth. Limited and predictable supply matters to people who
don't trust their state to manage currency supply well, a real problem
for some people in the world both historically and currently. (with
these currencies still being new, there is price volatility in them that
makes them only effective store's of value to patient steady hands)
* Crypto currencies that are issued by states as equivalent to their
physical notes, coins, and central bank balances. Here in Canada we have
a Canadian Mint experimental project alone these lines called MintChip.
Both types of crypto cash are desirable for all the reasons that cash is
desirable -- cash allows two parties to transact when it isn't possible
(or efficient) to deal with each other through direct or indirect credit
relationships.
Crypto cash removes the requirement of needing to exchange cash in
physical form. Even when a crypto cash transaction happens in person
it's not carried in an easy to steal form.
2) Crypto managed credit (debt relationships).
I'm not the sort of person that thinks all finance should be cash based
and that credit is evil. On the contrary, I actually think that striving
to be never trust anyone or any institution is an evil.
And I think it will be a real shame if too many of today's important
financial institutions lose all the trust people have in them by their
own fault. We'll have to build trust in new ones and this will be
painful and slow.
The trust that underlies credit relationships makes a lot of things
possible, particularly big capital investments like buying a house,
building an office tower, or obtaining business inventory with the help
of a line of credit.
/Debt: The First 5000 Years/ by David Graeber is a fairly fascinating
read on the history of credit.
http://www.mhpbooks.com/books/debt/
If you take as a given that credit relationships can be a good thing,
here's a problem to consider: Who should be responsible for the record
keeping in credit relationships and who should have the authority to
make adjustments to those records?
Not a problem at all if you're making a deposit with or borrowing from a
big financial institution -- you trust them to do a good job of tracking
your balance and making the correct adjustments when you make a new
deposit there.
But, there's a big social opportunity cost if only banks can be trusted
to do credit record keeping. Take for example pre-paying a merchant,
e.g. "gift cards" or "merchant scrip".
Why do people even give their money to dirty merchants in exchange for a
promise of future service? Why trust a merchant?
* When you give a gift card to a friend or family, you know there's a
good chance they'll use it and not sell it without a lot of effort and
slippage
* What if you really like their service and they offer you a discount
if you pre-pay for future service? I recall for example that Subway used
to have a "buy a gift card, get a free sandwhich" promotion and I went
for that a few times not because I was giving anyone the cards but
keeping them for myself as an effective 20% discount on total purchases.
There's no problem there with Subway -- I know they have one hell of a
system and aren't going out of business.
But, what about someone smaller? Does it inspire a lot of confidence if
they're issuing you a piece of cardstock with a guessable serial number?
(double spend attack just waiting to happen). Even if they invest in
something electronic, did they cut corners on the choice of vendor? Are
they getting murdered by their vendor? Is one of the employees messing
with the firmware in the middle of the night?
You might feel they're likely to stay in business. A community of
customers who loves that this business exists might even make some
significant pre-buys in exchange for some discount. (and the merchant
can kill off some of the high interest line of credit they have)
This is one of many situations where there's potentially a greater
willingness to extend credit if the record keeping system for the credit
isn't at risk of manipulation by either side.
Here's another example -- you and some friends go out for food and the
bill(s) arrive. Long story short, settlement complexities lead to at
least one friend owing another and not having the cash on hand to deal
with that. No problem, the two friends trust each other.
But, can they trust their own memories? Combine that with their fuzzy
memory of last weekend?
So, behold an innovation that allows us to have a reliable, trusted, and
neutral debt record keeper: Construct a decentralized p2p network to
track all the credit relationships, with cryptographically authorized
transactions and limitations and cryptographic validation by the network
nodes that it all makes sense.
The leading implementation of this concept is called Ripple (ripple.com).
The network nodes (validators) work with neighbours that they don't need
to trust per se, but trust not to collude with each other. If each node
has a decent number of diverse neighbours unlikely to collude then the
network will be robust -- there will be a hard to break consensus around
correct behavior and pockets of bad behavior will be identified and
routed around. Only massive collusion will bring the whole thing down --
a hard thing to do when a network has many diverse participants.
I like the way David Swartz describes this consensus protocol with his
"agreement room" analogy:
http://bitcoin.stackexchange.com/questions/7550/how-does-ripple-solve-the-double-spend-problem
This network is responsible for keeping track of everything, but
authorization for transactions fundamentally lies with creditors. It is
only by a message signed with the private key of a creditor that a
credit limit is extended to someone else. It is only by a message signed
by a creditor that they directly reduce a debt owed to them. (usually in
exchange for something they receive elsewhere in the network or outside
the system -- less often as a donation or forgiveness of debt)
You can also draw on credit that's been extended to you (by signing a
message of your intent to do so), but it's still fundamentally the
creditor that has allowed this to happen in the first place because they
authorized it in the first place with it already being in mind how and
when you're going to pay them back.
Now, here's the "woah!" part. Credit is more than just direct
relationships like a deposit and/or loan you have with your bank, an
amount a merchant owes you cause you have a gift card, or how much your
friend owes you from the other night. Credit balances can flow from
party to party to party as a means for payment to get from point A to
point B.
That's the rippling in Ripple. This older video is a little dry, but
does a good job of getting to the core of some of these fundamentals:
http://www.youtube.com/watch?v=xgGcVv04unM
Rippling of credit as a means of payment is actually at the core of all
our established institutional electronic payment systems. You can reduce
it all down to something like:
* Alice pays Carol by reducing her debt to Bob and getting Bob to owe
Carol instead.
(Carol has pre-authorized this by way of her credit limit with Bob, and
Bob has pre-authorized this by having at debt with Alice in the first place)
* Allice pays Carol by borrowing from Bob and asking Bob in turn to
reduce how much Carol owes him.
And so on an so on, pick your combination of credit limits, existing
balances, and as many billy bobs in between as you want. Note that Bob
isn't worse off or better off after each transaction, his assets and
liabilities have just shifted around within limitations he has accepted.
Such systems are rebalanced (debts paid off) in processes of "clearance"
or "settlement". In Canada this kind of thing is done by the Canadian
Payments Association and also our big banks all have direct accounts
with the Bank of Canada (a balance there is as good as cash).
Internationally we have the Bank of International Settlements (BIS).
What Ripple and similar systems are going to open up is the possibility
for much more diverse and efficient Bobs to pass through.
You can choose to be a "liquidity provider" (Bob) within safe limits
between some of your close friends. (Note, extending trust on Ripple
isn't like "friending" on Facebook, you don't just open yourself up to
any random person you might of met IRL once. And even with people you
know its good to have debt settlement terms)
You can also set yourself up between businesses that you believe are
honourable and set a transaction fee for credit flowing through you
between them. (be a Bob)
Many of the businesses expected to participate in Ripple early-on will
be professional money-changers/exchangers/transmitters with money
service licenses called called "gateways"
https://ripple.com/wiki/Ripple_gateway
The global remittance business will perhaps be the first big target for this
https://ripple.com/blog/disruption-target-1-remittance/
This is one area where crypto (cash-like) currencies and crypto+p2p
tracked credit will meet. There will be less credit based hops between
gateways when there also liquidity providers able to jump in-between and
re-balance their positions between gateways with crypto (cash-like)
currencies. Bitcoin in particular could play a huge role early on due to
it having the leading its market cap, market depth, wide range of
holders, and breath of implementations.
But is all of this a /revolution/? I'm not so sure, so I've said
"reform" in the subject line. We have a system of cash and credit, and
when these technologies mature and find real traction we'll have added
systems of crypto cash and crypto+p2p debt records on to them. I have my
doubts about economies becoming almost entirely cash based or almost
entirely credit based and I also have my doubts that state issued fiat
will die out and be replaced entirely with p2p crypto cash and p2p
stateless credit instruments.
You can support these things by working on the technology and promoting
it, but the most basic form of participation is to go go out of your way
to use these systems for transactions by identifying products and
services you'd want to buy anyway that are available by these new means,
charitable/non-profit causes you support that are payable these ways,
and if appropriate accept them as means of payment to yourself from others.
For this purpose, you might want to avoid the price volatility that
crypto (cash-like) currencies are going to continue to show while they
reach maturation. There's exchanges where you can hold a balance owing
in a conventional currency (CAD, USD) and just sell want you need to
obtain a crypto (cash-like) currency at the moment you need it
transactionaly. Crypto-based p2p credit systems like Ripple will also
play a role in helping people move credit balances out and through
various currencies in one clean step.
Same ideas apply if you're accepting it as a means of payment, one can
sell for something more steady and trusted right away.
Mark
p.s. Some may find my interest in Ripple surprising given that the
validator daemon is not yet released as free software. (its distributed
privately to small number of interested parties, not mass published).
The client code is free software and doesn't place real trust on the
validator daemon (rippled) -- most importantly, it signs it's own
transactions and the private key is only decrypted locally.
I take OpenCoin's commitment to publish it as free software eventually
seriously. I just don't recommend getting too carried away with adoption
beforehand, if you're in to it just get ready for that moment and the
adoption growth that will follow.
I also understand the goal of having a solid v0.1 cathedral in place
prior to poping open the doors and having a good old bazzar.
They've done an okay job documenting the protocol and answering questions:
https://ripple.com/forum/viewtopic.php?f=2&t=2785
Enough so that I've started working on what I call a "passive"
validator, a daemon that doesn't participate actively with other servers
in the consensus process, but watches the transactions they publish to
be sure they're signed with the claimed public keys and make sense in
terms of hashes, balances and credit limits.
Running this daemon you would be informed if the existing network of
rippled were acting dishonestly, stop transacting, and be able to
present the evidence to parties you have credit relations with so you
can wrap them up and leave the system.
With at least this in place, I'm going to be comfortable promoting some
small scale adoption and giving a presentation in September.
_______________________________________________
SkullSpace Discuss Mailing List
Help: http://www.skullspace.ca/wiki/index.php/Mailing_List#Discuss
Archive: https://groups.google.com/group/skullspace-discuss-archive/
