Proof of Forestry: a currency with a smart contract to combat climate change
132 views
Skip to first unread message
Christine Lemmer-Webber
Jan 2, 2022, 7:39:23 PM1/2/22
to cap-...@googlegroups.com
Here's an idea I've had kicking around. I've thought it might be an
interesting way to upend peoples' assumptions of what a "cryptocurrency"
entails, and for that matter, "smart contracts" (I hate the "smart
contracts are code that runs on a blockchain" redefinition).
It's at least a useful concept. Is it possible to actually work? I
think so but I'm not sure. I've thought about writing this on my blog
or something but I've held off fearing I'd be ridiculed. But maybe
cap-talk is the right space of people who can be critical thinkers about
this without ridicule.
Ok here goes:
- Start with a deterministic abstracted machine/vat running on Spritely
Goblins, Agoric's stuff, whatever. The vat/machine is deterministic
in that anyone who receives the agreed upon incoming set of messages
is able to run the same abstract machine.
- Select five countries with five very different legal jurisdictions
which you believe have a significant likeliness to allow this project
to continue. Now select five official stewards in those countries.
This is now a quorum: if three out of five machines agree that X is
the next message, then X is the next message. Transactions can thus
be wickedly fast, but no one (or even two) countries can shut it down.
- Now throw a mint on it, either the classic Ode-style E mint or
Jessica's Pebble Bank. It doesn't really matter, either of those two
will do.
- Wait, this sounds awfully fiat currency like! Well, we're going to
wire something special up to the mint-new-currency capability. We'll
come back to that in a second though.
- We'll also assume there's an ocap market/escrow-like system where
anyone can exchange between currencies pretty easily. Thus you don't
need "one really good" currency, you can have a lot of them, people
can converge on a curency as long as they trust it and move away when
they don't.
- Okay, we now the foundation to make these currency things, and one
that runs with a simple three-out-of-five quorum where we're going to
try to do something particularly nice that we hope people like and we
hope will take off but which is also efficient and cheap enough to
use that people both feel it's a solid system but also encourage its
use. We'll call it ForestCoin for now and hopefully what we hear
people say is, "You should accept ForestCoin!" and people feel pretty
comfortable using it for everyday purposes the way people presently
use USD.
- We're not aiming for something that's a good "investment", inflation
is okay; let's say about 1-2% a year. We won't prevent people from
having gold-like or stonks-like systems. In a sense we're embracing
Gresham's Law, that "bad money drives out good". People may be more
reluctant to spend something that seems like an investment because
it's too "good". But fast and loose everyday transactions may even
be desirable to have something that doesn't feel like an
"investment". (You can always convert to and from those!)
- So, that brings us to... mining! Or rather, our replacement...
forestry! Who gets it? Well, how about... people who plant and
maintain actual, traditional forests?
- You may be observing: this sounds like something that can't be
verified "on-machine" like proof-of-work or proof-of-stake. That's
true and ok, we're just embracing that. Trees don't "live in the
blockchain". So yeah, we need some sort of outside oracle.
- We'll need some seriously good governance here. I won't claim to be
the right person for it. For now I'll handwave massively: some
university, or maybe some UN commission or something, agrees to:
- Set up a panel of scientists who set up the initial set of
qualifications to receive a portion of that year's "grown" currency
- Something something delegation to an authorized set of beauraucrats
and surveyors to assess claims. Each delegation to a surveyor
is signed off from someone in the beauraucracy and is open to
public scrutiny.
- Every time money is minted, it's immediately handed over to the
relevant party doing the forestry. However, the capability can
only be invoked with submitted evidence and the sign-off from
someone who is authorized and can attest that the surveying seems
correct, which is bound to their professional reputation.
There's some more stuff that can be said, I have some thoughts on:
- How to incentivize governments to *not* shut this down and actually
buy in without controlling it
- How transaction fees could work in a sensible way, and how they could
be spent (including on social services... by governments or directly
funded by the program itself)
But I think I'm going to stop here. The obvious problem here is that
the oracle of the beauraucracy that can authorize the minting new
currency, as handed out to qualifying programs, is subject to
corruption. But for all the complaints about government that exist,
there is an impressive amount of *non*-corruption that occurs amongst
civil servants, and I've worked with and been amazed by the seriousness
that many civil servants can take their jobs, so I will say that this
isn't an impossible thing but it does require much care.
But yeah, I have no idea how to actually bootstrap the organization that
does all this. I can design all the technology they would use top to
bottom (or at least, I think I could), but I'm not in a position to
decide what's healthy for the earth or who's qualified to run such
programs. We really need some experts who think carefully about
governance.
I dunno if this is practical. But maybe it at least paints a picture
that isn't the default, popular assumptions of how this stuff has to
work. Hope you had fun reading it...
- Christine
PS: Oh yeah, and that very mechanism for signing off on minting new
currency: that's an ocap smart contract (or a series of smart contract
submachines), obviously. This whole thing can run on one server/machine
but for this use case/robustness, obviously we want a quorum. But
depending on your defition (are quorums blockchains?), this is also an
example of a smart contract that doesn't necessarily run on a
blockchain!
interesting way to upend peoples' assumptions of what a "cryptocurrency"
entails, and for that matter, "smart contracts" (I hate the "smart
contracts are code that runs on a blockchain" redefinition).
It's at least a useful concept. Is it possible to actually work? I
think so but I'm not sure. I've thought about writing this on my blog
or something but I've held off fearing I'd be ridiculed. But maybe
cap-talk is the right space of people who can be critical thinkers about
this without ridicule.
Ok here goes:
- Start with a deterministic abstracted machine/vat running on Spritely
Goblins, Agoric's stuff, whatever. The vat/machine is deterministic
in that anyone who receives the agreed upon incoming set of messages
is able to run the same abstract machine.
- Select five countries with five very different legal jurisdictions
which you believe have a significant likeliness to allow this project
to continue. Now select five official stewards in those countries.
This is now a quorum: if three out of five machines agree that X is
the next message, then X is the next message. Transactions can thus
be wickedly fast, but no one (or even two) countries can shut it down.
- Now throw a mint on it, either the classic Ode-style E mint or
Jessica's Pebble Bank. It doesn't really matter, either of those two
will do.
- Wait, this sounds awfully fiat currency like! Well, we're going to
wire something special up to the mint-new-currency capability. We'll
come back to that in a second though.
- We'll also assume there's an ocap market/escrow-like system where
anyone can exchange between currencies pretty easily. Thus you don't
need "one really good" currency, you can have a lot of them, people
can converge on a curency as long as they trust it and move away when
they don't.
- Okay, we now the foundation to make these currency things, and one
that runs with a simple three-out-of-five quorum where we're going to
try to do something particularly nice that we hope people like and we
hope will take off but which is also efficient and cheap enough to
use that people both feel it's a solid system but also encourage its
use. We'll call it ForestCoin for now and hopefully what we hear
people say is, "You should accept ForestCoin!" and people feel pretty
comfortable using it for everyday purposes the way people presently
use USD.
- We're not aiming for something that's a good "investment", inflation
is okay; let's say about 1-2% a year. We won't prevent people from
having gold-like or stonks-like systems. In a sense we're embracing
Gresham's Law, that "bad money drives out good". People may be more
reluctant to spend something that seems like an investment because
it's too "good". But fast and loose everyday transactions may even
be desirable to have something that doesn't feel like an
"investment". (You can always convert to and from those!)
- So, that brings us to... mining! Or rather, our replacement...
forestry! Who gets it? Well, how about... people who plant and
maintain actual, traditional forests?
- You may be observing: this sounds like something that can't be
verified "on-machine" like proof-of-work or proof-of-stake. That's
true and ok, we're just embracing that. Trees don't "live in the
blockchain". So yeah, we need some sort of outside oracle.
- We'll need some seriously good governance here. I won't claim to be
the right person for it. For now I'll handwave massively: some
university, or maybe some UN commission or something, agrees to:
- Set up a panel of scientists who set up the initial set of
qualifications to receive a portion of that year's "grown" currency
- Something something delegation to an authorized set of beauraucrats
and surveyors to assess claims. Each delegation to a surveyor
is signed off from someone in the beauraucracy and is open to
public scrutiny.
- Every time money is minted, it's immediately handed over to the
relevant party doing the forestry. However, the capability can
only be invoked with submitted evidence and the sign-off from
someone who is authorized and can attest that the surveying seems
correct, which is bound to their professional reputation.
There's some more stuff that can be said, I have some thoughts on:
- How to incentivize governments to *not* shut this down and actually
buy in without controlling it
- How transaction fees could work in a sensible way, and how they could
be spent (including on social services... by governments or directly
funded by the program itself)
But I think I'm going to stop here. The obvious problem here is that
the oracle of the beauraucracy that can authorize the minting new
currency, as handed out to qualifying programs, is subject to
corruption. But for all the complaints about government that exist,
there is an impressive amount of *non*-corruption that occurs amongst
civil servants, and I've worked with and been amazed by the seriousness
that many civil servants can take their jobs, so I will say that this
isn't an impossible thing but it does require much care.
But yeah, I have no idea how to actually bootstrap the organization that
does all this. I can design all the technology they would use top to
bottom (or at least, I think I could), but I'm not in a position to
decide what's healthy for the earth or who's qualified to run such
programs. We really need some experts who think carefully about
governance.
I dunno if this is practical. But maybe it at least paints a picture
that isn't the default, popular assumptions of how this stuff has to
work. Hope you had fun reading it...
- Christine
PS: Oh yeah, and that very mechanism for signing off on minting new
currency: that's an ocap smart contract (or a series of smart contract
submachines), obviously. This whole thing can run on one server/machine
but for this use case/robustness, obviously we want a quorum. But
depending on your defition (are quorums blockchains?), this is also an
example of a smart contract that doesn't necessarily run on a
blockchain!
Christine Lemmer-Webber
Jan 2, 2022, 7:46:03 PM1/2/22
to cap-...@googlegroups.com
Christine Lemmer-Webber <cwe...@dustycloud.org> writes:
> Here's an idea I've had kicking around. I've thought it might be an
> interesting way to upend peoples' assumptions of what a "cryptocurrency"
> entails, and for that matter, "smart contracts" (I hate the "smart
> contracts are code that runs on a blockchain" redefinition).
>
> It's at least a useful concept. Is it possible to actually work? I
> think so but I'm not sure. I've thought about writing this on my blog
> or something but I've held off fearing I'd be ridiculed. But maybe
> cap-talk is the right space of people who can be critical thinkers about
> this without ridicule.
>
> Ok here goes:
>
> - Start with a deterministic abstracted machine/vat running on Spritely
> Goblins, Agoric's stuff, whatever. The vat/machine is deterministic
> in that anyone who receives the agreed upon incoming set of messages
> is able to run the same abstract machine.
>
> - Select five countries with five very different legal jurisdictions
> which you believe have a significant likeliness to allow this project
> to continue. Now select five official stewards in those countries.
> This is now a quorum: if three out of five machines agree that X is
> the next message, then X is the next message. Transactions can thus
> be wickedly fast, but no one (or even two) countries can shut it down.
>
> - Now throw a mint on it, either the classic Ode-style E mint or
> Jessica's Pebble Bank. It doesn't really matter, either of those two
> will do.
know about the Ode-style E mint, but here it is anyway:
http://erights.org/elib/capability/ode/ode-capabilities.html
And here's Jessica's "Pebble Bank" design blogpost (posted here recently):
https://spritelyproject.org/news/pebble-bank.html
Alan Karp
Jan 2, 2022, 8:28:43 PM1/2/22
to cap-...@googlegroups.com
I'm not sure what you're proposing is doable, but just in case I have a couple of comments.
If majority voting was sufficient for such systems, Lamport would not have proposed Paxos. I think you need to consider various failure modes of your consensus mechanism.
Rather than a bureaucracy providing smart contracts to mint more currency, can that be done by locking in a smart contract at the beginning that mints new currency at a fixed rate? A fixed expansion rate might not be flexible enough, but perhaps we understand economics well enough to build in enough variability to be good enough.
--------------
Alan Karp
--------------
Alan Karp
--
You received this message because you are subscribed to the Google Groups "cap-talk" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cap-talk+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cap-talk/87zgodtzyh.fsf%40dustycloud.org.
Christine Lemmer-Webber
Jan 2, 2022, 8:41:48 PM1/2/22
to cap-...@googlegroups.com
Christine Lemmer-Webber
Jan 2, 2022, 8:49:22 PM1/2/22
to cap-...@googlegroups.com, Alan Karp
Alan Karp <alan...@gmail.com> writes:
> I'm not sure what you're proposing is doable, but just in case I have
> a couple of comments.
>
> If majority voting was sufficient for such systems, Lamport would not
> have proposed Paxos. I think you need to consider various failure
> modes of your consensus mechanism.
Not suggesting that consensus is not required, but rather that
> I'm not sure what you're proposing is doable, but just in case I have
> a couple of comments.
>
> If majority voting was sufficient for such systems, Lamport would not
> have proposed Paxos. I think you need to consider various failure
> modes of your consensus mechanism.
Earth-sized consensus is not required.
> Rather than a bureaucracy providing smart contracts to mint more
> currency, can that be done by locking in a smart contract at the
> beginning that mints new currency at a fixed rate? A fixed expansion
> rate might not be flexible enough, but perhaps we understand economics
> well enough to build in enough variability to be good enough.
could be set up front and be a hard limit of the system.
Justin Cormack
Jan 3, 2022, 5:35:56 AM1/3/22
to cap-...@googlegroups.com
Bill Frantz
Jan 3, 2022, 4:56:53 PM1/3/22
to cap-...@googlegroups.com
On 1/2/22 at 6:21 PM, cwe...@dustycloud.org (Christine
In-order message handling as seen by an external observer
Single-thread execution as seen by an external observer
No sources in indeterminism e.g. a clock
I will add that the idea that money grows on trees is highly amusing.
More seriously, proof of growth has some nice "green"
advantages. Here in New Hampshire, some entities tout their
greenness by saying that their HVAC systems run on wood pellets.
(There is a good argument that wood pellets are a reasonable way
of storing solar energy for a few years. Wood pellet furnaces
are almost as automatic and low maintenance as oil or gas burners.)
Cheers - Bill
-----------------------------------------------------------------------
Bill Frantz |Security, like correctness, is| Periwinkle
(408)348-7900 |not an add-on feature. - Attr-| 150
Rivermead Rd #235
www.pwpconsult.com |ibuted to Andrew Tanenbaum |
Peterborough, NH 03458
Lemmer-Webber) wrote:
>- Start with a deterministic abstracted machine/vat running on Spritely
>Goblins, Agoric's stuff, whatever. The vat/machine is deterministic
>in that anyone who receives the agreed upon incoming set of messages
>is able to run the same abstract machine.
So the vat/machine has at least these restrictions:
>- Start with a deterministic abstracted machine/vat running on Spritely
>Goblins, Agoric's stuff, whatever. The vat/machine is deterministic
>in that anyone who receives the agreed upon incoming set of messages
>is able to run the same abstract machine.
In-order message handling as seen by an external observer
Single-thread execution as seen by an external observer
No sources in indeterminism e.g. a clock
I will add that the idea that money grows on trees is highly amusing.
More seriously, proof of growth has some nice "green"
advantages. Here in New Hampshire, some entities tout their
greenness by saying that their HVAC systems run on wood pellets.
(There is a good argument that wood pellets are a reasonable way
of storing solar energy for a few years. Wood pellet furnaces
are almost as automatic and low maintenance as oil or gas burners.)
Cheers - Bill
-----------------------------------------------------------------------
Bill Frantz |Security, like correctness, is| Periwinkle
(408)348-7900 |not an add-on feature. - Attr-| 150
Rivermead Rd #235
www.pwpconsult.com |ibuted to Andrew Tanenbaum |
Peterborough, NH 03458
Tony Arcieri
Jan 5, 2022, 3:15:54 PM1/5/22
to cap-...@googlegroups.com
Not an endorsement, but you might check out https://www.regen.network/.
--
You received this message because you are subscribed to the Google Groups "cap-talk" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cap-talk+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cap-talk/87zgodtzyh.fsf%40dustycloud.org.
Tony Arcieri
Reply all
Reply to author
Forward
0 new messages