letstalkbitcoin #20

20.1: The How and Why of localbitcoins.com
0:12
Hi and welcome back to episode twenty of Let's Talk Bitcoin!, a twice-weekly show
exploring the ideas, people and projects building a new digital economy and the
future of money.
0:21
Twenty episodes. That number came up fast.
0:25
Today's an experiment: one of the most common complaints we get is that our show
is too long and some listeners would rather have the ability to pick and choose
what order they listen to the show in.
0:33
It's also very intimidating for new users who might be interested in a
particular segment but not in the half-hour discussion that came before.
0:40
Episode twenty will be released a number of parts. You should find a total of four
files.
0:45
In 20.1, that's what you're listening to now, Stephanie and Andreas share their
experiences with localbitcoins.com, an alternative to centralize exchanges
that's not against the rules.
0:57
In 20.2 the hosts have a long conversation about what the point of mining is and
if the incentive might have become the end result, losing sight of Satoshi's
vision. This is a must-listen-to segment, and clocks in at a little under half
an hour.
1:09
In 20.3, I sit down with Asher Ten and Ryan Joe, better known as ZhouTong, two
of the founders behind coinjar.io, a new Australian Bitcoin exchange. At the end
of this segment you can hear Andreas on Bloomberg West yesterday.
1:24
In 20.4 Andreas tells us about the upcoming Bitcoin Improvement Project Proposal
number 32 which proposes a new third type of wallet, a hierarchical
deterministic wallet. Sounds technical, and it is but the concept is powerful
and it's worth your time to try to understand the potential here.
1:40
We appreciate you hanging with us as we try these new things out.
1:41
Please send all your feedback to [email protected].
Enjoy the how and why of localbitcoins, from the team at LetsTalkBitcoin.com.
1:52
Stephanie: I received an email from a listener who was curious about how to get
Bitcoins in general, and said that he was having some real difficulties just
getting his hands on some Bitcoins and I know that this is a common problem that
a lot of people have.
2:05
He was specifically asking me about the website localbitcoins.com, which we've
talked about on the show before.
2:12
Personally I think with all the difficulties with exchanges and Bitcoin websites
you know, more options are popping up all the time to be able to get Bitcoins
and we've talked about a bunch of them like Coinbase and BitInstant and Bitstamp
and BTCQuick and all these different websites where you can buy Bitcoins, but
you've got to go through a verification process and link your bank account in
some cases not everybody wants to do that, sometimes they just want to buy
Bitcoins although you can do that on BitInstant, but it can be a little more
complicated; you've got to find a location and so forth and he was curious about
how to use LocalBitcoins to just simply buy Bitcoins with cash. There is a
website called localbitcoins.com where you can find people in your area who are
buying or selling Bitcoins and hook up with them through this website and then
you can broker Bitcoin transactions in person.
3:06
You can find people who are buying or people who are selling, you can be a
seller on localbitcoins, you can buy Bitcoins online even, there are online
sellers that are offering Bitcoins for various forms of online payment like
OKPay or Dwolla or PerfectMoney.
3:23
Localbitcoins.com itself is based in--
Andreas: It's in the free and democratic country of Finland.
3:30
Stephanie: Finland, yes thank you.
3:32
Localbitcoins itself is not based in the US, what it is right now is a way to
hook up with other people who want to you buy your Bitcoins or sell you
Bitcoins. What they have is like user profiles, it's kind of like e-bay, there's
a feedback system. They have an escrow method as well where it provides some
protection for both parties so that they can hold the Bitcoins in escrow and
then the seller can release them when they get the payment going through.
3:56
There is a little bit of protection built into it and then there's this feedback
system also. You probably want to do business with people who have positive
feedback and have a bunch of instances of positive feedback if you're going to
use localbitcoins.com and of course all the things that apply to for instance
online dating or anything where you would meet somebody that you don't
know apply to this. You always want to me up with a person in a public location
that's crowded, in case anything goes sour, whatever, there are at least a lot
of people around to witness what's going on so a lot of people meet up in
internet cafes or restaurants or places that are a lot of people in the
daylight.
4:37
You can put down hours when you're available and places that you want to meet;
the sellers will put their terms in their advertisements on localbitcoins.
4:45
You know honestly you can meet some really cool people this way, people who are
interested in Bitcoins. Not everybody who's out there on the internet is a
monster or shady or anything like that, a lot of people are just regular people
and they're very nice and they'll sell you some Bitcoins and they'll help you
out maybe even getting started setting up a wallet. I mean, some people do sort
of offer that, support, as well as being Bitcoin "dealers" will call it on
localbitcoins. You know i think it's possible to have a really great experiences
with localbitcoins if you are a little bit savvy about those things.
5:16
Just deal with people who have positive feedback, make sure they're
communicative with you, meet them in a public place in daylight; I think those
are the pointers that I would want to give. Do you guys have anything you want
to add to that?
5:26
Andreas: I don't believe that there's any sort of legally questionable in the
US. I believe the FinCEN guidance that came down a couple months ago
specifically said that so long as you're not doing this as a business,
but rather it's just person-to-person transactions, that that's something that
doesn't fall under their guidelines.
5:44
Announcement: Let's Talk Bitcoin! is heard each week by thousands of people who
are participating in the new digital economy.
Our listener base of Bitcoin owners, miners, investors, technologists, and
merchants is growing fast. We offer a limited number of short advertising slots
in each show to keep our listeners engaged and to provide maximum impact our
sponsors. If you'd like to talk to us about Let's Talk Bitcoin, send us an
email at [email protected].
6:11
Advertisement: If I showed you a website where you could easily purchase
electronics from the world's largest distributor with Bitcoins at zero percent
markup, would you think it was too good to be true? Good news: it's real, and
it's at BitcoinStore.com. Choose from half a million items, save money over
Amazon and Newegg, and convert your Bitcoins to real-world items. You can even
buy with privacy; all they need is a shipping address. But don't take my word
for it, see for yourself at BitcoinStore.com.
6:45
Andreas: So I did number a number of trades on localbitcoins.com, I have seller
accounts with a lot of positive feedback. I use it primarily for market research
because the best way to understand who's buying Bitcoin is to meet them.
6:59
So I also offered a service to help people set up wallets and to explain the
issues to them, and then I met more than a dozen people from every age, every
professional background, every political bent, different kinds of interests et
cetera, all very pleasant very nice interactions and no problems whatsoever. I
would also emphasize what Stephanie said, to keep in mind that as a buyer you're
the one holding cash and going to this meeting with the other person knowing
you're holding cash, so don't get mugged.
7:29
I did have one occasion when a younger female buyer had a handle or a pseudonym
hasn't made it very obvious that she was a younger female buyer and I advised
her to change her pseudonym because it's it's really not safe just like it
wouldn't be safe to be meeting people on Craigslist and advertising things like
that.
7:48
But other than that if you take those basic precautions and meet in a public
space I think this is a brilliant way to get Bitcoin, to meet other Bitcoin
people, and to establish relationships that survive regulators basically.
8:02
Adam: There's a variety of different fees that participants on a localbitcoins
offer, what do you guys think, you know Andreas when you were doing this what
sort of upcharge were you charging, and what do you think is reasonable?
8:13
Andreas: I was doing something slightly different which was I was only selling
up to one Bitcoin or actually about a hundred dollars worth of maximum purchase,
it was only for new users, new buyers so I wanted to see an account that didn't
have prior purchases so I actually offered them at a discount over the exchange
rate, most people add a slight premium and usually you do it with a formula so
you say, I want to take Mt Gox 24-hour price, or a blended exchange 24-hour
price and then add a ten percent premium or something like that.
8:47
But it kind of depends. You'll see different prices throughout the day and they
change dynamically a lot of the time.
8:52
Stephanie: There are a number of different formulas that you can use or you can
just do like a flat price over spot for Bitcoin, or under spot, I mean you can
really do whatever you want if you're a seller and if you're a buyer you can find
the best price that's the most convenient for you where you live or you know
depending if you want to buy Bitcoins online you could send somebody money
online if you're comfortable with that.
9:14
So it's really a pretty varied marketplace, it's a lot like e-bay I think. I
just want to mention that there is great tutorial about using localbitcoins.com
to grow Bitcoin adoption and this is about how to become a seller on
localbitcoins and basically do exactly what you were doing Andreas, with trying
to help people get interested in Bitcoin and get started with it and it's by a
friend of the show who's been on the show, Justus Ranvier and if you google
search using localbitcoins.com to help grow Bitcoin adoption you'll see a post
about that.
[Article: http://localbitcoins.blogspot.com/2013/06/this-is-first-in-series-of-posts.html]
9:47
There is actually one more thing that I wanted to mention about localbitcoins,
you can bypass localbitcoins if you have basically a friend who you know and
trust already who wants to sell Bitcoin or buy some Bitcoins from you and you
know it's similar to localbitcoins, but you wouldn't have to pay their fee
although it is very small, you know they do have a small fee for the sellers I
think and if you have a friend who's willing to do Bitcoin trades with you I
would prefer that option because it's somebody that you already know and trust
and you don't have to worry about any potential on safety issues with that.
10:20
But for those who don't have any friends who are into Bitcoin who are local to
them and want to sell or buy, localbitcoins is a great option.
Andreas: Or just get your friends they get all the games start buying that going
from you if your friends are not into Bitcoin you're doing something wrong.
10:39
Thanks to Dr. Stephanie Murphy, Andreas Antonopolous, and the listener who
emailed Stephanie providing content for the show.
10:45
Please send all listener mail to [email protected]. Especially this time
since we're trying something very new here I'd like as much feedback as possible
so chime in. Thanks for listening





20.2: Mining, and the Forest for the Trees
0:12
Hi and welcome back to episode twenty of Let's Talk Bitcoin!, a twice-weekly show
exploring the ideas, people and projects building a new digital economy and the
future of money.
0:21
Twenty episodes. That number came up fast.
0:25
Today's an experiment: one of the most common complaints we get is that our show
is too long and some listeners would rather have the ability to pick and choose
what order they listen to the show in.
0:33
It's also very intimidating for new users who might be interested in a
particular segment but not in the half-hour discussion that came before.
0:40
Episode twenty will be released a number of parts. You should find a total of four
files.
0:45
In 20.1, Stephanie and Andreas share their experiences with localbitcoins.com,
an alternative to centralize exchanges that's not against the rules yet.
0:57
In 20.2, that's what you're listening to now, the hosts have a long conversation
about what the point of mining is and if the incentive might have become the end
result, losing sight of Satoshi's vision. This is a must-listen-to segment, and
clocks in at a little under half an hour.
1:09
In 20.3, I sit down with Asher Ten and Ryan Joe, better known as ZhouTong, two
of the founders behind coinjar.io, a new Australian Bitcoin exchange. At the end
of this segment you can hear Andreas on Bloomberg West yesterday.
1:24
In 20.4 Andreas tells us about the upcoming Bitcoin Improvement Project Proposal
number 32 which proposes a new third type of wallet, a hierarchical
deterministic wallet. Sounds technical, and it is but the concept is powerful
and it's worth your time to try to understand the potential here.
1:40
We appreciate you hanging with us as we try these new things out.
1:41
Please send all your feedback to [email protected].
Enjoy mining for fun, profit, or the good of the land, from the team at
LetsTalkBitcoin.com.
1:55
Adam: Dan Kaminsky is a well-known security researcher and was recently on a
panel talking about the security of Bitcoin at the Bitcoin 2013 in San Jose last
month.
2:05
One of the most interesting things I think that came out of the entire
conference was actually something that he said during that talk and I want to
play the clip now so we can kinda discuss the implications.
2:16
[clip]
Dan: I assign zero percent probability that we will continue with the
present proof-of-work function.
2:23
The present proof-of-work function is not going to survive the year, period. If
there's one hard prediction I'm going to make, it's going to be that.
2:32
The reality is that's it's impossible to model the--
2:35
Look, you have to think of the system in terms of regulatory steps. One dude
spins up one batch of ASICs and gets double-digit percentages of the network.
2:49
That's ONE guy. That's not okay. Bitcoin depends on there being so many freaking
people to regulate that you can't do it. That when you try to regulate it, you
don't get power; you just get more people. Bitcoin uses, by design, was designed
to leverage the hundreds of millions of machines around the world that do
computation. That is an impossible set to control.
3:18
One dude? Not an impossible set to control. We think that the ASIC guys love to
say "Hey, we'll just have lots of people buy ASICs." Yeah, you like that
argument because you sell ASICs.
Other vocie: I mean, that's okay, there've always been economic motivators for people to make
their mining things
Other voice: I mean with ASIC even in the best case you're going to see a
massive industrialization of mining to a few small players. It's not going to be
an ASIC in every garage.
3:42
Kaminsky: if that happens, Bitcoin fails. I'm not sure what the proof-of-work
function, or functions, are going to look like. I had a great meeting, let's
just say I had a meeting with some people and it became clear that multiple
proof-of-work functions operating as a basket of currencies is probably the path
that's going to happen. I don't know where this is going to go yet, all I know
is that Bitcoin is at a life-or-death dependency on mining not centralizing.
4:20
You have one mining pool, BTCGuild, with 48% of the power. You think this is an
accident? You think they just "oh, you know what, we just ran out of power just
before 50%"? No! They're like "okay, everyone realizes that if we ever exceed
50%, the value of everything goes to zero, so we're going to stop here".
4:43
This aspect of the system has got to change. I'm not sure what it's going to
change into. Either Bitcoin changes, or something else happens to leverage the
hundreds of millions of machines, because right now if you have a CPU or GPU,
you shouldn't be mining.
4:58
Adam: So the question is, if we could turn back the clock and take this move
towards GPUs and ASICs and make it go away, make it so that that is not
something that is feasible any longer, is that is that a good thing, is that a
bad thing, I mean are ASICs, are they a curse, are they a blessing?
5:19
Andreas: Dan is absolutely right this is absolutely a critical problem for Bitcoin,
it is one that represents the greatest threat to the survivability of Bitcoin,
and I'd like to quote Saint Satoshi himself, this from page 3 of the Satoshi
paper:
5:37
"The proof work also solves the problem of determining representation in
majority decision making. If the majority were based on one IP address, one
vote, it could be subverted by anyone able to allocate many IPs. Proof-of-work
is essentially one CPU, one vote."
That's what Satoshi said. That was the vision. And that vision has very much
been subverted by ASICs because essentially what it says now is that the power's
getting concentrated.
6:05
The problem isn't so much the fact that people can buy ASICs and they're
expensive, the problem is that the ASIC is three, four, five, ten orders of
magnitude more efficient than CPUs and it basically crowds out everybody else so
centralization destroys neutrality, provides a single targets for regulators;
centralization is the thing that destroys Bitcoin.
6:32
Absolutely ASICs are a problem, the current proof-of-work system in the end ends
up centralizing too much power into too few miners and violates the basic
precept of Satoshi, proof of work is essentially one CPU, one vote -- no longer
the case.
Stephanie: I can't really blame them as a problem, I know you're not blaming
them Andreas, I think they're kind of inevitable.
6:53
What I mean by that is that there's such a great incentive to find a more
efficient way to do the proof of work, the pressure is so great and the
incentive is so great in there so much to gain by developing some technology
that helps that happen, that of course ASICs come out. If you know maybe
there'll be something even more powerful than ASICs in the future. I think I
agree with you that there is a potential problem there with the centralization
especially since ASICs right now are so expensive.
7:19
Definitely centralizing power. We've seen some ASICs come out, but there have
been so many problems with them I guess there's still some room for skepticism
right now like there's still room to talk about this and say we can still
prevent this problem before it really gets bad.
7:32
There hasn't been a huge hitting of the market so to speak of ASIC miners yet.
How could you even do that was Bitcoin? We can't turn back the clock; the only
thing that could be done I suppose is create an other cryptocurrency that is
somehow literally unable to do anything but CPU mining, and I know that there's
some folks who are interested in doing that, but then you know you have to get
people to adopt that and use that and value it, and will they when there is such
a strong profit incentive to go with Bitcoin and the ASIC miners which have the
potential to centralized power quite a bit?
8:06
Andreas: Hey, if the incentives create centralization, don't blame the actors
who are following the incentives as you said Stephanie, but look at the
incentives very carefully.
8:18
Are those incentives really aligned with the long-term vision of Bitcoin?
And the answer is simple, it's no.
8:25
Centralization is anathema to Bitcoin. It will be the things that could destroy
Bitcoin. So if the incentives are aligned so they create centralization, which I
think is obvious and very much happening right now, then that's a problem.
8:39
Can it be solved? I think it can be solved with a hard fork. It's not an easy
solution, it's not a soft fork solution, it's a hard fork solution and it looks
kind of something like this: after block X, where X is some block in the future,
the only blocks that will be accepted by the reference client are those that
have proof-of-work based on something else. That could be SCrypt. By the way,
people have probably heard us talking about SCrypt, and just to clarify that's
S-C-r-y-p-t, and it's an algorithm that is extremely difficult to expand into an
ASIC because it requires a trade-off between CPU and memory so if you try
to crunch it hard with an ASIC what you end up doing is using a lot of memory,
and you can't scale memory the way you can scale computation so essentially what
it does is it puts a counter-balancing incentive again centralization.
9:38
So yeah, we could change proof-of-work, it would be a hard fork, it would be a
difficult project that would require some work to get there but if anything is
worth a hard fork that would be it.
9:47
The other alternative--
Stephanie: but why not just use Litecoin?
9:49
Andreas: the other alternative is Litecoin.
The issue here is simply one of balance, right? Do you fork to a different
currency, or do you maintain the network effect and do the fork internally
within Bitcoin?
10:03
It's always a matter of if you think that Bitcoin will succeed in the long term
in the network effect and the fact that many others supported and and use it
is good enough then you go to Bitcoin. If you don't believe that evolution can
happen within Bitcoin, then you go for Litecoin.
10:18
And I think there will always be a balance between that natural kind of
incentive to fork and the incentive to stay on the Bitcoin block chain where the
network effect is greatest, so we'll see how this plays out.
10:29
The good news is, Litecoin already exist so the alternative is already there and
at this point fails what that does is it allows Litecoin to succeed. That's the
nice thing about having an ecosystem.
10:40
Adam: So now Litecoin aside, Peter was also on that panel, Peter from Coinbase
was also on this panel with Dan, and the thing that he brought up, I clipped
this out a little bit so it could be shorter, one of the things that he brought
up is that if you wanted to make a change like that to Bitcoin itself, you'd run
into a problem because the people who you're you're essentially trying to
obsolete, you know, the ASIC miners, probably by the time you get around to
doing this actually have more than fifty one percent of network.
11:05
So you're essentially asking them to be okay with and participe in a switch that
invalidates their rather large investments into.
11:17
So is that even possible, or do we think that is why Litecoin could in fact take
over, because ASICs wouldn't decide that they are going to invalidate themselves
and so the shift has to happen.
Andreas: A hard fork where you change the proof of work would be very difficult
but think of it from this perspective, the centralization has alreday resulted
in essentially CPU and GPU miners abandoning Bitcoin, the mining increasingly
being concentrated on ASICs. Now if you did, say sometime in the future, and
this could be six months out, eight points out, a year even, but as long as you
baked it into the protocol in advance, then you could essentially gradually
align the incentives. Instantly ASICs would be less popular. It wouldn't affect
the current generation of ASICs because they will continue being profitable
until that period of time.
12:07
But it would affect decisions about buying future versions of ASICs, and it would
also make people evaluate the ASICs and the new proof-of-work for future value
that they could bring.
12:18
But at the end of the day if you made that change it would also allow almost a
million people who are in Bitcoin today to resume CPU mining on this new
algorithm, and bring them into the fold.
12:31
So if ASIC miners walked away, ten thousand of them, and instead you had a
million people walk into the fold, I don't see how that's a problem.
12:41
It's only a problem if you accept the centralization is inevitable and that's
really a symptom of the proof-of-work, not the underpinning assumption. So if
you change the proof of work the balance of power changes and now you have a
million new miners who can enter the field. So if ASIC miners lose some money
you know, they lose some money over that speculation but I don't think that it'd
be a problem with there a block chain continuing to be more valuable while the
other blockchain with all of these users on its isn't.
13:09
I think it's quite the opposite essentially people would vote with their mining,
and that's exactly what mining is supposed to do, be a vote, so if the new proof
of work is decentralized, more people can vote.
13:24
Announcement: Let's Talk Bitcoin! is heard each week by thousands of people who
are participating in the new digital economy.
Our listener base of Bitcoin owners, miners, investors, technologists, and
merchants is growing fast. We offer a limited number of short advertising slots
in each show to keep our listeners engaged and to provide maximum impact our
sponsors. If you'd like to talk to us about Let's Talk Bitcoin, send us an
email at [email protected].
13:52
Advertisement: If I showed you a website where you could easily purchase
electronics from the world's largest distributor with Bitcoins at zero percent
markup, would you think it was too good to be true? Good news: it's real, and
it's at BitcoinStore.com. Choose from half a million items, save money over
Amazon and Newegg, and convert your Bitcoins to real-world items. You can even
buy with privacy; all they need is a shipping address. But don't take my word
for it, see for yourself at BitcoinStore.com.
14:26
Adam: You know taking a step back from this, I think that broadly speaking I
can say we all kind of agree that ASICs aren't great overall for the health of
the network because of like you said the centralizing effect, right?
14:37
Andreas: I would call it the disenfranchising effect, eve more strongly.
Stephanie. I'm neutral on them, but I think they do tend to centralize a little
bit.
14:43
Adam: Okay so following that logic, doesn't that also mean that GPUs are
basically also bad because in the same way that fewer people have ASICs
than have GPUs, similarly fewer people have GPUs that are powerful enough to
mine in a reasonable fashion compared to CPUs?
15:01
The question I'm asking here i guess is, does the total amount of power in the
network actually matter, or is just about the distribution of however much power
there is?
Stephanie: I think you're getting at something that's sort of on my mind during
this whole discussion Adam, which is that we need to redistribute the hashing
power on the Bitcoin network. You know, I don't like that because that itself,
to me smacks of some centralization when anybody talks about redistribution.
15:30
Now I'm not sure like, my thoughts aren't completely formed on this but yeah I
mean I guess if we take that logic to it's conclusion, if something is done to
eliminate ASICs or get rid of ASICs then you know why not GPUs, why can't we--
are we going to be saying you know everybody has to be mining, if you're gonna
mine on the Bitcoin network it has to be with this with a CPU and it has to be
this specific CPU so that everybody has equal hashing power, that could get
pretty ridiculous, who gets to make that decision and enforce it, you know?
Andreas: I would certainly not join in such a blockchain, one that prescribes
control over which CPU you use or something like that, and think it's a matter
of orders of magnitude right, if you use a proof of work that has less of a
difference between the orders of magnitude of mining of a CPU, a GPU, and an
ASIC, and SCrypt already does that, it flattens that advantage a tiny bit,
16:25
then essentially with the absolute minimum amount of change you achieve the
original incentives of Bitcoin which is to distribute the voting consensus among
as many participants in the network as possible.
16:36
In my ideal environment for Bitcoin, every single phone you have that's running
a client is also doing a tiny bit of mining, every single device you use use as
your wallet is doing a bit of mining.
16:49
If you could distribute it to that extent then it becomes harder and harder
to stop Bitcoin, and at the end of the day I think it's really a survival
mechanism.
16:59
Centralization in ASICs is not a problem because the ASICs people make too much
money or because we want to change who controls the network, it's a problem
because they can be attacked in regulatory terms, and also the mining can be
shifted to an organization that doesn't have Bitcoin's best interest in mind.
17:17
Someone puts up an ASIC farm, and it's not that hard to do, they could literally
takeover Bitcoin just to shut it down and that's a bad outcome. Now it's less
likely if you have a proof-of-work that is not as easy to scale up as an ASIC.
17:30
Adam: This is very tangential here but I'm really curious, on the flip side,
because CPUs are so ubiquitous, right, so let's assume for second that we make a
shift and we go, you know GPUs no longer work, ASICs no longer work, does that
put more power in the hands of corporations that have large amounts of computers
like workstations that employees use, you know and they can just have this
low-level mining process going in the background, is this a way to generate
revenue for companies that have large amounts of computers and workers on them?
17:59
Andreas: I think you'll find that the number of computers owned by corporations
is a lot less than the number of computers are owned by individuals, so from
that perspective I don't think that too much of a problem but at the same time,
we're not talking about GPUs no longer work or ASICs no longer work,
18:15
we're simply talking about ASICs don't give you a ten-thousand-fold advantage
over CPUs, they only give you a hundred-fold advantage or tenfold advantage over
CPUs. They still would have significant advantages, there'd still be plenty of
incentive for specialized computing devices to do this.
18:34
But it would go back more to the initial environment where you have people
building mining rigs that could be done on an individual basis.
18:42
I think it's just a matter of deciding which incentives give us the best
possible future for Bitcoin, and you know the core idea of the distributed
blockchain proof-of-work system is one CPU per-vote. If you can buy up all the
votes in the network, that's a problem.
19:00
Adam: I can tell you something that's a little bit exciting to me, you know on
this topic on June 23rd, a guy by the name of bytemaster on the bitcointalk
forums released a challenge for people to try and get more oomph out of a GPU
processor or an ASIC processor for this proof-of-work, and you know obviously
it's GPU because you'd have to design a specific ASIC, because they are
application-specific and this is a new algorithm,
19:25
basically the protocol is aimed entirely for being as efficient as possible on
CPUs to the detriment of GPUs and to the detriment of any sort of dedicated
application. The bounty if you can get 25 percent faster performance out of any
GPU compared to what they're getting out of their CPU code is 40 Bitcoins, so if
this is something that you have expertise in and you think that you can make
this happen then you should definitely go take a look. There'll be a link in the
show notes too, and I'm very curious if someone manages to come up with a way
to break through this new algorithm, because it could be pretty promising
otherwise.
Stephanie: If this passes the challenge, creating some kind of new
cryptocurrency with the new algorithm...
20:04
Adam: Yes there's a currency that they've been working on called BitShares,
I have actually spoken with their lead designer a couple of times, guy named
Daniel Larimer, it's gone through several iterations, this was crypto-USD the
first time we talked about, we looked at it on the show before, not on the show
itself but on the back end trying to figure out if it's something we should talk
about, and this is their latest iteration of it and they seem pretty confident
that it's entirely GPU-proof and ASIC resistant while being as fast as possible
to verify.
20:29
Direct quote: "it's good to see people working on this problem because the
centralization thing, especially as we get into a more ASIC-defined future, just
seems like you can see that this is going to be a problem. It might not be a
problem now but it's going to be a problem and so it's good to see people
working toward solutions, towards that, and if anybody else is working toward
solutions in this area please contact because it's an are we're actively
interested in."
20:48
Andreas: So could we just point out the fact that Bitcoin is just amazing,
simply from the perspective that if you don't like it you can go make your own,
and also if you don't like the current decisions going on in the Bitcoin code
you can fork and follow a different block chain.
21:09
Despite the problems we're talking about, the solutions to these problems are
enabled by the very nature of Bitcoin itself, by the fact that it's a consensus
driven mechanism so Bitcoin can solve it's own problems, and can evolve past its
own problems and that's really a key property of Bitcoin that will allow it, I
believe, to survive many such issues, and it has already survived issues like
this in the past.
21:37
Or, it will allow alternative coins to arise out of the ecosystem and take on
more of a leading role, and the incentives are perfectly balanced, the feedback
mechanisms are super fast, so this is as close as you can come to an efficient
market for picking the currency of your choice. Even if Bitcoin went away
tomorrow, you could jump on a new currency, and still exchange your Bitcoin for
this new currency relatively easily.
22:03
Stephanie: Taking this incentive idea I guess, if there is some new
cryptocurrency that's created and is GPU- and ASIC-proof, what if it's only GPU-
and ASIC-proof during this initial testing phase when it's kinda like, the
incentives aren't really there, what if it becomes a multi-billion dollar
currency and then suddenly there're huge incentives to make devices or machines,
maybe it's not GPUs or ASICs, maybe it's something completely different, but to
somehow maximize the profit, what if that happens in the future? I'm interested
to see how that plays out, I'm just thinking economics here, that as a
cryptocurrency becomes more valuable, then the profit incentives become greater
and greater and necessity is the mother of invention right, like everyday said
about Litecoin,
22:46
it's totally ASIC- and GPU-resistant, well now people are talking about making
ASICs that may be able to mine Litecoin on the SCrypt algorithm, even though the
spread wouldn't be as big as with the SHA-256 algorithm like for Bitcoin,
between ASICs and GPUs, or GPUs and CPUs. I'm just really curious to see what
happens with this.
23:07
Andreas: The organization offering 40 Bitcoin as a bounty to overcome their
algorithm is competing against the established economy of Bitcoin that offers
a 1.7 billion dollar bounty to anyone who breaks the encryption,
23:22
25 Bitcoin every ten minutes for anyone who is able to overcome mining
difficulties and mine more efficiently, so yeah I mean the incentives really
favor innovation, very rapid innovation towards achieving the rewards that are
on the block chain.
23:38
The better solution would be to introduce some flexibility into the
proof-of-work concept so that we can gradually change proof-of-work,
through consensus, as the network becomes more centralized. You know, a feedback
mechanism that acts to decentralize the network if mining becomes centralized,
by changing proof-of-work would be a great way to solve this problem, not just
this time but for the future. At the end of the day, too big to fail happens to
any organization whether it's a totalitarian system or a democratic system or
meritocratic system.
24:14
All of these systems gradually morph towards centralization, concentration of
power, and basically they need a reset every few years. I have no illusions that
Bitcoin won't also become more centralized, with the power more concentrated
in fewer hands unless it also gets a reset every few years in order to adjust to
the new reality.
24:36
That should be a goal not something to fear.
24:38
Adam: Do you think that that's something the winds up being automated into the
protocol itself, or is that something where you know the devs at that the
steering wheel at that particular time make the judgment call, say "okay well
it's been four years, it's time to have a big reckoning again"? You
know, how do you think something like that can happen?
24:53
Andreas: Well the devs have shown so far that their inclination is towards
creating market-based feedback mechanisms rather than making decrees by fiat, as
you might say, and making decrees as to how the algorithm should work and i
think that follows the spirit of Satoshi, which is be able to make these
decisions by consensus. It becomes a bit more complicated when the issue you're
trying to resolve is the disenfranchisement of consensus itself, and that's a
bit more tricky, but I think in the long term all we need to do is find self
balancing solutions that work through feedback.
25:28
That would make the currency much more robust, so if you think that innovation
in Bitcoin hasn't stopped but is only beginning, I don't see why we couldn't do
innovation in that particular area that also honors the original incentive
structures of Satoshi and achieves the one-CPU, one-vote promise more
effectively.
25:52
Adam: Thanks to Dr. Stephanie Murphy and Andreas Antonopolous for providing
content for this episode.
25:57
Please send all listener mail to [email protected], especially this time
since we're trying something very new here I'd like as much feedback as
possible, so chime in. Thanks for listening.











20.3: Coinjar.io, and Andreas on Bloomberg
0:12
Hi and welcome back to episode twenty of Let's Talk Bitcoin!, a twice-weekly show
exploring the ideas, people and projects building a new digital economy and the
future of money.
0:21
Twenty episodes. That number came up fast.
0:25
Today's an experiment: one of the most common complaints we get is that our show
is too long and some listeners would rather have the ability to pick and choose
what order they listen to the show in.
0:33
It's also very intimidating for new users who might be interested in a
particular segment but not in the half-hour discussion that came before.
0:40
Episode twenty will be released a number of parts. You should find a total of four
files.
00:45
In 20.1 Stephanie and Andreas share their experiences with localbitcoins.com, an
alternative to centralize exchanges that's not against the rules.
0:57
In 20.2 the hosts have a long conversation about what the point of mining is and
if the incentive might have become the end result, losing sight of Satoshi's
vision. This is a must-listen-to segment, and clocks in at a little under half
an hour.
1:09
In 20.3, that's what you're listening to now, I sit down with Asher Ten and Ryan
Joe, better known as ZhouTong, two of the founders behind coinjar.io, a new
Australian Bitcoin exchange. At the end of this segment you can hear Andreas on
Bloomberg West yesterday.
1:24
In 20.4 Andreas tells us about the upcoming Bitcoin Improvement Project Proposal
number 32 which proposes a new third type of wallet, a hierarchical
deterministic wallet. Sounds technical, and it is but the concept is powerful
and it's worth your time to try to understand the potential here.
1:40
We appreciate you hanging with us as we try these new things out.
1:43
Please send all your feedback to [email protected]
1:47
Enjoy my interview with Asher Tan and Ryan Joe of Coinjar.io from the team at
letstalkbitcoin.com
1:58
Joining me today we've got Asher Tan and Ryan Joe, two of the cofounders behind
Coinjar.io, a recently released merchant platform and exchange existing in the
australian Bitcoin market.
2:10
Asher, how did this come about, what what was a process that took you to where
we are today?
2:14
Asher: Well so Adam, my background is Economics; I used to work as an economist
for close to five years; I was working on startup-projects on the side, and at a
certain point I said I don't really want to have my day job writing about
interest rates any more, let's start a startup. So I found a Joe, my technical
co-founder, we started on a few projects, we were accepted into an incubator
program. The incubate told us we would rather much you have a Bitcoin idea than
your social eating idea, which was great because we were really bullish on
Bitcoin and I knew exactly who to call to join our Bitcoin project, which was
Ryan.
2:48
Adam: So Ryan, you actually have some background in this space, this is not the
first exchange that you've been part.
Ryan: Actually, I founded Bitcoinica two years ago, it was a margin trading
platform that allows people to trade with more deposit than they have, also it
allows trader to short sell, unfortunately after I sold Bitcoinica to a US
investor, after the new management went onboard, the exchange was repeatedly
hacked by security breaches and that resulted in huge monetary losses and
currency deliquidators working on the issues right now.
3:24
Adam: So I know that there's still a little bit of aggression in the communities
and some bad memories towards that for people who lost money, that's something
that my understanding is that you have no control over at this point, is that
right?
3:37
Ryan: Yeah that's right, but it's really unfortunate that this whole thing
happened, but well I'm starting over in Coinjar and I want to still continue to
contribute into the Bitcoin community with my experiences.
3:54
Adam: So tell me about Coinjar, what's the plan for the platform, what is
it that you're actually trying to do with this?
4:00
Ryan: Currently Coinjar is a full service Bitcoin platform, and we provide
wallets, and Bitcoin buying service as well as service for merchants, so Coinjar
is the wallets, which is highly integrated with our other parts of our platform,
and Coinjar Filler is actually the first service in Australia that allows you to
buy Bitcoin with not ridiculous fees, so we charge only two percent for buying
transactions and many Australians really love this service, and Coinjar Checkout
is our merchant solution, which is similar to BitPay, and our merchants can use
Coinjar Checkout to integrate with their website, and get settled in local
currency every day.
4:40
Asher: Yeah when we started this there were a lot of opportunistic players in
the market, and we didn't feel it was right even though we could probably come
in and charge much higher fees, there was no real reason to charge Bitcoin
enthusiasts eight percent to buy Bitcoin and we just thought that was wrong, so
4:58
I think our first move was to make sure that we have a really good service for
Australians to hold Bitcoin and subsequently working on merchant services as
well.
5:07
Adam: So what's the primary focus, I mean are you trying to be the big player in
Australia here or is the focus more on using that as a launchpad, mingling
international, what's the plan?
5:16
Asher: Yeah I think we do have international ambitions; right now the southern
hemisphere, Australia really lacks a strong player so I think if we can
prove ourselves, that we can do a really good job for the Australian market,
we won't have any problems branching overseas.
5:30
Adam: Usually when we talk to people who run exchanges, they're not in
incubators. At least in the United States and in some parts of Europe, the idea
of running exchange is something that requires you know like a legal staff and
basically you have to be really proactive in complying with a whole lot of
regulations that deal with money laundering and online crime and things like
that, is the situation different in Australia because I mean it sounds like you
guys are kind of just getting started in and you're looking to take over.
5:56
Asher: We are still in the embryionic stage of figuring out the direction of our
company, and Australia as well is quite unsure of what to do with Bitcoin,
I mean the tax office has come out to say that Bitcoin is no more anonymous than
physical cash, and it doesn't raise major issues that don't currently
exist. Bitcoin is still a digital commodity in Australia and not yet classified
as a digital currency, or a currency per se, so we're still exempt from many of
the regulatory rulings that exist to in America and other countries.
6:26
Adam: So Ryan, the thing that was interesting about the Bitcoinica, like you
said it applied leverage in a way or it allowed traders to apply leverage in a
way that wasn't really possible with any other platform and to my knowledge
still is impossible with any other platform. Why is it that you were able to do
that when it's been so difficult for everybody else?
6:42
Ryan: well Bitcoinica actually adopted a really unique model that is the market
maker model, because you for every new exchange there is always a
chicken-and-egg problem: buyers will only go there if they have a lot of sellers
and sellers and sellers will only go there if there are a lot of buyers, this is
a problem with every new exchange and Bitcoinica kind of stopped this problem by
providing guaranteed liquidity by taken the spread out of the current market
price at Mt Gox and then providing liquidity to buyers and sellers so that they
can start trading. Also the margin trading platform was inspired by a lot of
wthe foreign exchange trading platforms that I have experience in, actually it's
really similar to other forex brokers.
7:25
Adam: I guess the thing that's confusing to me is you had a lot of success with
that, and there've been so many markets that have come out and so many exchanges
that have come out since Bitcoinica went away that haven't really been able to
get much traction, surely they might have been able to scrape a percentage or
two off of Gox but it's not really anything substantial and this is something
that over and over again, people are saying you know this is something that
would make for a more mature market, this is something that the market needs or
at least a lot of players think that the market needs.
7:50
Do you think that there's going to be a player that comes back and does
something like that, is that something that Coinjar.io is looking to get into,
or are you guys taking a more traditional exchange and merchant services
approach?
7:59
Ryan: Currently we don't really want to expand our services into the margin
trading territory, so for now we just focus on simple buying and selling; buying
for consumers and selling for merchants, and as a service platform we just want
to facilitate Bitcoin transactions rather than speculation.
8:16
Asher: Early on, Ryan told me the success of Bitcoinica had to do with two main
things: one was design and usability, so a lot of that's lacking at a lot of
other Bitcoin platforms at the moment, and the other one was service because
Ryan was always on the bitcointalk forums trying to help users solve user
issues. I think those two things help Bitcoinica succeed.
8:38
Adam: So with Coinjar.io, given what you see as the potential for the Australian
market here. Do you think that this is something that could really enter wide
adoption; have you participated in any meetup groups, or do you know of, I mean
is there a scene for startups there that are focused on the Bitcoin space?
8:54
Asher: Not so much for startups, right now in Australia pretty much every major
city has a Bitcoin meetup, I host Melbourne's Bitcoin meetup and it's
co-sponsored by Coinjar and an international Bitcoin company which, we're still
finalizing the details but we're expecting close to fifty people at our next
meetup so, lots of interest in Bitcoin. About a month ago we also traveled down
to Singapore for a conference and a hackathon, the main reason we attended this
hackathon was we found out that the development bank of Singapore, DBS Bank
would be there. So we went to really pitch to the innovation team and gauge
interest in the Asia-Pacific, Singapore market to see if Bitcoin was feasible
choice for them. So we went in the with a point-of-sale terminal that we made
overnight, showed it to them and they were quite keen but at the same time I
don't think they took us quite seriously in terms of Bitcoin and P2P payments.
They did agree that there's a lot of people unserviced by current banking
markets but at the same time I think they're just a bit complacent and don't
really see the disruptive force of Bitcoin.
10:05
Adam: Do you think it's that they don't see the destructive force or do you
think that it's that there's not really much in it for them, you know I mean
because the people who aren't serviced by the banking system aren't serviced by
the banking system because they're not profitable, and so if it's not profitable
than there's no reason to do it. Do you think that they just don't get it, they
look at this and they don't see the potential or that it's something else?
10:24
Asher: Well I think one is banks have just become complacent, especially in a
place like Singapore where a banking monopoly is easy because of a small
population.
10:33
Ryan: The bank is not too much interested in P2P payments mostly because it may
not be profitable for them in the future; most banks are quite happy being a
monopoly in financial services and make a lot of money from the money-creation
process, but Bitcoin actually prevents that from happening, Bitcoin belongs to
the people and there's no central banks so everything becomes very competitive
because the barriers of entry becomes so low in the Bitcoin economy, the banks
may not have a big role in this kind of things if it becomes the future.
11:07
Asher: Yeah, the banks may not really want to get the concept because that's
exactly what they are not trying to do.
11:16
Adam: For our australian listeners, if they'd like to get in touch with you, do
business with you, what have you, it's Coinjar.io
11:21
Ryan: Yeah, that's correct.
11:24
Ryan, Asher, they very much for being on the show.
11:30
Let's Talk Bitcoin! is heard each week by thousands of people who are
participating in the new digital economy.
11:36
Our listener base of Bitcoin owners, miners, investors, technologists, and
merchants is growing fast. We offer a limited number of short advertising slots
in each show to keep our listeners engaged and to provide maximum impact our
sponsors. If you'd like to talk to us about Let's Talk Bitcoin, send us an
email at [email protected].
11:57
Advertisement: If I showed you a website where you could easily purchase
electronics from the world's largest distributor with Bitcoins at zero percent
markup, would you think it was too good to be true? Good news: it's real, and
it's at BitcoinStore.com. Choose from half a million items, save money over
Amazon and Newegg, and convert your Bitcoins to real-world items. You can even
buy with privacy; all they need is a shipping address. But don't take my word
for it, see for yourself at BitcoinStore.com.
12:31
Emily: Cameron and Tyler Winklevoss, the twins best known for their role
in the history of Facebook, are looking to create the first publically traded
fund for Bitcoin. An entity called the Winklevoss Bitcoin Trust has filed for an
Initial Public Offering of one million shares to raise up to twenty million
dollars in capital. For more I want to bring on my partner and our
editor-at-large, Cory Johnson. Cory?
12:54
Cory: Yeah Emily, along with the digital currency plan this is being met with
lots of skepticism for just the ETF alone. I'm joined right now by Andreas
Antonopoulos, who's the cohost of Let's Talk Bitcoin!, the show about the new
economy, the future of money. Andreas, this notion of an ETF that'll trade an
amalgammated value of twenty million dollars worth of Bitcoins, how is that seen
in the Bitcoin community?
13:20
Andreas: I think it's seen as good news. It's going to bring some volatility,
it's going to bring some mainstream views and perspectives and some mainstream
investors.
13:27
Cory: And twenty million bucks, at least.
13:29
Andreas: Well yes, I mean compared to the 1.5 billion dollar total market size
it's small but at the same time for daily volume that's great.
13:37
Cory: Are there certain, the way it's going to work if this works at all is
they're going to try to raise twenty million bucks and then go out and acquire I
guess Bitcoins themselves and so they'll kind of track it like a currency
almost, but I guess the difference is among other things with this you'll now be
able to short, for all of those Bitcoin haters out there who think this
technology and this currency is a joke, they can actually go short the ETF.
14:03
Andreas: Absolutely. It's an index fund so they say they already have one
percent of the Bitcoins that are in the markets, the Winklevoss twins, so
therefore presumably they already have the basis to capitalize this from the
Bitcoin perspective, and as an index fund it's going to be great because they
can support the bulls and the bears.
Cory: It wasn't clear from the filing and maybe you know more than I, but it
wasn't clear from the filing that they were going to take the money raised and
sell their own Bitcoins to the ETF; could the actually go in the marketplace and
buy them, would there be a price difference at all?
14:32
Andreas: At the moment it's probably trading a bit higher than when they
originally bought their own Bitcoin, so they would probably make a loss if they
did it that way but it's not clear from the filing.
14:43
Cory: What are the tax advantages to owning Bitcoin as opposed to owning
currency, and would that change with an ETF? If you're holding an ETF you will
be fully taxed on things and capital appreciation so on. Are you in Bitcoin as
well?
14:58
Andreas: Yes, absolutely. I'm in Bitcoin, and I'm taxed on Bitcoin just as I am
on sterling, or Euro, or any other foreign currency. You know, if I keep it
long-term, long-term capital gains--
15:06
Cory: So there's no real tax advantage or difference with this thing.
15:10
Andreas: No no, it's treated just like a currency, the IRS considers it
essentially capital gains if you trade it, income if you're actually creating
Bitcoins through mining but other than that capital gains.
Cory: Now within the S1 filing for this new thing, one of many risks factors
that I thought really got to the crux of it which says essentially that until
Bitcoins is accepted by retailers or is used in commerce more it is purely
speculative tool, therefore sufferers excessitudes of trading and the volatility
of that as well,
15:41
within the Bitcoin community is there a belief that there needs to be a push to
actually use these in commerce more?
Andreas: Absolutely. I think in fact it is using commerce quite a lot, you know
there are already thousands of retailers who accept it and for the average
retailer it presents some big opportunities--
Cory: A thousand retailers is not a lot
Andreas: It's small, it's tiny, yes
Cory: I can't think of a single place I've been that accepts Bitcoin as payment.
16:03
Andreas: There's a few here in San Francisco, they gain the advantage of
essentially non-reversible transactions and very low cost for the payments,
so you know from that perspective it has a lot of advantages over traditional
payment systems and merchants love it.
16:15
Cory: Interesting stuff, alright Andreas Antonopoulos thank you very much to the
cohost of Let's Talk Bitcoin! I appreciate you talking Bitcoin with us.
16:21
Andreas: Thank you so much.
16:32
Adam: Thanks to Asher Tan and Ryan Joe for providing content for this segment.
Please send all listener mail to [email protected] - especially this time
since we're trying something very new here, I'd like as much feedback as
possible so chime in. Thanks for listening.











20.4: Understanding Hierarchical Deterministic Wallets
0:12
Hi and welcome back to episode twenty of Let's Talk Bitcoin!, a twice-weekly show
exploring the ideas, people and projects building a new digital economy and the
future of money.
0:21
Twenty episodes. That number came up fast.
0:25
Today's an experiment: one of the most common complaints we get is that our show
is too long and some listeners would rather have the ability to pick and choose
what order they listen to the show in.
0:33
It's also very intimidating for new users who might be interested in a
particular segment but not in the half-hour discussion that came before.
0:40
Episode twenty will be released a number of parts. You should find a total of four
files.
0:45
In 20.1, Stephanie and Andreas share their experiences with localbitcoins.com,
an alternative to centralize exchanges that's not against the rules.
0:57
In 20.2 the hosts have a long conversation about what the point of mining is and
if the incentive might have become the end result, losing sight of Satoshi's
vision. This is a must-listen-to segment, and clocks in at a little under half
an hour.
1:09
In 20.3, I sit down with Asher Ten and Ryan Joe, better known as ZhouTong, two
of the founders behind coinjar.io, a new Australian Bitcoin exchange. At the end
of this segment you can hear Andreas on Bloomberg West yesterday.
1:24
In 20.4, that's what you're listening to now, Andreas tells us about the
upcoming Bitcoin Improvement Project Proposal number 32 which proposes a new
third type of wallet, a hierarchical deterministic wallet. Sounds technical, and
it is but the concept is powerful and it's worth your time to try to understand
the potential here.
1:40
We appreciate you hanging with us as we try these new things out.
1:41
Please send all your feedback to [email protected].
Enjoy Hierarchical Deterministic Wallets: The Seed that Becomes the Mighty Oak,
from the team at LetsTalkBitcoin.com
1:55
Andreas: I'm very interested in a new standard, called hierarchical
deterministic wallets, defined in Bitcoin Improvement Proposal 32.
2:03
So BIP-32 is a standard for creating wallets in a tree structure that can be
derived from a root key. This is based on original conversation and suggestion
by Greg Maxwell who's one of the core developers, who is really a genius with
cryptography and has these incredibly innovative ideas around key management.
Pieter Wuille developed that into a full proposal and it's also been implemented
by Tamas Blummer in the Bits of Proof implementation of the Bitcoin enterprise
server as well as many other places. I'm currently writing an implementation
based around that.
2:40
I'd like to talk a bit about the topic. So what are hierarchical deterministic
wallets? Type-1 wallets, as we know them today, are wallets where the private
key is derived from a random number. They allow you to generate a completely new
private-public key pair whenever you want. The problem with that is that you
have to backup each wallet individually, because there's no way to recreate that
private key, it's based on a random number.
3:06
Type-2 wallets are the type we see in the original implementation in Electrum,
where you have a seed, and that seed generates not just the first wallet but
wallets after that in a sequence, so essentially all the wallets that are
generated derive from that original seed, and you can regenerate them and that
works very well as a backup process.
3:28
So taking that one step further, BIP-32 allows you to have a Type-2 wallet that
actually has structure in it. So you still have a master seed, but then you can
develop sub-wallets to that and for each sub-wallet you can have sub-accounts,
and sub-wallets and sub-wallets, but basically it's a tree structure.
3:49
You have at the root a master key, and then you can derive all the children
underneath that and from each of the children you can derive more children. The
nice thing about this is that you can share any part of the tree, so for example
I could give one of the sub-trees, I could give the private keys or public keys
to someone else and they could then generate branches further down that tree for
example in order to do one-key per transaction, or if they could use it to
monitor some of those keys, for example to do accounting or transaction
counting.
4:23
If you have an organization it would allow you to do departmental spending, etc.
So it's a very interesting protocol. What do you guys think?
4:29
Stephanie: I was gonna say, would that mean they're adopting your grandchildren
or something?
4:33
Andreas: Yes, absolutely, so it really does have that kind of structur. You
can think of the master key as the ancestor and then you can create
descendants, and then you can essentially assign different families of keys for
different uses.
4:49
Stephanie: Yeah, I'm glad you mentioned the uses for like business accounting, I
mean that sounds super useful to me and it's great to have the more concrete
idea of like what this could be used for but I imagine there's lots of other
uses besides just in business, it could even be used to manage like a household
budget or something like that.
5:07
Andreas: This is not just for business and you don't need a hierarchical
structure. Let me give you one specific example:
5:12
Let's say you have a web server that's doing content monetization, and you want
to be able to generate lots and lots and lots of keys, essentially once per
transaction or one per page.
5:24
Just like you can do that with a master public in Electrum, in a hierarchical
deterministic wallet what you can do is take a branch, share the public key
that's the head of that branch, and then from that derive millions, billions,
2.1 billion keys in fact, that are all derived from that master public key, but
then you can also unlock any of those wallets with a private key that never
touches that server, so that way you can have essentially a completely untrusted
server generating public keys without any knowledge of the underlying private
key that can unlock all those transactions, that's a great use case.
6:05
Announcement: Let's Talk Bitcoin! is heard each week by thousands of people who
are participating in the new digital economy.
Our listener base of Bitcoin owners, miners, investors, technologists, and
merchants is growing fast. We offer a limited number of short advertising slots
in each show to keep our listeners engaged and to provide maximum impact our
sponsors. If you'd like to talk to us about Let's Talk Bitcoin, send us an
email at [email protected].
let's talk between his herd each week by thousands of people who are
6:33
Advertisement: If I showed you a website where you could easily purchase
electronics from the world's largest distributor with Bitcoins at zero percent
markup, would you think it was too good to be true? Good news: it's real, and
it's at BitcoinStore.com. Choose from half a million items, save money over
Amazon and Newegg, and convert your Bitcoins to real-world items. You can even
buy with privacy; all they need is a shipping address. But don't take my word
for it, see for yourself at BitcoinStore.com.
7:07
Andreas: Another use case is, you give the public key to someone who's doing
audits, and they can see all of the transactions of all of the subtree, but they
can't unlock any of them.
7:17
So it gives you this enormous flexibility to share parts of a key structure, and
share either the public or the private side, and then derive an infinite number
of sub-levels, or near infinite number of sub-levels, very very quickly.
7:33
It has a lot of uses; I'm planning on using it for example for doing paper
wallets that are all derived off one master key, so that you can not only store
them individually but you could also recover any of the paper wallets
deterministically.
7:49
Stephanie: So how do we get one? I want one.
7:52
Andreas: So it's still a standard in development, it's draft, but here's what's
interesting: in the last development announcement that Gavin did, which was the
development announcement number 4 where he talks about version 0.83, one of the
things he said is that his plans include by 0.9 the implementation of
hierarchical deterministic wallets. I'm not sure if he's endorsing BIP-32 at
this point, but certainly there's enough momentum behind it, there's already
three or four implementations that work very very well, and so I think this is
going to be moving forward quite fast.
8:25
Adam: This might be totally wrong, when you talk about the seed that you used to
generate this hierarchy of wallets basically, is their any sort of connection
between that seed and the brain wallet concept?
Andreas: So a brain wallet is when you generate a single key off a seed that's
human-memorable, so for example a passphrase.
8:45
Theoretically you could use a brain passphrase as the root for hierarchical
deterministic wallets. I would consider that a potential security risk, because
if you are able to unlock that master key, you unlock the entire trees so that's
a very bad thing, and bring wallets have the distinct disadvantage that include
human patterns, and if it's memorable it's brute-forceable.
9:10
So I generally consider brain wallets to be extremely insecure other than for
very sophisticated users who can generate high entropy passphrases and not
forget them.
9:22
For the purpose of hierarchical deterministic wallets, usually you start with a
random number generator and then you apply some kind of passphrase using a hard
to brute-force algorithm like SCrypt, that's the implementation I've seen so
far, but generally you start with a random number. What you would do is you
would only need to keep a very strong backup of that master key, and the
everything else from that level below could be derived algorithmically.
9:49
It makes backup a lot easier, but it also creates more risk on that single key,
so you have to balance it very carefully.
9:57
Adam: Once this makes it past the standards process and the decision making,
what type of software is needed to enable this is, is this something that will
need to be built into wallets, or will there need to be a specific application
that is just about hierarchical deterministic addresses or what's the, what do
you think the process is going to be?
Andreas: It's probably going to be built into wallets and other applications, so
for example if it's built into bitcoind and bitcoin-qt, then you would be able
to get something very similar to what you can do with Electrum today, which is
have some kind of recovery seed that you can store, and then you can recover all
of your keys in the case of loss.
10:32
So you could lose your entire wallet, you wouldn't need to necessarily encrypt
your wallet on the device because you could you could store the master key
offline and in your wallet only have public keys.
10:44
So as a result you can derive payment keys, but you can't actually send any
transactions, so there's all kinds of nice you can do.
10:50
I think you'd see it first in wallets, then you'd see it as part of online
infrastructure, and then finally you'd see it in more complex services.
11:00
Thanks to Andreas Antonopolous and Dr. Stephanie Murphy for providing content
for this segment. Please send all listener feedback to [email protected],
especially this time since we're trying something very new here I'd like as much
feedback as possible, so chime in. Thanks for listening.