Thanks everybody for showing up, interested to give you a little overview of one of the
many projects that BitGo has been working on over the past year.
And Prova is basically our first foray into a custom permissioned blockchain.
This is something that came about after several years of talking with enterprises who wanted
to get into this space but did not necessarily want your fully fledged permissionless open
blockchain that is very difficult to control.
So after doing a lot of research and evaluating different technologies, we dove in and after
about a year of work both on the protocol side and our own infrastructure side, we've
got an interesting new system that is going into production use right now, it should actually
be starting to get used by real customers in the next couple of months.
So what is Bitcoin?
That's kind of where we have to start with a lot of these things, hopefully don't need
to explain to most of the people here because we certainly do not have time for all of that.
Of course we're here because back in 2009 this new technology appeared on an obscure
mailing list for cryptographers and cypherpunks and ever since then we've all been kind of
stumbling forward trying to figure out what it is that we've actually got here and what
can we actually do with it.
So we of course began this journey because the current financial infrastructure has a
lot of deficiencies, a lot of inefficiencies, a lot of vulnerabilities and as I'm sure all
of us are aware there are a number of different problems that can result in, shall we say,
unfair practices, unfair advantages by different players in the system who can take advantage
of the way that the infrastructure is currently set up.
So just due to the nature of centralization and concentration of power within these various
hierarchical systems we find that people manage to get themselves in positions of power, positions
of trust where they can then take advantage of really the infrastructure of the system
itself in order to make changes or convince people to do things that are actually against
their own best interests.
So as we are aware of the way that traditional ledgers and databases work generally you're
going to have one master copy of your database that is the source of truth and that is generally
quite efficient from a technical standpoint because you don't have to worry about replication
errors or reconciling multiple different sources of data but the downside is what you're doing
is you're putting all of your data under the trust of one entity and usually this is in
a very obscure format where it's not open, only a few trusted people within an organization
might have the ability to actually do an audit on the database or the ledger and this can
result in issues that grow out as the network around this ledger grows where you end up
creating systemic problems due to the trust and due to the lack of visibility.
So as a result this can create a very powerful single point of failure which can result in
the entire system potentially collapsing or becoming unstable if tiny little things go
wrong at the main point of your source of truth.
So that of course can be anything like hacks or fractional reserve banking that results
in bank runs or of course really like what we saw in 2008 is just sort of this extreme
like layering of complexity upon complexity to the point that basically nobody actually
understands what's going on in the system anymore and eventually enough people figure
out what is going on and the whole thing kind of falls over and a lot of trust gets lost.
So then comes along this completely different concept of distributed ledgers where we're
really inverting the entire system where instead of having a single trusted source of truth
we then say okay everybody checks everybody else's work right so we are basically copying
replicating data all around a network of entities that is participating and they are all checking
each other's work and if anybody finds anything that has gone wrong they reject the changes
and as a result we have this new sort of automated consensus system but of course the tricky
part is you have to build a protocol that can basically automate whatever it is that
you want to come to a consensus about amongst your entities to then have the machines basically
do the auditing for you automatically and throw a red alert through a lot of alarms
if the consensus logic itself finds any problems with the activity that is happening on the
network.
This results in a much more robust system because we no longer have single points of
failure if something happens you know at one particular node that's going to be a localized
failure rather than a systemic failure that can cascade across the entire ecosystem.
It also just creates a much more open type of model where you could argue that it's
easier for people to build on top of the system because they have a better idea of what's
actually going on on the layers that they're building on top of as an infrastructure.
So for your standard like marketing checklist here of why would you want blockchains versus
traditional databases of course we've got transparency we've got what you often hear
called immutability but I would say a better description is just tamper evidence.
Tamper evidence is really what comes along with any type of blockchain record because
you're really you're creating this cryptographically chained history or audit log of everything
that has ever happened within the system and while of course it is possible for like a
hacker to go in and change data they cannot change the data in a way that is not easily
evident because they're going to be breaking one of those links in the cryptograph chain
somewhere so at the very least you have this tamper evidence where you can easily tell
your computer you know go back and audit all of the data and if anything goes wrong we
know that we need to get some human auditors in and check what happened right here you
know maybe it was a software failure maybe it was actual you know malfeasance by some
actor within the system.
Immutability I would say is tamper evidence plus some sort of extreme cost to tampering
with the system so within Bitcoin of course that is the proof of work within other systems
you might have some sort of proof of stake or federated consensus.
The actual consensus algorithm for creating that high cost of tampering is less important
if you're talking about a permissioned system where most of the actors who are operating
the network somewhat trust each other and we'll talk a little bit about that later with
Prova and then of course with regards to the actual the trust nature of it you're no longer
relying upon one entity to be telling you the truth about what the state of the system
is because you validated everything yourself and with regard to costs I would say it's
less efficient from a technical like database standpoint of you know actual queries and
updates but what you really find is like you're potentially lowering a lot of costs if you
can get the sort of human auditors out of the equation as much as possible.
So really what we're trying to do is automate away a lot of the compliance costs within
a particular system so that you know you can still have some human auditors that check
on it every once in a while but the vast majority of the auditing gets automated so that it's
happening 24 7 365 without necessarily needing oversight of a human and of course we believe
that these systems are much more robust for a number of reasons one of which of course
is just the fact that we're now using public private key cryptography so that instead of
having like all of your secrets or all of your private data once again in one database
that needs to be highly guarded instead what we're doing is we're spreading out the actual
keys to the system amongst all of the users so you can think of it instead of having like
a single highly vulnerable honey pot of valuable data that is going to get attacked we instead
have sort of edge security so that instead we have hundreds if not thousands or tens
of thousands of users that are actually controlling their own private keys that allow them to
access the system so then once again any one of those people of course can be hacked and
might have their their assets stolen from them but that is a localized problem rather
than a systemic problem it results in a much less fragile system you could even say anti
fragile system in some cases.
So if you want to use a how do you actually figure out like what kind of blockchain or
what kind of distributed ledger do you want do you want a permissionless ledger do you
want permission does it need to be public private what's the simplest way to build it
what's the most robust technology available can the technology scale to meet whatever
your throughput requirements or latency requirements or other technical needs are there's a lot
of questions that need to be asked before you just go off and start building on some
sort of blockchain or distributed ledger technology.
So what is Prova well Prova is open source blockchain protocol we specifically designed
it for digitizing physical assets the the goals that we're going for here is transparency
in ownership and transfer of ownership reduced cost of auditing the database and just general
global access to this network now unlike Bitcoin it it is permissioned it will have you know
governance and administrators but it will also have some additional safety rails that
you don't see in Bitcoin itself and a lot of this stuff that we built was directly as
a result of speaking to traditional financial institutions talking about what their needs
are from regulatory requirements and from you know investor requirements.
So in terms of you know what type of blockchain is Prova on this spectrum it is a public permissioned
blockchain and public permissionless blockchains are like Bitcoin, Ethereum, Litecoin, whatever
private blockchains you can probably say I guess hyper ledger is is much more private
it is definitely also permissioned blockchain I believe yeah so yeah you can do either with
hyper ledger right yeah so that has even more flexibility and from a technical specification
Prova is a fork of BTCD which is one of the older Bitcoin implementations specifically
written in Go and it is using the UTXO style accounting model as opposed to accounting
style which is really more what like Ethereum and Ripple use.
One of the differences though is that because at BitGo we are a multi-signature wallet provider
and we require all of our blockchains to support two out of three multi-sig for our security
model we actually set up Prova to be a multi-sig only blockchain so all addresses all wallets
all transactions within Prova are two out of three multi-sig and this actually gives
us some interesting properties one of those is this concept of the safe address and what
that basically means is that in Bitcoin or in any public permissionless blockchain all
you need to do to create an address is generate a private key with a little bit of entropy
and then you've got your your public key and you can drive an address from that then you
can send money to it and then you can throw that private key away or lose it or some act
of nature causes you to not be able to recover it once that private key is gone nobody in
the entire world has the ability to recover that money or those digital assets for you
so we wanted to prevent that from ever being able to happen within our own system and the
way that we do that is that whenever you are generating an address in the system it has
to be two out of three key set up and one of those keys has to be provisioned by an
administrator to a wallet service provider and one of the the third keys has to be provisioned
to a key recovery service so basically long story short the user can generate any key
that they want and screw up and lose it but there are always two other entities within
the system who will have the other two of those three keys for that address and will
thus be able to recover you know given if the user goes through their recovery process
with their AML KYC process and will thus be able to recover any any funds that they manage
to lose from their own ignorance or bad luck or what have you now we do use proof of work
for our consensus algorithm but it is not quite the same as Bitcoin I believe we're
using the Keckac algorithm which is related to Ethereum I believe but yes so we you still
need miners but for example you could basically have just a single CPU on on each miner that's
really what we're expecting to happen with the Royal Mint Gold when it rolls out there
will be you know maybe like a dozen miners on the network but these miners aren't going
to be using ASICs they're probably not even going to be using GPUs they're just going
to be using GPUs so very low electrical costs and the reason why that is okay is because
we added some additional logic which is the throttling that we're talking about which
is basically that amongst all of the other stuff within the system that is permissioned
the miners are permissioned so in order to be a miner on this system you actually have
to get a special key provision to you from the administrator of the system so that's
going to keep a lot of bad actors out and then on top of that it's not going to be possible
for anyone with a single mining key to 51% attack the system because we have some throttling
built in that basically says like if three out of the past ten blocks were created by
you know this provisioned miner key and they try to create another block reject it so question.
Yeah why did we use proof of work versus like proof of stake or anything else or proof of
elapsed time or whatever yeah basically simplicity it doesn't really really matter because it's
all permission in the first place we could have just used some sort of like federated
round robin signing but I mean it's all essentially going to end up with the same security model
because you will have some super administrator sitting on top of these miners and if they
start causing havoc then the administrator is just going to whip out their key and deprovision
that miners ability to create blocks at all so yeah we could we could swap out any type
of consensus algorithm and probably will long term one of the other reasons why we did keep
it proof of work though is that that potentially gives the flexibility for this to turn into
a more of an open permissionless mining system basically all that would be required to turn
it into that is that the administrator releases the keys into the public for mining and says
okay have at it you know we're going to try to make this a fully fledged proof of work
system where anybody who wants to can join in it's flexibility but because this whole
system is permissioned and I think you're going to find that like the Royal Mint and
CME and and companies that end up using it they're going to be tweaking stuff as they
go forward experimenting basically well yes so so that is another thing at least within
the Royal Mint gold there is no subsidy transaction fees so that incentive but we expect that
at least in the beginning as we're bootstrapping the Royal Mint gold network it's only going
to be you know the companies that partner together in the first place they're already
highly incentivized to keep the system up and running because they've spent who knows
how many millions of dollars getting it built in the first place then long term it may eventually
transition over to something where you know we're at they're actually paying for hash
power if they feel they they want to change that.
So yeah so there's a lot of permissioning that we'll also go into we've got multiple
layers of stuff going on yeah yeah so you account modeling there are I'm not aware of
like any legal issues or reasons why you might prefer one over the other I mean the reason
why we preferred UTXO over account model is that it provides a stronger history of the
flow of that money like you you have you have not only a blockchain of blocks with you know
the stamp transactions you have a chain a historical chain of inputs and outputs so
input output input output it just it seems to be more robust and you know the as a result
of building of going through that whole blockchain and and processing all the transactions you
then you end up with you know the UTXO set which is like your your current snap of your
state of the system.
Accounts can also work well too but now we ended up going with the UTXO model mainly
because we felt like Bitcoin had a better longer more proven track record and I'll also
talk about like why we didn't do any smart contract stuff for any of that in a bit but
this this system gives you a number of features mainly you know related around the the asset
issuance and the enabling and disabling of various actors within the system so we have
permissioning basically you have your administrators who own the system for from multiple perspectives
and then they give permission to wallets they give permission to validators or miners and
then the the wallets are generally going to be like the AML KYC type entities that then
give permission to the individual users who get into the system and start using it as
a you know peer to peer digital asset.
So to kind of take a look at how that works we've got here you can see the the wallet
service provider they're always going to have one of those three keys the user is always
going to have one of their three keys and then there's going to be a key recovery service
that is a completely third party that will keep recovery key you know just in case somebody
manages to lose one of their keys and we need to go and reconstitute enough data to be able
to recover somebody's assets and so if a user comes in and they pass through the KYC
AML compliance stuff with the wallet service provider they get into the system now they
can basically act the same as any user acts once they get into Bitcoin or Ethereum or
Litecoin or whatever and they will use their key to half sign a transaction they will then
send that half sign transaction to their wallet and the wallet will decide based upon any
number of security parameters whether or not to co-sign the transaction if they do co-sign
the transaction then it becomes valid from a protocol standpoint gets broadcast out on
the network the miners confirm it it gets into the blockchain and now you have your
tamper evidence audit log that will remain as such for all time from a kind of higher
level similar type of stuff happens with the administrator where you have you'll have multiple
administrators with multiple keys and we have some sort of offline signing software that
we've developed to basically think of it as like when your guys who are manning the minute
man missiles for the army and they have to put in their keys at the same time and you
know turn them synchronously in order to initiate a rocket launcher what have you similar type
of stuff with making administrative level changes to a prova based system where we have
to get you know X out of Y administrators to come together and assign a administrator
transaction which can do any number of things such as create or destroy assets in the system
provision or deprovision the wallet service providers or provision or deprovision the
validators or miners on the system and once enough administrators come together and create
that transaction just like any other transaction it gets broadcast out on the network it's
confirmed in a block and at that point some aspect of the system gets changed.
So we are like I said we're focused on narrowing here we're not trying to be everything for
everybody we're trying to be highly secure and highly specific in what the system is
trying to do I think I've already really talked about all the stuff on here and come again
right so this is its own blockchain right so the UTXO is the asset like there's nothing
complicated about you know using like op returns to color assets or really anything else it's
one asset for the entire blockchain so this is yeah it's not like we're I mean if you
wanted to have multiple digital assets you would need to have multiple different networks
like we're not trying to support you know multiple assets on the same network let's
see so yeah so in Prova the UTXO is the asset I'll talk a little bit about like how that
works with Royal Mint and Gold in just a second Prova of course is open source I think there's
a URL on here somewhere Provachain.com I believe but it's on our github just github.com slash
bitgo you'll find Prova there and of course all the instructions you need to get it up
and running it's pretty easy to get a Go project compiled and running and with the one of the
reasons that we chose Go as opposed to like the Bitcoin Core implementation which is C++
is that we just felt like Go was the most developer friendly because we wanted this
to be an implementation that the vast majority of developers would be able to get into and
start you know playing around with and I found I haven't written much Go myself but found
that it wasn't particularly challenging I'm still terrible at C++ so I don't contribute
to Bitcoin Core very much who uses Prova at the moment this was really like for the Royal
Mint and Chicago Mercantile Exchange and the idea there being that Royal Mint is this thousand
year old institution that has been vaulting gold bullion and basically they are probably
one of the most trustworthy entities in the entire world at this point and what they wanted
to do was just kind of upgrade their own infrastructure and be able to enable new types of functionality
for their own users so the ability to have like a peer-to-peer and auditable digital
asset was appealing to them they're able to offer a number of things that most like physical
gold vaulting institutions have not been able to offer up to this point let's see a little
marketing material yeah so so Royal Mint and gold is the first like production use of this
yeah we've been I mean we've been in testing phases for at least like six months now and
at this point like the only thing stopping them from turning it on and having you know
real users sign up I believe is just like their own internal bureaucracy like trying
to plan stuff last I heard it was supposed to be you know yeah it's supposed to be before
the end of the year so we're coming up on the end of the year should it should be going
live yeah yes everyone the blockchain transactions yes yeah so that's the public and permissioned
you could I'm sure it would be possible to create you know a private closed source version
of it yeah yeah so they're at least within RMG and now of course because it's open source
and you can fork off and add whatever you want anything is possible but at least within
RMG there while the the blockchain data itself is pseudonymous every actor within the system
is going to be identified so so every user has to go through AML KYC and you know you
won't be able to look at the blockchain and say oh that was Bob and you know sending a
hundred grams of gold to Jill or whatever but the administrators of the system would
be able to figure that out it's the same as Bitcoin right yeah yeah so you know if you're
running a full node for for Prova or RMG or whatever you would be downloading validating
on the transactions of course you could perform the same type of chain analytics that a number
of people perform on the public blockchains that are already out there but you would not
be able to really identify who's behind it unless you were in one of the systems like
they have that mapping private keys that have been handed out to the users yes yeah yeah
if there was some sort of breach if there was some sort of breach and of course you
know you could you could from from the top down you know the system is controlled by
a set of like root administrator keys these got stolen the whole system could get screwed
up so that's why we built you know additional offline signing software and have additional
security practices for how we believe the administrators should go about creating those
transactions so they will never be creating any of those administrator transactions on
any device that is you know touching the internet for example like those keys will remain on
specific hardware that never touches the internet oh yeah initial values hey if if anybody wants
to add that functionality or I guess maybe is it in these projects yeah pull in well
elements of course is the c++ so you have to port it but yeah I mean if there's demand
for that it would be great you know this is an open source project and we're trying to
get people interested you know contributing new features and whatever there's demand for
hopefully will get added in you know I doubt that the Royal Mint would want that but who
knows somebody else might want it yeah so to be specific the at least the normal trades
the normal gold trades that happen are still going to happen on like a centralized exchange
in most cases now you can you can use the peer-to-peer aspects of this to send the digital
assets to other people who are in the network but we expect that most of the trades are
still going to happen at a like traditional centralized trade exchange and in fact we'll
get to that in a second where they're one of the partners in this project is Alpha Point
which I believe is one of the trading engines so since I am like way on the back end of
doing infrastructure stuff I have no idea what the customers think yeah I mean we would
expect that a lot of the trading will still happen in the centralized exchanges and then
you know they're going to withdraw into their own wallets and then you know they could potentially
send back and forth with other users and other wallets in the system but you know that would
be up to them if somebody ended up building some like peer-to-peer you know decentralized
trading exchange then that would be a possibility let's see so some of the the reasons why the
Royal Mint wanted this basically trying to you know get out of the very old-school traditional
physical assets and start to bridge the gap into digital assets and really because they
have so much gold in their vaults they saw this as an opportunity actually to to offer
a a no vault fee solution for the customer so like one of the the driving reasons why
people might want to get in and use this system is that you will have a cryptographically
verifiable proof of ownership or at least proof of IOU to physical gold and you can
call in of course that IOU and get delivery of the physical gold but the difference between
this and other physical gold IOU systems is that they will not be charging you any vaulting
fees so you can keep your your digital gold assets as long as you want and they're not
going to be you know chipping away the the amount that you have held in their platform
I guess got a whole bunch of stuff if that's even readable so yes
Yeah, I have a question. Take for instance, you know, you have your vault in Singapore
the Royal Mint is not going to be charging the customers that use this Royal Mint gold
platform they're not going to be charging them ongoing vaulting fees so if you have
you know a kilogram of digital gold a lot of vaulting services would be charging you
some sort of ongoing maintenance costs just to keep it in their vaults but the Royal
Mint has I guess they're just large enough and have been around long enough that they
are able to offer a no vault fee solution just probably because they just already have
so much gold on hand that's going to be sitting there one way or another I forget how much
we're talking about but it's probably hundreds of millions if not billions there's a lot
of gold in there and actually in the top right here we have noted that their their gold in
their vault gets reconciled daily and is fully audited every six months so it would also
be interesting to see like how accessible those audits are they must be posted somewhere
because they have garnered quite a bit of trust over the past thousand years let's
see so you know this is still a trusted solution of course you know you you have a trusted
centralized third party that is holding the actual asset and basically we're building
a decentralized layer that just gives us more auditability of like what's actually going
on within the system and gives you some new abilities you know to to send this money in
a more efficient type of format like because we're now offering a network that is running
like 24 7 365 it's not like a lot of the traditional financial services that are only going to
be operating in set hours on set days of the week you are going to have a lot more freedom
to move around these digital IOUs for the digitized assets I know you someone was asking
like can you actually see the transactions that happen and let's see explore.rmgchain.info
should be up and running and you're not going to find a lot of transactions on there of
course because I don't I don't think they've actually taken the system live with their
users yet but we are creating blocks we you know have issued some gold into the system
and at this point it's just spinning away waiting for actual users to start getting
into the system so what is the rmg you can basically replace most of you know these instances
of rmg with bitcoin or litecoin or whatever it's using the same type of wallet platform
that we have used and expanded to support a number of different crypto assets over the
years and the like I said the onboarding process is going to be like AML KYC that is just required
you know from a regulatory standpoint but once you get into the system you can then
send the money around send the assets around to anyone else with inside the system without
having to go and you know ask ask your favorite banker or ask your favorite administrator
for permission to do that also we had the questions about trading and there is you know
fully fledged trading platform that is baked into this whole royal mint gold system I believe
yeah this was alpha point and their trading engine and so you know the trader folks market
makers I think are going to find this interesting in comparison to I guess like normal spot
gold like day traders this don't want to make any claims off hand but this might be one
of the first systems where it really is going to be like 24 7 365 gold trading at least
most other markets that I'm aware of you know are only open during specific hours no no
so so the trades the trades are not going to happen on the blockchain it's just going
to be like other assets where you can deposit into the exchange trade around and then withdraw
back into your wallet or or whoever else transaction performance so I'm trying to even remember
I believe we set our block times to be like two minutes or two and a half minutes and
I think we set our block sizes to be a couple of megabytes but we haven't done like a full
stress test of the system because it doesn't have any volume and it's going to be really
easy to hard fork so you know if we ever get anywhere near you know capacity and and you
know folks are actually using it so well I guess to actually answer your question that
would put us at what like eight or ten times Bitcoin transaction capacity so that would
be like what 30 transactions per second yeah so 30 transactions per second not bad we're
just gonna have to wait and see how it goes yeah so this is specifically for RMG but but
this is alpha points trading platform so alpha point makes their trading engine available
for anyone who wants to basically I guess pay them for like a white label solution
yes yes in this case yeah
you have the user key so yeah it's all two out of three multi-sig but like once you get
your RMG and it's deposited into your wallet that you have created you're now you're in
a two out of three key setup where you have one key the wallet provider has one key and
then a third party recovery service has the third key let's see we right so this is specifically
using like bit 32 key derivation I we do not support the bit for the like 12 12 word recovery
phrase we have we create like when you create your wallet you get a PDF basically with your
encrypted recovery data on it it's just like blobs of JSON basically but then we have recovery
tools and recovery processes and the key recovery service is its own type of protocol that we
developed year year and a half ago but at the at the beginning of like RMG launch the
only wallet is going to be bit go we are a web wallet but we also have API's and SDK's
so really the vast majority of people that use bit go do not use us as a web wallet they
actually have our SDK running on their own servers and they're doing all their key signing
and stuff on their own computers and sending a half signed transaction to our server and
then we then you know decide whether or not to co-sign yeah yeah so so this is specifically
gold trading with RMG but you know from a sort of stepping back at a higher level you
could potentially do this with any physical asset that you want to digitize using Prova
software to you know create a new blockchain a new network for whatever digital asset it
is that you want to use so this is kind of just like going more into the specifics about
you know the first production use case of RMG these are some of the partners that came
together to do this basically bit go which is my employer we were the ones who actually
developed the protocol and we were able to reuse a lot of work that we've done over the
years with regards to key signing and key recovery type of technologies and then Alpha
Point came in with their trading software CME is of course on the financial side and
the Royal Mint is really the the gold vaulting side and they're the primary drivers of all
of this and so you know bit go we have a long track record we've been used by a lot of different
digital asset related companies and that basically covers all of it so if we have more questions
looks like we've got one back there yep that is a good question I don't as far as I'm
aware that no I believe Alpha Point is going to be the one actually managing the trading
aspect of it and of course long term we don't want that to be the only exchange like because
because this is a blockchain protocol that is very very very similar to Bitcoin we expect
that if this system gains significant adoption then you will see a lot of other like crypto
exchanges out there that may want to add support for it so long term RMG will hopefully get
traded on many different exchanges yeah if there is you know sufficient demand and volume
then any number of trading pairs could get created and I think you know I think that
there have been like one or two attempts at like crypto tokens that are supposedly backed
by gold over the years and that they've all really fallen apart and I think that most
of that is due to the fact that none of the vaulting institutions had really anywhere
near the level of trust that the Royal Mint has it's always been sort of like fly-by-night
companies that you know pop up out of nowhere and like you're not even really sure if they're
being honest with you so hopefully you know this particular system will be able to to
do a lot better and go a lot further than the previous tries at a gold token yeah.
So is Royal Mint actually a royalty?
Well I believe it was Her Majesty's Royal Mint so I believe there is some you know lineage
that goes back to royalty basically saying you know we need to have a mint for the crown
basically right the crown mint and then we will issue our money from that so I believe
that there is a lineage that you know goes back with royalty but like the Royal Bank
of Scotland for instance you know is it a derivative of some actual physical mint but
you say that it's kind of dispersed over time?
Yeah well I'm not actually you know completely familiar with the like the history of you
know ownership and management of the mint.
You'd have to ask our CEO that because he's the one who does the business side I'm just
the tech guy.
So is there a take at some point to be able to do payments across this between like the
user and sort of self-service transfer role to me as payment?
Yeah I mean there's no reason why you couldn't yeah there's no reason why you couldn't use
it as a payment rail.
Of course.
Do you think that's their intent?
Well you know this is definitely starting with like store of value.
This is like this is next generation technology for gold bugs right.
When you invest in this you're basically buying some portion of the gold they maintain right.
They sell it anyway so I mean.
Yeah so.
So you can use this.
So I can see this becoming a real value a gold backed cryptocurrency base with what
it looks like.
Is there a do they do anything with NASDAQ when they get along?
Are they associated with that?
Not that I'm aware of but I'm not sure.
In the consensus I thought that they were talking about some kind of association with
that.
Yeah but no I don't know but no the so yeah so the like one RMG which is like the same
unit as one Bitcoin one RMG is equivalent and will always be equivalent to one gram
of gold and I don't I don't remember what the minimum limitation is but whatever when
you have these tokens these IOUs you can send them to the Royal Mint and cash in you know
for physical delivery of that amount of gold.
I'm sure they have some minimum order there.
I'm not sure what it is and there's probably a fee associated but that is I think you know
why this system has a lot more promise than previous like gold cryptocurrencies.
This one you know nothing is a hundred yeah exactly yeah so I mean we could you can feasibly
use this network to build a new type of payment system or you know you could integrate it
into your current crypto payment systems and say hey do you want to send Bitcoin or Litecoin
or Ethereum or RMG or what have you and of course we certainly hope that does happen
and that we get to the point where we have to start worrying about on-chain scalability
problems.
Are there any implementations other than RMG kind of coming out of the pipe of Croma or
anything you've considered maybe have used before?
So the way that I think this is going to play out is that we are constantly in talks with
all kinds of traditional financial institutions and that inevitably some of them will have
you know a similar need to a system like this but it won't be exactly the same and so they're
going to I guess this is kind of one of one of the the new business avenues that Midgo
is going down is you know custom blockchain type of service and so you know when we talk
to these various institutions if they have a need for their own permissioned blockchain
and we believe that you know Prova is a good fit for that but might need a little tweaking
then it would probably come down to you know us working on some new specifications for
new features that they would want in their protocol and we roll it into Prova and of
course it gets open source to use at that point but that would be kind of my ultimate
hope is that a lot of these institutions end up with Midgo to write open source software
that can be used by anybody else who wants to use it.
Yeah so you know it's going to use the exact same type of probably level DB data store
well at the node level you know whatever BTCD uses which is probably level DB but then you
know no you can't build enterprise infrastructure on level DB and really this is what has happened
in the entire blockchain ecosystem but so like you can build you can build nodes and
consensus systems that use like level DB type stuff but if you're then if you then want
an enterprise grade like we're doing you know hundred thousand queries every you know few
seconds type of thing then you're going to want to have probably some more scalable type
of database so like what we do this is really what my job is at BitGo is to build these
backend infrastructure and really what I do is I build services that connect to the nodes
that are running on level DB or BDB or whatever and just suck all the data out of them and
then put them into a more scalable data structure like you know Hadoop or HBase or even Mongo
works pretty well for a lot of small medium sized stuff.
Yeah so it's we've tried not to stray too far from like what the standard technology
is that technology stack that has been used and no no so all of the permissions happen
like at a protocol level well you could argue I mean there is data in the data store that
says you know who has permission to do what and that is the the administrator you know
chain of actions throughout the history of the Prova blockchain basically says you know
we're changing this permission setting or that permission setting or what have you and
so from that you build sort of administrative UTXO set that is like what is the current
settings of the system from a permissioning standpoint.
Yes.
They say they have a hundred million times more with it so it's something that's not
going to happen in the future but it's something that's not going to happen in the future.
Yeah so you know unlike a lot of the paper ETFs I think the auditing requirements are
a lot higher and so Toilment does audits twice a year they say they reconcile their reserves
daily into a full audit twice a year I I imagine those audits are available publicly they if
they're not then that would be kind of worthless it would be particularly yeah yeah they've
been doing that for who knows how long.
A thousand years yeah yeah definitely yes so this this is not meant to be the like record
of their gold R&D is actually only a tiny tiny fraction of their vaulted gold product
so percentage but basically as they onboard new users they will allocate more of their
vaulted gold to be RMG gold and as they do that they will that will get reflected in
their audits and it will get reflected in the system as they provision or allocate and
deallocate the digital assets within RMG yeah yeah yeah so of course amongst the ability
to provision and deprovision the the validator keys and the wallet keys there has to also
be a way to provision and deprovision the administrator keys and these of course there
these will be like initialization parameters that need to get decided by the creators of
the system when they originally create it I forget I'm not even sure like what RMG settled
on it was probably something like five out of seven or five out of ten or something like
that so you know you want to have both redundancy and requirement of you know multiple trusted
people that you not going to all collude against you at the same time right
well at the end of the day like the whole system can get changed in any way that the
owners want so it's not it's not a it's not a like permanent security concern there are
of course basically if enough administrator keys got compromised someone could really
screw with the system for a short period of time but people would notice pretty quickly
and then you know the rest of the administrators would come in and say no we need to fix this
you mean for like trying to detect compromised keys
or something yeah it wasn't I guess too much of a concern at least for RMG yeah so that's
what like what we're really trying to do in this system is create automated logic that
can find you know anything that's going wrong in the system it's a lot more important to
have that like fully fledged 100% coverage when you're talking about a public permissionless
system you can be a little bit more lax when you're talking about like a system that is
controlled by a number of companies or entities or what have you so yeah I mean it's certainly
possible it might it might not be worth it you know the risk reward like it I imagine
they might think the risk is so low and the payoff like you know saving numbers of downtime
or something might not be worth like the effort that would require to go into that
yeah they they must be saving some money somewhere right it's also as good good marketing I guess
you know get in on on the blockchain craze while it's it's still really hot sector
yeah just like performance issues of having too many transactions yeah you know we we
hope that that is a problem that we will have to deal with but like I said like at the finding
the bottlenecks and whatnot yeah yeah we hope that that is a problem that we have to deal
with and you know then have to spend more engineering time solving scalability issues
but as for right now we're just waiting for them to actually turn the system on and and
then you know we'll adjust as things go I'm sure Bicco has some sort of contract like
in place already you know with the Royal Mint that basically says you know we're going to
be monitoring and making adjustments as necessary you know as the system grows and has growing
pains it's pretty much inevitable yeah
scalability is really going to depend on a number of different things like we we don't
have Segwit support in Prova because I don't think BTCD has Segwit support I'm pretty sure
the the beat most of the BTC developers don't work on it anymore I think they work on Decred
mostly they've created their own cryptocurrency and like with regard to lightning payment
channel stuff maybe of course you would need Segwit first that's that would be a really
really good problem to have like if we got to the point where we needed a second layer
network but then we've talked about this a little bit creating second layer networks
could potentially cause regulatory compliance issues maybe you then you start getting into
you start enabling all these really cool cypherpunk things like you know cross-chain atomic swaps
and all that stuff and so then the question becomes you know what happens if if someone
is you know using or owning or trading RMG and they're actually an unidentified person
you know what actors within this regulated system might be in like legal trouble if there
are people who are doing that I don't know I'm not a lawyer it sounds cool to me but
I think it is like some sort of potential regulatory issue I mean it's a brave new world
and we're creating these and completely new systems that have never enabled this type
of stuff before yeah
You mean like if if the administrators or the wallets were to like feign ignorance
and say you know we can't we can't let you you know claim your physical delivery of the
gold anymore I mean just I mean in a world where the regular societal rules are present
you have the court system you have ways to sort of fight your battles but if some of
that stuff falls apart especially when you're talking about gold which is a durable physical
asset and if they're able to keep it there's not really if you have no record of what you
own that verifiable outside of the blockchain itself they could just keep all that gold
right?
Yeah so this would not be the most ideal system for like a doomsday prepper gold investor
so I would say like really like your best recourse of
Yeah but or or they're buying gold because they know other people are buying gold because
that doesn't mean all these people buying these gold ETF products like they're never
going to be able to get the gold
Yeah so that's what happened I mean when people started trying to buy physical gold they cold
up the Royal Mint or anybody else and said hey I want to buy some gold and they found
out it was pretty hard to do to figure out if they could have it but the Royal Mint said
well okay I don't want to lose that customer I'll just give them something like this so
I think that's what they're doing
Yeah so I would say that like your best recourse in that type of adversarial situation would
have to be the courts and the legal system and there's actually a strong precedent there
of you know cryptographic signatures and and and chains of history you know we had a we
actually had a lawyer speak here last year and she talked about how like there's already
a number of court cases with precedence of of you know proving ownership of cryptographic
assets so
So there's a record of outside of their interest system that you'd be able to save somehow
Well there's the blockchain itself so anyone who wants to can run the software download
and verify the entire history of the blockchain and have a copy of it and then be able to
show you know I have the private cryptographic key that unlocks the funds in these addresses
and so you know you would probably have to pull in an expert witness in the court but
my understanding is that this has already been done a number of times for other cases
Have you seen this one the blockchain, the lawyers in Switzerland for the you know you
sign up I mean to monetize and everything it's a blockchain for lawyers for the goal
and the lawyers have designed that you know saying you have to sign up that just in case
anything happens you know it's I mean they've designed that then you sign up first with
them and then for them to go if anything goes wrong with that the lawyers already have the
blockchain for that and they can prove to them that it's already been set up for that
and my one is pulled out and all that stuff Switzerland is already been up and running
now you know what I'm saying for that you know what I'm saying
No I mean there's so many of these you know blockchains and various crypto token products
You already have one which is up and running right now with things like trading and all
that stuff it's up and running and all that stuff just in case anything you know to prove
and everything and all that stuff
Yeah I think you know there is going to be well a lot of industries are going to have
to adapt you know the legal industry is already adapting you know to incorporate this technology
into their day to day use it's just another tool to add to the toolbox whatever it is
that you're doing
Yeah
What inside the vault?
What inside the vault?
So the problem with I guess blockchains in general is the same problem of any database
is garbage in garbage out right?
So at the end of the day some human is going to be putting those numbers in until we get
like some you know now I've been hearing rumblings of like blockchain plus AI which I don't even
really know what that means yet but you know if we get to the point where you can have
like trustless robots you know that they've been programmed to do their task and they're
set up in such a way that you can be 100% sure that they're only doing the exact thing
that they were programmed to do and then they can go like those robots can go audit the
gold you know maybe you can start to create some sort of new you know trust chain from
that standpoint
Yeah yeah so that's definitely one of the primary differences even though we have we
can be comparing two different systems that have administrators your traditional database
system the administrator if they really know what they're doing can probably go in there
and delete stuff and basically cover their tracks but while the administrators in this
system can technically screw you over and delete your money out of the system they have
to do that in a transparent way that everybody else in the system has a record of
Go call for a bus man that's a new ICO
Yeah yeah I think healthcare is also gonna get revolutionized by all this
Thank you very much