Amoveo ♥🧿
Previous messages
Next messages
28 December 2020
MF
09:22
Mr Flintstone
what if they just prevent US users?
09:22
I guess that isnt technically a delist
mx
09:22
mr x
i put halt trading
09:23
what bitstamp announced "In light of the recent SEC filing against Ripple Labs Inc., which alleges that XRP is a security, we are going to halt all trading and deposits of XRP for our US customers on 8 January 2021 at 9 PM UTC."
09:24
yeah for us customers only i guess
mx
09:43
mr x
"Coinbase to halt XRP trading for US customers before February 2021" ?
MF
09:51
Mr Flintstone
yeah
09:51
Makes sense to me
mx
10:16
mr x
Ok created new and sold liquidity shares of the old one
Z
10:17
Zack
In reply to this message
after selling the liquidity shares, you probably want to use the contracts.html page to combine the subcurrency types back into veo
mx
10:17
mr x
i used contracts.html for doing both
10:18
how to sell liquidity in wallet.html?
Z
10:18
Zack
I see the xrp bet showing up in the dropdown
mx
10:18
mr x
yeah
Z
10:18
"lookup price to sell all liquidity shares"
mx
10:20
mr x
hmm
10:21
"you can sell all your liquidity shares to receive:
NaN of currency type AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=,0
NaN of currency type de01XJM9elbsxRgvHp5ELLTqgWHEHguvibeq5zXFTNs=,1
NaN of currency type de01XJM9elbsxRgvHp5ELLTqgWHEHguvibeq5zXFTNs=,2"
NaN of currency type AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=,0
NaN of currency type de01XJM9elbsxRgvHp5ELLTqgWHEHguvibeq5zXFTNs=,1
NaN of currency type de01XJM9elbsxRgvHp5ELLTqgWHEHguvibeq5zXFTNs=,2"
10:21
if i try it now
Z
10:21
Zack
NaN is javascript for "not a number".
you don't have any liquidity shares to sell.
you don't have any liquidity shares to sell.
10:22
I guess I should make a better error message in that case
mx
10:22
mr x
i have imo
10:22
balance: 1.99999
10:22
no wait
10:22
ugh
10:24
okay
10:24
"you can sell all your liquidity shares to receive:
0.00001998 of currency type AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=,0
1.99999998 of currency type pG4Byp2kBbxzaegWZPN5OsBc75+FedNOSufqDXXn074=,1
1.99999998 of currency type pG4Byp2kBbxzaegWZPN5OsBc75+FedNOSufqDXXn074=,2"
0.00001998 of currency type AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=,0
1.99999998 of currency type pG4Byp2kBbxzaegWZPN5OsBc75+FedNOSufqDXXn074=,1
1.99999998 of currency type pG4Byp2kBbxzaegWZPN5OsBc75+FedNOSufqDXXn074=,2"
10:24
looks good
29 December 2020
I
JS
MF
06:46
Mr Flintstone
the prediction market was right!
Z
07:00
Zack
http://159.89.87.58:8080/contracts_list.html
I made a page to list a lot more of the contracts
I made a page to list a lot more of the contracts
mx
07:22
mr x
haha that was quick
S
07:53
Sy
whats with the unknown oracle texts?
Z
07:55
Zack
The database of oracle texts got deleted at some point
07:57
Also i occasionally make contracts when i run all the servers on my local computer, and it pushes the tx to mainnet. But the oracle text is on the server on my laptop.
OK
栾
08:32
栾峰
8
Nick Albernas invited Nick Albernas
Z
OK
Z
09:47
Zack
In reply to this message
is the address you are sending to formatted correctly? is there an error message in the browser console? ill try sending to make sure it works.
09:48
I think I reproduced the error
09:49
oh nevermind. it is working for me. the problem was that I tried spending to myself
09:50
JS
OK
x
10:24
x
i'm not in US
10:24
but i can use qtrade
JS
10:56
Jon Snow
In reply to this message
Even after you registered to them that you are not in the US?
x
10:59
x
i think qtrade is used by people from every country
OK
11:23
I know non-US who use it.
11:23
I can maybe buy some for you for a fee.
JS
11:31
Jon Snow
In reply to this message
That’s good to know. I thought they are US registered exchange and could only service people in the US. Good to know it is more than that.
11:39
Very sad qtrade cannot offer service in North Korea
11:40
Kim Jung Un might want to have some VEO to get USD stable coin
D
12:14
😅
Davis Smith invited Davis Smith
FR
20:52
Fahim Ryhan
Are you going to listed VEO on moonswap? I think moonswap is a good choice. It has high speeds and 0 gas fees
Z
20:58
Zack
Amoveo is a blockchain. Veo is the native currency of amoveo. Veo is not an ethereum subcurrency.
R
21:45
Rom
Would it be possible to create a Wrapped VEO on uniswap for example to give more visibility to the project ?
30 December 2020
MF
00:03
Mr Flintstone
i think exantech was working on this but it wouldnt likely be out for some time:
http://159.89.87.58:8080/contract_explorer.html?cid=fwwL387sbGiDeTfPa73fEcHNdKkoafJdMeYipkboChE=
http://159.89.87.58:8080/contract_explorer.html?cid=fwwL387sbGiDeTfPa73fEcHNdKkoafJdMeYipkboChE=
00:03
this reads as a 10% chance by end of February
S
07:24
Sebsebzen
go go Exan
07:24
is that the same as a DAC?
Z
07:52
Zack
not the same as a DAC, since anyone can trade in it.
It is just a prediction market
It is just a prediction market
MF
07:54
Mr Flintstone
it can incentivize people similar to a DAC and get to the same endpoint though
07:56
there isnt enough liquidity to incentivize them in this market to actually build software they otherwise were not going to build, but there might be enough liquidity to incentivize someone who has information about the exantech software development process to move the price to the correct probability
31 December 2020
MF
00:56
Mr Flintstone
lol
00:56
do you guys think it is not obvious what you are doing
00:56
at least this is more sophisticated than other attemps
Z
00:57
Zack
Banned
Z
05:53
Zack
http://159.89.87.58:8080/contract_explorer.html?cid=jz5+NialTR/Ktymo2jb7PBDdVIGcSKVXwVw0bA6dEqY=
I added open interest to the contract explorer.
I added open interest to the contract explorer.
L
06:53
L
What's the GitHub link for the contract explorer frontend?
Z
06:55
here is the html that loads that javascript https://github.com/zack-bitcoin/light-node-amoveo/blob/master/src/js/contract_explorer.html
06:55
it is using other javascript libraries from that same folder. it is all part of the light node.
L
06:58
L
Yeah thanks! Also I was wondering if you had explored using gitcoin or a similar site for bounties. Say for the todo list.
Z
06:59
Zack
I haven't looked into gitcoin.
We want to be able to use Amoveo smart contract to fund Amoveo development https://github.com/zack-bitcoin/amoveo-docs/blob/master/use-cases-and-ideas/dominant_assurance_contract.md
To make it trustless and decentralized.
We want to be able to use Amoveo smart contract to fund Amoveo development https://github.com/zack-bitcoin/amoveo-docs/blob/master/use-cases-and-ideas/dominant_assurance_contract.md
To make it trustless and decentralized.
L
08:03
L
I see. Will look into it.
B
16:17
Ben
the problem is the usability, gitcoin could attract a very wide range of devs, if you rely only smart contract funding, you basically have to lure the devs in here, since how should they know otherwise about it.
16:17
using the exposure of a platform like gitcoin could be beneficial.
Z
16:18
Zack
If it is just about advertising, there are plenty of places to advertise, right?
B
16:22
Ben
if you look into that platform, you will realize it is not
Z
16:23
Zack
I dont want lose money to be some developer's babysitter.
Writing up proposals, and reviewing their work, and arguing with them about whether it meets the spec.
If this is going to be sustainable and exponentially growing, without a central point of failure, then developers need a way to get involved and get paid without me doing anything.
They need to make their own proposals, raise their own funds from the community, and a decentralized oracle needs to decide if their work matches their proposal.
Writing up proposals, and reviewing their work, and arguing with them about whether it meets the spec.
If this is going to be sustainable and exponentially growing, without a central point of failure, then developers need a way to get involved and get paid without me doing anything.
They need to make their own proposals, raise their own funds from the community, and a decentralized oracle needs to decide if their work matches their proposal.
B
16:23
Ben
it is a platform where you can attract devs and basically motivate them to start working on Amoveo
16:24
if you want get to the next level you have to involve other people, amoveo itself isn't attracting enough by themself, so we need more exposure and rewards to get people on boards
16:24
will there be fails? of course, but in the end we might end up with an usable MVP
Z
19:35
Zack
Im looking into putting price charts on the contract_explorer.html page
19:35
it is a little complicated, because each contract can have many markets connected to it
19:36
im thinking to just get us started, I should focus on the market between the 2 kinds of shares in the same contract
19:37
the default tool will keep all the markets displaying the same price. and if someone tries to manipulate one of the markets, they will be creating arbitrage opportunities. so it should heal itself fairly quickly.
19:39
but if we also want a liquidity history chart, we will probably need to account for the other markets as well.
I
Z
19:42
Zack
In reply to this message
if the price of a market is very far from 50-50, like if it is 90-10, then it is more efficient to have your liquidity shares in a market between veo and the more valuable side, instead of having liquidity shares in the market between the 2 types of subcurrency.
19:44
Imagine if the price is 10-90, and you want to convert veo to liquidity shares, but you also don't want to change the price of the market, and you also don't want to end up holding a bunch of subcurrency in either side.
if the only market was one between the 2 kinds of subcurrency, it would be impossible to achieve all your goals.
if you deposited veo into the market, you would end up holding the same number of each kind of subcurrency.
if you bought liquidity shares, you would end up spending all of the less valuable kind of subcurrency, and be left holding lots of the more valuable kind.
if you try to trade the more valuable kind for the less valuable kind, so you only hold liquidity shares, then you end up moving the price in the market.
if the only market was one between the 2 kinds of subcurrency, it would be impossible to achieve all your goals.
if you deposited veo into the market, you would end up holding the same number of each kind of subcurrency.
if you bought liquidity shares, you would end up spending all of the less valuable kind of subcurrency, and be left holding lots of the more valuable kind.
if you try to trade the more valuable kind for the less valuable kind, so you only hold liquidity shares, then you end up moving the price in the market.
19:45
by default a new contract creates 3 markets.
1 between the 2 subcurrencies.
1 between the first subcurrency and veo.
1 between the second subcurrency and veo.
it is also possible to make more markets between any pair of currencies.
the swap tool is set up to use all available markets to calculate the best possible price for the trade you want to make.
1 between the 2 subcurrencies.
1 between the first subcurrency and veo.
1 between the second subcurrency and veo.
it is also possible to make more markets between any pair of currencies.
the swap tool is set up to use all available markets to calculate the best possible price for the trade you want to make.
I
19:46
Instinct
Understood, good design
19:47
In reply to this message
Yep I think this is the best option, that’s the price people will want to see & then know they can add to either side separately
Z
19:48
Zack
does it make more sense to have the price charts on the market_explorer.html page then?
19:48
maybe we want charts on both
19:52
Having charts on the market page would have a nice advantage.
for futarchy contracts, we can make 1 more market to compare the expected consequences of our decision.
Then we would have a page showing a price chart of the expected consequence of our decision according to futarchy.
for futarchy contracts, we can make 1 more market to compare the expected consequences of our decision.
Then we would have a page showing a price chart of the expected consequence of our decision according to futarchy.
1 January 2021
Z
06:33
Zack
I started making the historical price chart
http://159.89.87.58:8080/market_explorer.html?mid=XohY+Qg9myFT88VehQHMwXyUFh129gnKz/T+BAgfS/g=
http://159.89.87.58:8080/market_explorer.html?mid=XohY+Qg9myFT88VehQHMwXyUFh129gnKz/T+BAgfS/g=
06:33
most contracts don't have enough trading to graph anything.
06:47
http://159.89.87.58:8080/market_explorer.html?mid=r2Zu2OiGrT9vOwZHDtHF5G2wfeKir/uP/zB+beao8hY=
Here is the chart for the XRP market
Here is the chart for the XRP market
06:48
it still needs dates and price info, and some way to indicate the amount of liquidity that existed at each point in time
06:48
I wonder what is the best way to indicate liquidity?
Maybe have the color change darker/lighter?
a second chart?
I could draw a line on top of this chart?
Maybe have the color change darker/lighter?
a second chart?
I could draw a line on top of this chart?
06:49
im thinking if the liquidity is too low, we should just cut that info off the chart.
So we only chart the region with sufficient liquidity to have a meaningful price.
So we only chart the region with sufficient liquidity to have a meaningful price.
b
s
Z
19:10
Zack
http://159.89.87.58:8080/market_explorer.html?mid=XohY+Qg9myFT88VehQHMwXyUFh129gnKz/T+BAgfS/g=
I did more progress on the graph.
now it has some numbers, and it is graphing liquidity as well.
I did more progress on the graph.
now it has some numbers, and it is graphing liquidity as well.
19:10
it is more accurate too.
19:20
http://159.89.87.58:8080/market_explorer.html?mid=6LxMVydAuMa0gquIGHfsRhmhZ6KHOQ0AmWOpF4iKb7w=
Something is wrong.
this market has more subcurrency than veo.
Either it is displaying incorrectly, or else there is a serious arbitrage opportunity.
And how come the other market keeps having trades and this one does not?
Something is wrong.
this market has more subcurrency than veo.
Either it is displaying incorrectly, or else there is a serious arbitrage opportunity.
And how come the other market keeps having trades and this one does not?
Z
21:22
Zack
looks like the new laplace strategy in the swap tab is making it so this market is never used.
21:33
I found a fix.
in the swap ai, there was a test to see if the current guess of how to trade is good enough. and if so, it stops trying to improve.
but for the new laplacian method, this test succeeded too soon, before we actually found the best possible trade.
So I turned this optimization off for now, and it seemed to fix the issue.
in the swap ai, there was a test to see if the current guess of how to trade is good enough. and if so, it stops trying to improve.
but for the new laplacian method, this test succeeded too soon, before we actually found the best possible trade.
So I turned this optimization off for now, and it seemed to fix the issue.
2 January 2021
Z
05:17
Zack
http://159.89.87.58:8080/contract_explorer.html?cid=jz5+NialTR/Ktymo2jb7PBDdVIGcSKVXwVw0bA6dEqY=
The graphs are getting better.
The graphs are getting better.
05:17
I still need to convert the block heights into dates.
05:18
units for liquidity don't really make sense on the contract_explorer page. for liquidity it is graphing data related to one of the 3 markets.
05:19
I think for liquidity I need to request data from the 3 markets, and put them together to get the total liquidity, and graph that.
Z
07:23
Zack
http://159.89.87.58:8080/contract_explorer.html?cid=jz5+NialTR/Ktymo2jb7PBDdVIGcSKVXwVw0bA6dEqY=
I changed the block heights into dates on this chart.
I changed the block heights into dates on this chart.
S
19:06
Sebsebzen
Exan tech make over soon?
s
R
21:12
Rom
The volume is low, the number of members on Telegram and Twitter is starting to decrease, what is the strategy to get the project adopted ??? What is the point of monetizing this project if there is no commercial strategy, we must meet the demand of people who have invested is the least we can do !
JS
x
22:33
x
don't worry i will shill it
R
23:25
It's hopeless, I'm not here for FUD, contrary I would rather be enthusiastic especially now with the BTC💥 ... I will wait a few more days if nothing changes I am reselling all my VEO even at a loss
3 January 2021
x
00:55
x
maybe you are right
00:55
it's not incorrect
Z
02:00
Zack
If Amoveo is to succeed, it seems like the next critical thing we need is some small group of regular users.
Some people who are willing to put up with the volatile illiquid currency, because Amoveo is delivering something they can't get elsewhere.
Once we have some kind of user-base, then we will have something that we can pay attention to and grow.
Things are steadily improving. Every new feature and interface is making it that much easier for the first users to get started.
The recent features have been more focused on futarchy type applications, because I feel like there is no good futarchy tool right now, and that futarchy markets have the potential to bring a lot of attention.
Some people who are willing to put up with the volatile illiquid currency, because Amoveo is delivering something they can't get elsewhere.
Once we have some kind of user-base, then we will have something that we can pay attention to and grow.
Things are steadily improving. Every new feature and interface is making it that much easier for the first users to get started.
The recent features have been more focused on futarchy type applications, because I feel like there is no good futarchy tool right now, and that futarchy markets have the potential to bring a lot of attention.
Z
05:00
Zack
http://159.89.87.58:8080/contract_explorer.html?cid=pG4Byp2kBbxzaegWZPN5OsBc75+FedNOSufqDXXn074=
I set up the graph to only display the region of time with high liquidity.
I set up the graph to only display the region of time with high liquidity.
05:09
As far as getting our first small group of users.
It might be a good idea to go after people already involved in cryptocurrency. They are more comfortable downloading wallets and holding altcoins.
Cryptocurrency projects have a lot of controversy. They are bad at making decisions.
We should use futarchy markets to show them what the correct decisions are.
For example, the drivechain community wants to activate BIP 300 and 301 on bitcoin.
We can make some futarchy markets to show how these updates will impact the price of bitcoin.
Especially in regards to hard updates, futarchy is needed.
Ethereum is going through a lot of controversial updates, we can make markets for those.
On the other hand, other cryptocurrency projects might see us as rivals, and refuse to try out Amoveo for that reason.
Another way to get started would be to target the handful of people who already own VEO in light wallets.
They have the lowest barrier of entry to becoming our first users.
What kinds of markets would they actually use?
It might be a good idea to go after people already involved in cryptocurrency. They are more comfortable downloading wallets and holding altcoins.
Cryptocurrency projects have a lot of controversy. They are bad at making decisions.
We should use futarchy markets to show them what the correct decisions are.
For example, the drivechain community wants to activate BIP 300 and 301 on bitcoin.
We can make some futarchy markets to show how these updates will impact the price of bitcoin.
Especially in regards to hard updates, futarchy is needed.
Ethereum is going through a lot of controversial updates, we can make markets for those.
On the other hand, other cryptocurrency projects might see us as rivals, and refuse to try out Amoveo for that reason.
Another way to get started would be to target the handful of people who already own VEO in light wallets.
They have the lowest barrier of entry to becoming our first users.
What kinds of markets would they actually use?
BJ
05:28
Bear Jackson
haha I left my damn coins on a1 exchange and theyre gone now
Z
05:29
Zack
In reply to this message
If you contact them, I think you can still get your coins back.
They left contact info on the discord.
They left contact info on the discord.
BJ
Z
05:30
looks like the webpage says something similar https://a1.exchange/
TG
05:32
Toby Ganger
the good news is that when you’re down 99% of the value of your coins…and then you lose your coins…you’ve only lost 1% of your initial value! sigh
BJ
05:33
Bear Jackson
@zack_amoveo nice! I just emailed them. I had left them from when it went from amoveo.exchange to a1.exchange, then they just disappeared. I was trying to get in touch with them for a while, but was never able. Now, looks like there may be hope!
05:33
@FreeMachine yeah lol happened many times
mx
09:26
mr x
My top feature requests for wallet.html: autosync, autorefresh all balances, remember keys, remember latest sync
Z
09:30
Zack
I think it does autorefresh balances
09:30
is there some situation you know about when it does not?
mx
09:30
mr x
hmm or autorefresh market prices etc
Z
09:30
Zack
remembering keys is tough, because security is a higher priority than usability.
but remembering last sync and auto-sync should be doable
but remembering last sync and auto-sync should be doable
mx
09:31
mr x
yeah
Z
mx
09:31
mr x
maybe batch request for balances
Z
mx
09:32
mr x
it asks for balance for each currency separately?
09:33
no biggie
Z
09:34
Zack
oh I get it. make a way to do one http request that is like, multiple of the current api requests. that would be faster
mx
09:34
mr x
yeah
09:53
risks of informal twitter bets
JS
12:30
Jon Snow
Here is a question
12:31
Is qtrade the single point of failure of Amoveo?
12:31
If they delist veo, are we going back to the dark stone age of OTC trading?
x
12:32
x
no
OK
12:49
O K
The OTC days were the first glory days.
TG
13:34
Toby Ganger
The OTC days had prices 15-20 times higher than now
Z
16:56
Zack
looks like the hard update to turn off channels failed.
It made the network crash.
But we have a fix, and are resyncing now.
block production should be back in 20 minutes or so.
It made the network crash.
But we have a fix, and are resyncing now.
block production should be back in 20 minutes or so.
borisko_ invited borisko_
b
17:40
borisko_
I m selling veo on qtrade.io cheaper than market price but its already more than a week noone buying it. What can be wrong ?
Z
18:31
Zack
im thinking the next time we try this hard update, we should do it like this:
1) soft fork to turn off the channel txs that are no longer valid.
2) I use normal spend tx to give part of the funds I received from the developer reward to refund people who had money in channels.
3) a hard update to make one block's developer reward bigger, to refund me for paying people for those channel txs.
1) soft fork to turn off the channel txs that are no longer valid.
2) I use normal spend tx to give part of the funds I received from the developer reward to refund people who had money in channels.
3) a hard update to make one block's developer reward bigger, to refund me for paying people for those channel txs.
J
20:53
JOHNwick3's dog
Did you delete my message?
OK
Z
23:05
Zack
https://github.com/zack-bitcoin/amoveo-docs/blob/master/failure_reports/3_1_2021.md
our first failure report in over a year.
our first failure report in over a year.
x
23:28
x
In reply to this message
there's an 1000+veo buy order at usd market price, which is 43000sats
Z
23:38
Zack
I set up the light node to remember the top header in localstorage, so you don't need to resync headers if you reload the web page.
23:40
In reply to this message
Johnwick's dog asks if we need a lot of liquidity to make amoveo contracts possible.
No, we don't.
You just need someone to take the other side of your contract.
No, we don't.
You just need someone to take the other side of your contract.
23:41
He asks if it is possible to use an AI program to overcome margin calls.
No. if we hit a margin, and the value of your asset is zero, then there is nothing for an AI to do. you have nothing left.
No. if we hit a margin, and the value of your asset is zero, then there is nothing for an AI to do. you have nothing left.
Deleted invited Deleted Account
Z
23:44
Zack
He asks about combining financial contracts to create more powerful tools.
I recommend looking into derivatives. There are classes on youtube.
This kind of technology is not new at all. You can learn all about what is possible and not possible.
A lot of people like to talk like blockchains have somehow re-written the rules of the game, and all sorts of new tech is possible.
but really, almost nothing has changed.
blockchains just get rid of the central trusted enforcer.
financial contracts work the same as they always have.
I recommend looking into derivatives. There are classes on youtube.
This kind of technology is not new at all. You can learn all about what is possible and not possible.
A lot of people like to talk like blockchains have somehow re-written the rules of the game, and all sorts of new tech is possible.
but really, almost nothing has changed.
blockchains just get rid of the central trusted enforcer.
financial contracts work the same as they always have.
23:45
He asks what Amoveo is capable of.
I recommend checking out the white paper, and then looking into derivatives more.
Finance is about trading risks. Derivatives are the contracts that are used to trade risk.
I recommend checking out the white paper, and then looking into derivatives more.
Finance is about trading risks. Derivatives are the contracts that are used to trade risk.
4 January 2021
J
00:10
JOHNwick3's dog
ok thank you.
Z
04:49
Zack
Because of the way difficulty is adjusting so slowly, the number of block rewards skipped is currently only 10.
It looks like it is going to get near to zero.
It looks like it is going to get near to zero.
04:50
If all the mining pools get shut off at once, and we recover relatively quickly, it looks like the value of block rewards lost is practically none.
05:00
I set up syncing headers in the light node to be automatic, and I got rid of the button to sync headers.
Z
06:01
Zack
http://159.89.87.58:8080/contract_explorer.html?cid=jz5+NialTR/Ktymo2jb7PBDdVIGcSKVXwVw0bA6dEqY=
I added a couple links to this page. to load up the swap tab with bet information pre-loaded.
I added a couple links to this page. to load up the swap tab with bet information pre-loaded.
mx
DavidF invited DavidF
mx
10:53
mr x
I chose to buy 2 veo of liquidity shares but it actually spent 3 🤔
Z
16:42
Zack
In reply to this message
Did you use the pool tab?
Why do you think it spent 3?
Can you link me to the tx, or block with the tx? Or at least your pubkey?
Why do you think it spent 3?
Can you link me to the tx, or block with the tx? Or at least your pubkey?
mx
16:44
mr x
BOnCHOqw/vE9qHuJi5iPJUg50ApD7YMO/3L3HR6pdecojRywvFm76W3wrUfAHwzCCGqxNR9MgZBjZ54jKnqCY9s=
16:44
pubkey
16:44
i put 2 in the amount to spend field
Z
16:45
Zack
Did your balance decrease by 3?
16:45
2 units of liquidity shares is not the same as 2 units of veo
mx
16:46
mr x
hmm
16:47
all right amount to spend = amount to receive?
16:47
uhh
Z
16:47
Zack
If there is lots of trading in the markets for this contract, the value of your liquidity shares, in veo terms, would increase.
16:47
So 1 liquidity share cannot be the same as 1 veo
mx
16:48
mr x
i sold veo to get liquidity shares
16:48
i put 2 in the amount to spend field
Z
16:48
Zack
In the pool tab, right?
Or on contracts.html?
Or on contracts.html?
mx
16:48
mr x
pool tab
Z
16:49
Zack
Ok
Ill double check this, and look up your tx too
Ill double check this, and look up your tx too
mx
16:49
mr x
but now it seems to give correct values when i lookup price
Z
16:50
Zack
Maybe the pool tab should have a confirmation where it displays all the amounts and you click one more button.
mx
16:50
mr x
it said i would lose 3 units of source
16:50
i just put 2 in the amount field
Z
16:51
Zack
Ok, there is a bug somewhere then
mx
16:51
mr x
yeah maybe
16:51
cant replicate it now
Z
18:24
Zack
im looking into mr x's tx now.
looks like it is in block 146348.
looks like it is in block 146348.
18:26
lets look at the parts that can involve veo.
{market_swap_tx,0,0,0,
<<232,188,76,87,39,64,184,198,180,130,
171,136,24,119,236,70,...>>,
14883384,26549003,1,
<<0,0,0,0,0,0,0,0,0,0,0,0,...>>,
0,
<<143,62,126,54,38,165,77,31,202,183,...>>,
1},
{market_swap_tx,0,0,0,
<<214,69,132,214,127,234,195,204,148,96,
144,24,130,180,45,...>>,
100,297,1,
<<0,0,0,0,0,0,0,0,0,0,0,...>>,
0,
<<143,62,126,54,38,165,77,31,202,...>>,
2},
{contract_use_tx,0,0,0,
<<143,62,126,54,38,165,77,31,202,183,41,
168,218,...>>,
87873150,2,
<<0,0,0,0,0,0,0,0,0,0,...>>,
0},
{contract_use_tx,0,0,0,
<<143,62,126,54,38,165,77,31,202,183,41,
168,...>>,
166216165,2,
<<0,0,0,0,0,0,0,0,0,...>>,
0}
18:29
The first swap is turning 0.149 veo into subcurrency.
the next is turning 0.000001 veo into subcurrency.
the first contract_use is turning 0.879 veo into subcurrency.
the second is turning 1.662 veo into subcurrency.
all together, it looks like this is spending around 2.69 VEO.
the next is turning 0.000001 veo into subcurrency.
the first contract_use is turning 0.879 veo into subcurrency.
the second is turning 1.662 veo into subcurrency.
all together, it looks like this is spending around 2.69 VEO.
18:31
Now looking at the parts that are buying liquidity shares.
18:32
{market_liquidity_tx,0,0,0,
<<94,136,88,249,8,61,155,33,83,243,197,
...>>,
134853097,
<<143,62,126,54,38,165,77,31,202,...>>,
1,
<<143,62,126,54,38,165,77,...>>,
2},
{market_liquidity_tx,0,0,0,
<<232,188,76,87,39,64,184,198,180,130,...>>,
43536636,
<<0,0,0,0,0,0,0,0,...>>,
0,
<<143,62,126,54,38,165,...>>,
1},
18:32
1.35 + 0.435 is about 1.78 units of liquidity shares.
18:33
which is a lot nearer to 2 than the amount of veo.
18:34
next im trying to build a similar tx in the pool tab.
18:35
when I try to spend 2 veo, the message is
you can sell 2.0012047635910126 of the source currency to gain liquidity shares that are currently worth:which looks like what I want to do
0.3378586635910126 of source currency.
1.6621413231031354 of subcurrency type 1.
1.6621413273453902 of subcurrency type 2.
mx
18:35
mr x
yes
Z
18:35
Zack
Now lets look at the tx that it actually builds
mx
18:36
mr x
my tx seemed to also change the price hmm?
Z
18:36
Zack
In reply to this message
oh, I think I under-counted the veo here. it is not 2.69.
because one of the liquidity txs is spending some veo
because one of the liquidity txs is spending some veo
18:37
{market_liquidity_tx,0,0,0,
<<232,188,76,87,39,64,184,198,180,130,...>>,
43536636,
<<0,0,0,0,0,0,0,0,...>>,
0,
<<143,62,126,54,38,165,...>>,
1},this tx spent some veo
mx
18:39
mr x
not too much
Z
18:42
Zack
In reply to this message
here is the info about this market from contracts.html
market id: 6LxMVydAuMa0gquIGHfsRhmhZ6KHOQ0AmWOpF4iKb7w=seems like 0.43 shares is going to cost 0.43 * 2.46 / 3.17 = 0.33 veo
currency 1: veo
amount 1: 246235231
cid 2: jz5+NialTR/Ktymo2jb7PBDdVIGcSKVXwVw0bA6dEqY=
type 2: 1
amount 2: 409344382
shares: 317319024
18:42
that brings you up to 3 veo, as you were saying before
mx
18:43
mr x
yup
Z
18:44
Zack
In reply to this message
notice that the prices moved opposite directions in the 2 markets.
I think there was an arbitrage opportunity that you profited from.
I took most of the arbitrage when I fixed that error related to the new laplacian strategy for swapping, but there was some left, and you took it.
I think there was an arbitrage opportunity that you profited from.
I took most of the arbitrage when I fixed that error related to the new laplacian strategy for swapping, but there was some left, and you took it.
mx
18:45
mr x
Cool if true
Z
18:45
Zack
ok, now back to my test. the confirmation message says I should be spending 2 veo, lets see if the tx it generated matches.
mx
18:46
mr x
I fixed mispricing because previous bug and now it works as intended?
Z
18:48
Zack
The parts that spend veo:
I am losing 0.435 * 2.46 / 3.17 from the liquidity tx, 0.33. same as in your case.
Which adds up to 2 veo.
So it seems like it is working for me.
Is there any possibility that you had typed a 3 accidentally instead of a 2?
["contract_use_tx",0,0,0,"jz5+NialTR/Ktymo2jb7PBDdVIGcSKVXwVw0bA6dEqY=",166214134,2,"AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=",0],looks like I lose 1.66 veo from the contract_use_tx.
["market_liquidity_tx",0,0,0,"6LxMVydAuMa0gquIGHfsRhmhZ6KHOQ0AmWOpF4iKb7w=",43535710,"AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=",0,"jz5+NialTR/Ktymo2jb7PBDdVIGcSKVXwVw0bA6dEqY=",1],
I am losing 0.435 * 2.46 / 3.17 from the liquidity tx, 0.33. same as in your case.
Which adds up to 2 veo.
So it seems like it is working for me.
Is there any possibility that you had typed a 3 accidentally instead of a 2?
mx
18:49
mr x
maybe
18:55
i just feel like i tried typing 1 at one point and for it showed i would lose 2 veo
18:55
then typed 2 and it showed i would lose 3
18:55
oh well
19:04
someone made new contract about twitter bot creation? what does it show if you try to buy liquidity shares?
19:05
i have pending tx and now it shows 3 if i put 2
Z
19:06
Zack
In reply to this message
the explorer is showing this contract has a negative price, haha
mx
19:07
mr x
🤔 .
Z
19:07
Zack
that happens if the amount of veo in a market is bigger than the amount of subcurrency on the other side
19:08
someone else pointed this out a few hours ago.
it seems like if you create a binary market with a very low probability, it is initializing the VEO<>false market very incorrectly.
it seems like if you create a binary market with a very low probability, it is initializing the VEO<>false market very incorrectly.
19:08
so there is a big arbitrage opportunity in this market
19:09
since you can combine the 2 subcurrency types into veo, it is impossible that either would have a price bigger than the price of veo.
constant product markets are supposed to have the same value of currency locked on each side. so this market is saying that veo is worth less than the false-type-subcurrency, which is impossible.
so that is why it is calculating an impossible negative price.
constant product markets are supposed to have the same value of currency locked on each side. so this market is saying that veo is worth less than the false-type-subcurrency, which is impossible.
so that is why it is calculating an impossible negative price.
19:09
ill get the binary creation tool fixed soon.
mx
19:10
mr x
alright
Z
19:24
Zack
I confirm the 2->3 issue you had described
"you can sell 2.9969976710191153 of the source currency to gain liquidity shares that are currently worth:
0.8888867510191157 of source currency.
1.1111132108572026 of subcurrency type 1.
1.1111132318741153 of subcurrency type 2."
"you can sell 2.9969976710191153 of the source currency to gain liquidity shares that are currently worth:
0.8888867510191157 of source currency.
1.1111132108572026 of subcurrency type 1.
1.1111132318741153 of subcurrency type 2."
19:25
and
0.1 -> 1.1
0.1 -> 1.1
19:28
I think this is related to the arbitrage opportunity, because I can see that it is actually spending 2.1 veo, and then it takes the arbitrage to get a refund of 1 veo, to be left spending 1.1
19:28
so maybe the arbitrage refund is being double-counted
Z
21:28
Zack
the arbitrage was double counted, I fixed it.
now im trying to find out why the market price starts wrong in the veo<>false market
now im trying to find out why the market price starts wrong in the veo<>false market
Z
21:49
Zack
I tested the binary creation tool. looks like the markets are starting with the correct prices.
I found that some markets explorer pages were displaying the wrong price and fixed that.
but some markets are still having a negative price, so I investigated more, and I found that a swap tx was created that moved the price negative.
I found that some markets explorer pages were displaying the wrong price and fixed that.
but some markets are still having a negative price, so I investigated more, and I found that a swap tx was created that moved the price negative.
21:50
in block 146413
{market_swap_tx,0,0,0,
<<60,85,110,145,87,251,133,145,53,146,248,189,24,102,...>>,
100000000,76475057,1,
<<0,0,0,0,0,0,0,0,0,0,...>>,
0,
<<134,232,233,132,181,81,221,172,...>>,
2},
21:51
it is a part of a multi-tx, so I guess it had to have been made from the wallet.html swap tab.
22:04
In reply to this message
I can't reproduce this tx at all.
Maybe someone is using an old version of the light node that has a bug that was already fixed?
Maybe someone is using an old version of the light node that has a bug that was already fixed?
22:05
it reminds me of the bug with the laplacian that we had fixed a day or two ago.
Z
22:42
Zack
I found out that some of the markets in the explorer were being stored in reverse, and that was causing the wrong prices on some pages. so I fixed that.
Z
23:15
Zack
im thinking it is really high priority to combine the order book with the AMM.
That is going to give us much better liquidity.
More efficient liquidity and lower fees are our big advantages currently.
The way we can use flash loans to combine depositing money into contracts with doing swaps in the AMMs, it gives us 2x more liquidity for the amount of money deposited into AMM.
And the way we can utilize different AMM in a single contract, I think that is giving us another factor of around 3 or 4x better liquidity for the amount of money locked up.
That is going to give us much better liquidity.
More efficient liquidity and lower fees are our big advantages currently.
The way we can use flash loans to combine depositing money into contracts with doing swaps in the AMMs, it gives us 2x more liquidity for the amount of money deposited into AMM.
And the way we can utilize different AMM in a single contract, I think that is giving us another factor of around 3 or 4x better liquidity for the amount of money locked up.
23:16
comining AMM with order books will give us much deeper liquidity. and putting it all in flash loans will give a great user experience to use them both.
x
23:39
x
https://twitter.com/SGBarbour/status/1345937337704873985?s=20 I'm surprised about the comments under this tweet , very interesting,
5 January 2021
Axopoa invited Axopoa
Z
06:42
Zack
im trying to withdraw my winnings from an expired contract.
here is the contract PX/VkaUaTJfHk7VWm0jordTlJ1MTZQJ2jczoyrXf5LY=
here is the oracle
f42a/1Q9+y/F9IY6xyjA1ER+38K7tBPPA9KMsrfDnHY=
but it seems like the "scalar contract winnings" tool in contracts.html is no longer working.
here is the contract PX/VkaUaTJfHk7VWm0jordTlJ1MTZQJ2jczoyrXf5LY=
here is the oracle
f42a/1Q9+y/F9IY6xyjA1ER+38K7tBPPA9KMsrfDnHY=
but it seems like the "scalar contract winnings" tool in contracts.html is no longer working.
06:43
I was able to successfully close the oracle
06:57
oh, maybe I need to "resolve scalar contract" first
07:03
looks like if I try to do "resolve scalar contract" with bad data, it publishes the tx and charges me the fee anyway.
Z
07:31
Zack
I disassembled the contract and evidence being generated from the light node.
It seems like the evidence value isn't being read from the UX, instead it is hard-coded as 1/2.
oops
It seems like the evidence value isn't being read from the UX, instead it is hard-coded as 1/2.
oops
07:38
I fixed that bug, and was able to get the contract resolved, but the interface is really bad. I think ill make a simpler version that doesn't require copying and pasting such complicated info.
EA
07:43
Eric Arsenault
light node is looking really good
Z
07:43
Zack
thanks
EA
07:44
Eric Arsenault
We just need some lipstick to make it a bit more usable, but most of the functionality seems there
EA
08:08
Eric Arsenault
maybe you should pin a msg @zack_amoveo (with link to light node, github, discord, etc)
Z
08:10
Zack
In reply to this message
github links to everything else. maybe we should reorganize the README for the github
08:10
maybe you want to do a pull request for that?
or you can leave suggestions here.
or you can leave suggestions here.
x
08:22
x
In reply to this message
it can be as popular as satoshidice was https://www.buttcoinfoundation.org/satoshidice-is-killing-bitcoin/
08:25
satoshidice seems to have contributed to most of bitcoins volume in the early days. I've no idea how it got popular, but people like that kind of gambling. prediction market is a better form of gambling
Z
08:31
Zack
I made a simpler interface for resolving scalar contracts.
08:33
I think merging the AMM and the order book with a single UX could be a killer feature. I think that is what I will focus on next.
M
08:34
Minieep21
woah didn't realise mining became so centralised :/
08:35
veopool top miner is 95%+ of hashrate
Z
09:12
Zack
I was able to get my money out of the smart contract. I fixed a bug in the claim winnings tool.
09:13
In reply to this message
whoever else has money in this smart contract, now they can just claim their winnings without doing the other steps.
Deleted invited Deleted Account
Deleted invited Deleted Account
Z
17:41
Zack
im looking at the swap AI, and it seems like the actual AI is fast.
The problem is that it is doing a ton of http requests to re-download the same data over and over.
I think if I just store a copy, and read the copy instead of re-downloading when possible, it can be fast.
The problem is that it is doing a ton of http requests to re-download the same data over and over.
I think if I just store a copy, and read the copy instead of re-downloading when possible, it can be fast.
17:42
I think im going to do this before combining with the order book, just so testing the order book wont be so slow.
YiJun invited YiJun
Z
20:44
Zack
the swap tool is much faster.
http://159.89.87.58:8080/wallet.html
And after the first time you look up a price, it memorizes the relevant consensus state, like the contracts and markets and stuff.
So the next time you look up a price, it is instantaneous.
http://159.89.87.58:8080/wallet.html
And after the first time you look up a price, it memorizes the relevant consensus state, like the contracts and markets and stuff.
So the next time you look up a price, it is instantaneous.
mx
20:52
mr x
nice
Z
20:59
Zack
I am thinking about how we can combine the AMM and limit order tools.
It seems like giving an option to make your trade as a limit order instead is going to be the easy part.
So im trying to do the other first.
Matching limit orders in a multi-tx along with making trades in the AMM.
Starting with a simple example.
Lets say you have currency A, and you want currency C. and one of the paths you are taking uses market A->B, and market B->C.
The paths we calculate in the AMM AI, they are set up to leave you with zero excess B at the end.
Lets say there is a limit order you can match, which allows you to swap A for B at a better price.
But if we include this limit order, that means you are starting out with some A, as well as some B, and you want to change them both into C.
So do we run the AMM AI twice? once from A->C, and once from B->C, then we compress all the resulting txs into a single big multi-tx?
or what if the limit order, it requires you to spend more than all the A you had wanted to spend, but it is at a great price, giving you lots of B.
So in that case you would want to use the AMM to transform some of the B back into A, and some of the B into C.
It seems like giving an option to make your trade as a limit order instead is going to be the easy part.
So im trying to do the other first.
Matching limit orders in a multi-tx along with making trades in the AMM.
Starting with a simple example.
Lets say you have currency A, and you want currency C. and one of the paths you are taking uses market A->B, and market B->C.
The paths we calculate in the AMM AI, they are set up to leave you with zero excess B at the end.
Lets say there is a limit order you can match, which allows you to swap A for B at a better price.
But if we include this limit order, that means you are starting out with some A, as well as some B, and you want to change them both into C.
So do we run the AMM AI twice? once from A->C, and once from B->C, then we compress all the resulting txs into a single big multi-tx?
or what if the limit order, it requires you to spend more than all the A you had wanted to spend, but it is at a great price, giving you lots of B.
So in that case you would want to use the AMM to transform some of the B back into A, and some of the B into C.
21:00
In reply to this message
it might be cool if the swap tool automatically updated the price when you changed the amount you want to spend. without having to click "lookup price".
Z
21:20
Zack
So, what is the overall process we want to do for the swap with limit orders?
I am guessing it will go something like this:
1) set up a normal trade with the AMM AI swap tool.
This will generate a multi-tx with many market-swap tx inside of it. each market swap tx is telling us the price we are evaluating for a pair of the currencies.
We can evaluate the relative prices of currencies that don't have direct market connections by multiplying the prices along a multi-step market path.
2) look up all the limit orders between pairs of currencies involved in the multi-tx the AMM AI had generated.
Filter out any limit orders that are at worse prices than the pure AMM solution would pay.
3) Find any arbitrage loops created by your new trade. If there is a way to include one or more limit orders, along with your trade, and you end up net profiting in a currency without losing any other currency, then we should include those limit orders.
4) look at all the limit orders you want to include, and forget about the AMM swap we had planned in step (1). Calculate how many of each subcurrency you are left with if you accept all these limit orders.
For each of these subcurrencies, use the AMM AI swap tool to calculate the optimal way to convert that subcurrency into the currency you want to purchase.
5) compress all the txs generated by all the AMM AI swaps, to make the smallest multi-tx possible. it also matches all the limit orders you had wanted to include.
I am guessing it will go something like this:
1) set up a normal trade with the AMM AI swap tool.
This will generate a multi-tx with many market-swap tx inside of it. each market swap tx is telling us the price we are evaluating for a pair of the currencies.
We can evaluate the relative prices of currencies that don't have direct market connections by multiplying the prices along a multi-step market path.
2) look up all the limit orders between pairs of currencies involved in the multi-tx the AMM AI had generated.
Filter out any limit orders that are at worse prices than the pure AMM solution would pay.
3) Find any arbitrage loops created by your new trade. If there is a way to include one or more limit orders, along with your trade, and you end up net profiting in a currency without losing any other currency, then we should include those limit orders.
4) look at all the limit orders you want to include, and forget about the AMM swap we had planned in step (1). Calculate how many of each subcurrency you are left with if you accept all these limit orders.
For each of these subcurrencies, use the AMM AI swap tool to calculate the optimal way to convert that subcurrency into the currency you want to purchase.
5) compress all the txs generated by all the AMM AI swaps, to make the smallest multi-tx possible. it also matches all the limit orders you had wanted to include.
21:21
I think the difficult part is step (3).
Because there could be many arbitrage loops, and some of them could be contradictory.
like if there are limit orders A, B, and C.
maybe A+B makes one loop, and A+C makes another loop.
How can we know whether A+B or A+C is better?
Because there could be many arbitrage loops, and some of them could be contradictory.
like if there are limit orders A, B, and C.
maybe A+B makes one loop, and A+C makes another loop.
How can we know whether A+B or A+C is better?
21:22
I guess since we have a way of calculating the relative value of subcurrencies from step (1), we could look at how much free money we get from A+B vs A+C, and go with whichever one is giving more free money.
21:23
How exactly do we calculate all these arbitrage loops anyway?
21:30
I guess if a limit order has a better price than we would have paid in the AMM solution, then it necessarily allows for arbitrage loops.
But we need to account for the loss due to slippage in the AMM when we try to complete the loop.
If we can combine multiple arbitrage loops that would have moved some AMM in opposite directions, then the slippage cost for that AMM cancels out.
But we need to account for the loss due to slippage in the AMM when we try to complete the loop.
If we can combine multiple arbitrage loops that would have moved some AMM in opposite directions, then the slippage cost for that AMM cancels out.
21:44
One way to think of it is that we have a field that we are taking a path integral through, and we want to find the minimum distance. and the field has non-zero curl. and we are trying to identify regions that curl in the same direction, so we can loop around that region as a shorter path.
So we divide up the field into big regions that curl one way or the other. Because bordering regions that curl in the same direction can combine constructively.
And for each region, we calculate if the arbitrage profit for a loop around that region exceeds the losses due to slippage in the AMM markets to complete the loop.
If 2 bordering regions are curling opposite directions, and we can profit off either of them, then they interfere destructively. One loop makes the other more expensive.
Maybe it sometimes makes sense to leave some neutral territory between the conflicting regions?
So we divide up the field into big regions that curl one way or the other. Because bordering regions that curl in the same direction can combine constructively.
And for each region, we calculate if the arbitrage profit for a loop around that region exceeds the losses due to slippage in the AMM markets to complete the loop.
If 2 bordering regions are curling opposite directions, and we can profit off either of them, then they interfere destructively. One loop makes the other more expensive.
Maybe it sometimes makes sense to leave some neutral territory between the conflicting regions?
6 January 2021
Z
01:04
Zack
The current amm ai.
It converts one currency into another.
We have precise control over the amount spent, and less control over how much we receive.
I think we need to generalize this tool, so you can spend a precise negative amount in one currency, and receive an amount you have less control over in another.
Basically the same tool, but the markets go backwards. All the values are negative.
Then, i think, it should be easy to combine the amm ai with the order book.
The process would be like this:
1) make a normal swap using the amm ai.
2) identify which limit orders are at better prices than the trade built by the ai.
3) one by one, try including these limit orders in the multi tx. When we add a limit order, it causes a precise excess of one currency, and a precise deficit of another.
We can use the normal amm ai tool to turn the excess into the currency we are buying. We can use the negative amm ai to turn the deficit into negative of what we are buying.
If the net result is that we earb more, then keep that limit order. Otherwise, discard it.
It converts one currency into another.
We have precise control over the amount spent, and less control over how much we receive.
I think we need to generalize this tool, so you can spend a precise negative amount in one currency, and receive an amount you have less control over in another.
Basically the same tool, but the markets go backwards. All the values are negative.
Then, i think, it should be easy to combine the amm ai with the order book.
The process would be like this:
1) make a normal swap using the amm ai.
2) identify which limit orders are at better prices than the trade built by the ai.
3) one by one, try including these limit orders in the multi tx. When we add a limit order, it causes a precise excess of one currency, and a precise deficit of another.
We can use the normal amm ai tool to turn the excess into the currency we are buying. We can use the negative amm ai to turn the deficit into negative of what we are buying.
If the net result is that we earb more, then keep that limit order. Otherwise, discard it.
7 January 2021
Z
08:09
Zack
I am working on the negative version of the swap tool.
I got it working well for trading very small amounts. like, less than 1 veo.
but if I try trading a little more, it breaks.
The swap AI, it calculates how much output we get, given how much we could spend on a path.
And for each step of each path it is solving some quadratic equations based on how to combine market swaps with contract-use-txs to calculate how to push the money through without being left with any subcurrency.
The problem is that when we put larger negative numbers through, the solutions to some of these quadratic equations end up imaginary.
the part inside the square root is negative.
I'm struggling to interpret what that means, and how we can fix it.
I got it working well for trading very small amounts. like, less than 1 veo.
but if I try trading a little more, it breaks.
The swap AI, it calculates how much output we get, given how much we could spend on a path.
And for each step of each path it is solving some quadratic equations based on how to combine market swaps with contract-use-txs to calculate how to push the money through without being left with any subcurrency.
The problem is that when we put larger negative numbers through, the solutions to some of these quadratic equations end up imaginary.
the part inside the square root is negative.
I'm struggling to interpret what that means, and how we can fix it.
08:14
maybe when we are using negative numbers, we should be doing different combinations of markets with contracts?
It seems like these quadratic equations are in pairs. maybe for negative amounts, I need to swap out each pair?
It seems like these quadratic equations are in pairs. maybe for negative amounts, I need to swap out each pair?
08:17
6 pairs
Z
08:59
Zack
oh, I think I see the problem.
You can push as much money as you want through a constant product market, but you can only pull as much as exists on the side that you are pulling from
You can push as much money as you want through a constant product market, but you can only pull as much as exists on the side that you are pulling from
09:02
so maybe I just set the imaginary part of the number to zero?
09:09
that doesn't seem to help.
09:13
so maybe d^2(received)/d(spent)^2 doesn't have sufficient bounds to do gradient descent.
09:15
I guess I need to transform d(received)/d(spent) onto a space s(received) that does have enough bounds, and also has the same minimum, and preserving continuousness and differentiability.
09:24
What if we calculated the most negative value that we could want to push through each path, and if our gradient descent ever takes us out of bounds, we take a step directly back into bounds.
Taking value away from whichever was out of bounds, and distributing among all the parts of the vector that are in-bounds, without causing any of them to go out-of-bounds.
Taking value away from whichever was out of bounds, and distributing among all the parts of the vector that are in-bounds, without causing any of them to go out-of-bounds.
09:25
There is going to be some maximum value you can pull out, just like if you spent infinite money, there is some maximum amount you could push through.
Z
09:46
Zack
So I found a way to test the limits.
I can push a lot of money through the paths, and see how much comes out. The amount that comes out is about the max of how much we could pull through.
So I tested this on the negative set of paths, and it showed that several of the paths had maximums that could cause it to break.
I can push a lot of money through the paths, and see how much comes out. The amount that comes out is about the max of how much we could pull through.
So I tested this on the negative set of paths, and it showed that several of the paths had maximums that could cause it to break.
Z
10:10
Zack
so my strategy for calculating the maxes for each path, it didn't work.
simply pushing a large amount through and looking at what comes out doesn't work, because process_path( is doing math with floating point, not integers. so it can have a fraction of a satoshi on one side of a market.
So my next plan is to start with the market that can output the least. calculate if it loses everything but one satoshi, and follow that amount through the rest of the market to see what comes out.
simply pushing a large amount through and looking at what comes out doesn't work, because process_path( is doing math with floating point, not integers. so it can have a fraction of a satoshi on one side of a market.
So my next plan is to start with the market that can output the least. calculate if it loses everything but one satoshi, and follow that amount through the rest of the market to see what comes out.
Z
22:50
Zack
pushing negative numbers through the AI seems like it really is not going to work.
But the AI is really fast. maybe we should do some kind of optimized binary search to figure out how much positive value to push through to make a precise amount come out the end.
But the AI is really fast. maybe we should do some kind of optimized binary search to figure out how much positive value to push through to make a precise amount come out the end.
22:54
for small intervals the amount you spend and receive are nearly linearly related, so I think we can make the binary search very fast.
And we are calculating the derivative and second derivative anyway, so we could use that information to calculate a second order polynomial which very nearly matches for the interval, and have something a whole lot faster than binary search.
Maybe it would only be 4 or 5 times slower than the positive direction.
And users probably don't care if they end up holding ~1000 satoshis of dust in various subcurrencies.
And we are calculating the derivative and second derivative anyway, so we could use that information to calculate a second order polynomial which very nearly matches for the interval, and have something a whole lot faster than binary search.
Maybe it would only be 4 or 5 times slower than the positive direction.
And users probably don't care if they end up holding ~1000 satoshis of dust in various subcurrencies.
22:56
The AI seems to be converging really fast. I think we can lower it from 90 iterations to like 5 without losing meaningful accuracy.
So adding the optimized binary search on top shouldn't be a problem.
The step I am concerned with is looking at combinations of limit orders we could include.
That step is NP complete, so it is going to take some creativity to make something usable.
I was thinking for that part, maybe users could have a configuration option for how many seconds they want their computer to spend trying to optimize their trade.
For people trading very large amounts, getting a 0.5% better price for their trade can be significant.
So adding the optimized binary search on top shouldn't be a problem.
The step I am concerned with is looking at combinations of limit orders we could include.
That step is NP complete, so it is going to take some creativity to make something usable.
I was thinking for that part, maybe users could have a configuration option for how many seconds they want their computer to spend trying to optimize their trade.
For people trading very large amounts, getting a 0.5% better price for their trade can be significant.
8 January 2021
Deleted invited Deleted Account
T
03:03
Tromp
How can I generate my pubkey from my private key please?
OK
T
03:48
Tromp
Or what wallet I could use to input private key that shows my pub or balance
Z
Z
04:58
Zack
Seems like 12 iterations of binary search gets it accurate to 5 significant figures, and this is using linear interpolation between the bounds, not accounting for the derivative or second derivative.
and it takes much less than a second.
and it takes much less than a second.
Z
05:22
Zack
http://159.89.87.58:8080/wallet.html
I updated this swap tool so you can try it out.
If you try to swap a negative number, you have control over how much you will receive instead of how much you will spend.
after the first time you look up a price it caches the markets and contracts and paths, if you look up a price again it is practically instantaneous.
Even with the 12 iterations of binary search, the delay is unnoticeable.
So now I can start integrating the limit order tool into the AMM swap tool.
I updated this swap tool so you can try it out.
If you try to swap a negative number, you have control over how much you will receive instead of how much you will spend.
after the first time you look up a price it caches the markets and contracts and paths, if you look up a price again it is practically instantaneous.
Even with the 12 iterations of binary search, the delay is unnoticeable.
So now I can start integrating the limit order tool into the AMM swap tool.
mx
11:28
mr x
Cool
Z
17:23
Zack
Im thinking about combining the limit order and amm tools.
It seems like you would be interested in any limit orders that are moving value in the reverse direction of your trade, and at a better price than you are willing to pay.
And you would be interested in any orders that move value in the same direction as your trade, at a worse price than you are willing to pay.
It like, if me and my friends want to buy hamburgers, and i collect money and walk to the store to pick them up for everyone, and i charge slightly more than the cost of a hamburger to each of my friends, then i can pay less for my hamburger, and the average price is still good enough for the restaurant.
In particular, i think we are looking for combinations of buy orders and sell orders, which along with your trade, complete a cycle that earns you arbitrage profit.
It is a little tricky though, because if there is a trade going the opposite direction as you for a very good price, but it is too big for you to match entirely, it can be worth it to include orders going in the same direction as you for better prices than you will pay, so that you have enough volume to match the big order.
So we cant just filter out all the orders in the same direction as you at better prices. We want to also consider the orders that have better prices, within some bound set by the orders we are considering in the opposite direction, how cheap they are compared to what you are willing to pay.
It seems like you would be interested in any limit orders that are moving value in the reverse direction of your trade, and at a better price than you are willing to pay.
And you would be interested in any orders that move value in the same direction as your trade, at a worse price than you are willing to pay.
It like, if me and my friends want to buy hamburgers, and i collect money and walk to the store to pick them up for everyone, and i charge slightly more than the cost of a hamburger to each of my friends, then i can pay less for my hamburger, and the average price is still good enough for the restaurant.
In particular, i think we are looking for combinations of buy orders and sell orders, which along with your trade, complete a cycle that earns you arbitrage profit.
It is a little tricky though, because if there is a trade going the opposite direction as you for a very good price, but it is too big for you to match entirely, it can be worth it to include orders going in the same direction as you for better prices than you will pay, so that you have enough volume to match the big order.
So we cant just filter out all the orders in the same direction as you at better prices. We want to also consider the orders that have better prices, within some bound set by the orders we are considering in the opposite direction, how cheap they are compared to what you are willing to pay.
s
17:27
sanket
Yes this sounds like a good idea.
Do you think people who will use it realise if it's a limit order or an Amm based order
Do you think people who will use it realise if it's a limit order or an Amm based order
Z
17:30
Zack
If we just consider a single market that swaps between 2 currencies, and ignore all the other paths, im pretty sure deciding which limit orders to include is equivalent to the knapsack problem, it is exponential in the number of limit orders you are considering.
But we are considering long n-stwp paths, which can have n^2 different markets. And we are considering many paths, not just one.
So a perfect solution is something like (((path length)^2)/2)*(# limit orders available)^2
And since we are layering it on top of an ai it is worse.
The ai has 15 rounds of gradient descent layered on 12 rounds of binary search. And the gradient descent vector is based on the number of paths, and calculating the gradient is lauered with the path length.
((((15*12* #paths * pathLength)^2)/2)*(# limit orders available)^2
So i think we need to be creative to get this to run fast enough.
But we are considering long n-stwp paths, which can have n^2 different markets. And we are considering many paths, not just one.
So a perfect solution is something like (((path length)^2)/2)*(# limit orders available)^2
And since we are layering it on top of an ai it is worse.
The ai has 15 rounds of gradient descent layered on 12 rounds of binary search. And the gradient descent vector is based on the number of paths, and calculating the gradient is lauered with the path length.
((((15*12* #paths * pathLength)^2)/2)*(# limit orders available)^2
So i think we need to be creative to get this to run fast enough.
17:31
In reply to this message
Im thinking the interface will give you a price to match your trade immediately, and there will be a button to instead create a limit order that defaults at the current market price, and gives a field so you can choose a different limit price.
17:32
I wont bother to tell the user whether their trade is being done purely with amm or if limit orders are involved.
17:38
In reply to this message
Maybe the way to keep this efficient is like how erlang solved the packing problem in their parallelization of threads system.
Their idea is that the packing problem is only difficult when you are packing big things.
If you can cut up the things really small, it becomes like packing your knapsack full of sand.
You dont have to worry about which sand grains you use where, you just pour it in, and it takes up practically all the space.
If we set up the limit order tool so that big limit orders are divided up into many small orders at the same price, then it becomes computationally easy to calculate near-optimal trades.
The cost is that more txs end up on-chain.
But i think if you match multiple identical limit orders in the same multi-tx, you dont need to include any more merkle proofs to make it work. And the different limit orders collapse well if we use compression on the txs and blocks.
The only part that doesn't collapse is the signature.
Maybe we can find a clever solution for one signature to make many limit orders valid.
Their idea is that the packing problem is only difficult when you are packing big things.
If you can cut up the things really small, it becomes like packing your knapsack full of sand.
You dont have to worry about which sand grains you use where, you just pour it in, and it takes up practically all the space.
If we set up the limit order tool so that big limit orders are divided up into many small orders at the same price, then it becomes computationally easy to calculate near-optimal trades.
The cost is that more txs end up on-chain.
But i think if you match multiple identical limit orders in the same multi-tx, you dont need to include any more merkle proofs to make it work. And the different limit orders collapse well if we use compression on the txs and blocks.
The only part that doesn't collapse is the signature.
Maybe we can find a clever solution for one signature to make many limit orders valid.
17:39
Or maybe we can set it up so like, there is one big limit order that increases your nonce by 100. Or 2 half-sized ones that increase your nonce by 50 each. Etc.
So the number of signatures can be kept smaller.
So the number of signatures can be kept smaller.
17:40
Or maybe the signature isnt big enough to matter yet.
17:44
Maybe the limit order can be partially matched, and it has a variable effect on your account nonce that depends on how much of the order was matched.
17:44
So the same limit order can be matched multiple times until it is fully filled
s
9 January 2021
Z
05:37
Zack
im looking at the limit order swap tx type.
It seems like it does not increment the nonce at all, but it does check what your account nonce is. So you can increment your account nonce to cancel an open order.
When an order gets matched, currently we store 32 bytes of trade ID in a merkel tree. once the trade is stored, it can't be matched again.
It seems like it does not increment the nonce at all, but it does check what your account nonce is. So you can increment your account nonce to cancel an open order.
When an order gets matched, currently we store 32 bytes of trade ID in a merkel tree. once the trade is stored, it can't be matched again.
05:39
the trade merkle tree also has some extra space where we are currently storing the block height for when that trade was matched.
We aren't using the block height info for anything.
So I am thinking we can convert that space we are using for a block height into a nonce.
And the nonce increments as more of the trade gets matched.
We aren't using the block height info for anything.
So I am thinking we can convert that space we are using for a block height into a nonce.
And the nonce increments as more of the trade gets matched.
05:41
the trade offer will have some number written on it for how many parts the limit order can be broken into and matched seperately.
So lets say it has 1000 parts.
if you match 10% of this limit order, then the nonce increases by 100.
the nonce can keep incrementing until it reaches 1000.
So lets say it has 1000 parts.
if you match 10% of this limit order, then the nonce increases by 100.
the nonce can keep incrementing until it reaches 1000.
05:42
im thinking I will leave the trade tree as is, and write a new version of the swap offer and swap_tx.
once the new version of swap offer and swap tx are working correctly, we can do a soft fork to turn off the old versions.
once the new version of swap offer and swap tx are working correctly, we can do a soft fork to turn off the old versions.
Deleted invited Deleted Account
Z
09:11
Zack
Maybe we should make a different merkle tree for limit orders, so we can keep a record if more details in the consensus state.
Like the currency types, and quantities being exchanged, and how much of the order was already matched, and expiration date?
Maybe it is better to keep it off chain, so details can be changed without having to create a new entry on-chain?
Like if you wanted to increase the size of your order, or increase the time until expiry.
Like the currency types, and quantities being exchanged, and how much of the order was already matched, and expiration date?
Maybe it is better to keep it off chain, so details can be changed without having to create a new entry on-chain?
Like if you wanted to increase the size of your order, or increase the time until expiry.
09:18
How about if someone used or-logic.
Like, they want to buy X, and they dont care if they spend Y or Z.
So they could make 2 limit orders that share the same trade id. So if one is partially matched, it reduces how much is left to match of both.
Is that something we would want?
Like, they want to buy X, and they dont care if they spend Y or Z.
So they could make 2 limit orders that share the same trade id. So if one is partially matched, it reduces how much is left to match of both.
Is that something we would want?
EA
15:26
In reply to this message
Not sure tbh, maybe it just makes things too complicated. Maybe start simple and eventually get to this point?
15:26
Maybe Amoveo should have a forum so we can discuss Futarchy proposals, etc
Z
16:59
Zack
If we keep the limit order details off chain, you can adjust the price downward off chain.
If no one wants to buy your stable bitcoin for 100 veo, you can change it to 95 veo, without putting anything on chain or waiting for any confirmation.
Adjusting upwards requires canceling your trade and making a new one.
Another feature.
You could reuse the same trade id for more trades.
After one trade is fully matched, the trade id entry in the consensus state could be used for your next.
So you can make more trades without having to pay to use more of the blockchain consensus space.
If no one wants to buy your stable bitcoin for 100 veo, you can change it to 95 veo, without putting anything on chain or waiting for any confirmation.
Adjusting upwards requires canceling your trade and making a new one.
Another feature.
You could reuse the same trade id for more trades.
After one trade is fully matched, the trade id entry in the consensus state could be used for your next.
So you can make more trades without having to pay to use more of the blockchain consensus space.
Z
17:38
Zack
Maybe it is better to keep the on-chain part as simple as we can, since it is costly to store and adjust data there.
We want to maximize how much we can do without having to go on-chain.
We want to maximize how much we can do without having to go on-chain.
Z
17:55
Zack
At some point we need to circle back to the off-chain single-price batch market plan, like we had built out for state channels before realizing state channels weren't going to work.
As long as we are doing everything with on-chain AMM and off-chain limit orders, we are going to be suffering from front running, and miners extracting value from traders. And it is really hard to 10x the value proposition of Ethereum when our tools are so similar.
Limit orders are necessary infrastructure, because when people want to pull their winnings out of the single-price-batch market without waiting for contract resolution, they will be using limit orders.
AMM are necessary infrastructure so we can have a clear price in consensus state, to enable futarchy and prediction market applications, and for a sanity check that the off-chain single-price-batch tool is working with correct prices.
As long as we are doing everything with on-chain AMM and off-chain limit orders, we are going to be suffering from front running, and miners extracting value from traders. And it is really hard to 10x the value proposition of Ethereum when our tools are so similar.
Limit orders are necessary infrastructure, because when people want to pull their winnings out of the single-price-batch market without waiting for contract resolution, they will be using limit orders.
AMM are necessary infrastructure so we can have a clear price in consensus state, to enable futarchy and prediction market applications, and for a sanity check that the off-chain single-price-batch tool is working with correct prices.
Z
18:44
Zack
What do we have that Ethereum can't copy from us?
Ethereum is committed to the plan of a turing complete world wide computer.
Amoveo has the more specialized goal to be optimized for derivatives.
There are a couple cases where Ethereum's goal contradicts with their ability to be a good derivatives platform, and we can beat them in those places.
* our oracle system is more accurate, and less costly. So, if everything else is equal, we could charge less per trade, and we could do parasite attacks to break any oracles on Ethereum. We integrated the oracle into the consensus mechanism in a way that exceeds the capacity of the sandbox for Ethereum contracts.
* if Amoveo txs fail because the prices moved, or you don't have enough gas, those txs aren't on-chain and you don't pay any fees. Unlike ethereum where you can end up paying for a tx fee and your trade is unexecuted. Ethereum needs to work this way to maintain turing complete flexibility of it's txs. Amoveo has specialized txs that don't contain any turing complete elements. This is especially important for flash-loans/flash-minting, and for minimizing miner extractable value when making trades in the limit order/AMM system.
In ethereum you need to make a trade-off between using a flash loan to prevent miners from front running you, vs avoiding over-loading a single tx to minimize the chance that you end up paying a fee for an un-executed tx. In Amoveo this trade-off doesn't exist. You can make flash loans involve tons of markets to prevent miners from front running, without worrying about paying a fee if it isn't executed.
* Amoveo contracts are more limited in what they do in ways that make it easier to verify security. This makes it easier for developers to build with confidence, and easier for users to verify that developers have built something that wont break. Amoveo contracts only get executed once, and all they do is divide up their collateral between the subcurrencies defined by that contract. Things like re-entry bugs or the DAO hack could never happen.
It is easy to calculate the total on-chain cost of an Amoveo smart contract, so there is no worry that it could grow in complexity and become unaffordable to use.
Because of how Ethereum contracts can have infinite loops and how they can reference each other, Ethereum suffers from a version of the halting problem that makes it impossible to be sure if or how their contracts will break. Security analysis can't provide certainty.
Amoveo contracts are independent and we can write them with only finite-loops to completely sidestep the halting problem, so we can have certainty that a contract does what it is expected to do.
Our odds to win are not very good, but we haven't lost yet. We still have a chance to build the platform for the most popular application of money.
Ethereum is committed to the plan of a turing complete world wide computer.
Amoveo has the more specialized goal to be optimized for derivatives.
There are a couple cases where Ethereum's goal contradicts with their ability to be a good derivatives platform, and we can beat them in those places.
* our oracle system is more accurate, and less costly. So, if everything else is equal, we could charge less per trade, and we could do parasite attacks to break any oracles on Ethereum. We integrated the oracle into the consensus mechanism in a way that exceeds the capacity of the sandbox for Ethereum contracts.
* if Amoveo txs fail because the prices moved, or you don't have enough gas, those txs aren't on-chain and you don't pay any fees. Unlike ethereum where you can end up paying for a tx fee and your trade is unexecuted. Ethereum needs to work this way to maintain turing complete flexibility of it's txs. Amoveo has specialized txs that don't contain any turing complete elements. This is especially important for flash-loans/flash-minting, and for minimizing miner extractable value when making trades in the limit order/AMM system.
In ethereum you need to make a trade-off between using a flash loan to prevent miners from front running you, vs avoiding over-loading a single tx to minimize the chance that you end up paying a fee for an un-executed tx. In Amoveo this trade-off doesn't exist. You can make flash loans involve tons of markets to prevent miners from front running, without worrying about paying a fee if it isn't executed.
* Amoveo contracts are more limited in what they do in ways that make it easier to verify security. This makes it easier for developers to build with confidence, and easier for users to verify that developers have built something that wont break. Amoveo contracts only get executed once, and all they do is divide up their collateral between the subcurrencies defined by that contract. Things like re-entry bugs or the DAO hack could never happen.
It is easy to calculate the total on-chain cost of an Amoveo smart contract, so there is no worry that it could grow in complexity and become unaffordable to use.
Because of how Ethereum contracts can have infinite loops and how they can reference each other, Ethereum suffers from a version of the halting problem that makes it impossible to be sure if or how their contracts will break. Security analysis can't provide certainty.
Amoveo contracts are independent and we can write them with only finite-loops to completely sidestep the halting problem, so we can have certainty that a contract does what it is expected to do.
Our odds to win are not very good, but we haven't lost yet. We still have a chance to build the platform for the most popular application of money.
M
19:01
Minieep21
How would one create a parasitic contract? The efficiency of Amoveo needs to be brought forward to capture wealth from other ecosystems. That would start a self sufficient loop with more users joining in an effort to benefit from its efficiency
19:02
Increase the odds with more users
Z
G
19:16
Goldy
Where VEO is listed currently ?
Z
Z
20:23
Zack
To kick off trading, i think we need to do something to stabilize the price and increase liquidity.
Maybe sometime soon, we should all start making buy and sell orders on qtrade with an agreed upon prices. Maybe $25 and $27.50, something like that.
This will make it possible for people to buy some veo, participate in a bet, and have confidence that they can exit amoveo with their profits.
If we keep it up long enough, the people providing entry and exit would even be profiting off of this service.
Maybe it is easier to link to the btc price instead of usd, since qtrade is a btc/veo market.
Maybe sometime soon, we should all start making buy and sell orders on qtrade with an agreed upon prices. Maybe $25 and $27.50, something like that.
This will make it possible for people to buy some veo, participate in a bet, and have confidence that they can exit amoveo with their profits.
If we keep it up long enough, the people providing entry and exit would even be profiting off of this service.
Maybe it is easier to link to the btc price instead of usd, since qtrade is a btc/veo market.
x
20:35
x
USDC seems good
20:38
possibly , paypal will integrate stablecoins too,, if that happens, it will probably be easier for people to use blockchain apps. I don't know whether that is possible.\
20:39
short term USD price of veo is stable i think
Z
Z
21:22
Zack
Maybe we can do the stabilization program from inside amoveo. Using the order book.
10 January 2021
Z
04:19
Zack
Im thinking that for canceling limit orders, using account nonces is a mistake.
instead we should have a transaction type to increment the trade-nonce.
That way you wont accidentally cancel your limit orders by making some other kinds of transactions.
instead we should have a transaction type to increment the trade-nonce.
That way you wont accidentally cancel your limit orders by making some other kinds of transactions.
Z
09:18
Zack
I wrote the first draft of the new limit order tx types https://github.com/zack-bitcoin/amoveo/blob/master/apps/amoveo_core/src/consensus/txs/test_txs.erl#L2948
Z
17:58
Zack
How about some terrorism prediction markets, to guess what the pro-trump crazies will do next?
> amazon warehouse bombed before Jan 20.
> Policeman in DC killed in riots before Jan 20.
anything else?
> amazon warehouse bombed before Jan 20.
> Policeman in DC killed in riots before Jan 20.
anything else?
ŽM
18:00
You are biased
18:01
I expected more from you as our spiritual aryan leader
Z
18:04
Zack
https://en.wikipedia.org/wiki/Policy_Analysis_Market
Predicting when and where violence will occur can help to prevent it from happening.
Whichever side of the political spectrum a person is on, I think they could get behind a platform that is working to prevent violence.
Predicting when and where violence will occur can help to prevent it from happening.
Whichever side of the political spectrum a person is on, I think they could get behind a platform that is working to prevent violence.
Z
19:30
Zack
I wrote more tests for the new tx types. Now they work in multi-txs.
11 January 2021
G
00:23
Gregory
In reply to this message
you can hire a marketmaker, will do great liquidity if needed
Z
G
01:03
Gregory
Coz you suck at 90% of things?
01:03
To focus on what you do best
Z
01:03
Zack
Is there some kind of skill involved in market making?
B
02:34
Ben
yep, there are companies build around that topic
02:34
like altonomy
02:41
there are diffrent approaches and ways.
L
08:19
L
In reply to this message
It would seem like if we wanted to do this wrapping VEO for ETH and providing liquidity on uniswap or an alternative is a better, less centralized solution.
08:19
Also more stable than a one off event.
08:19
Or incentivise automated market making on VEO.
Z
08:42
Zack
In reply to this message
Why not use a constant product market maker on amoveo?
Instead of a centralized tool like wrapped veo.
It is a lot cheaper than uniswap too.
Instead of a centralized tool like wrapped veo.
It is a lot cheaper than uniswap too.
Z
10:05
Zack
https://github.com/zack-bitcoin/amoveo-docs/blob/master/todo.md
I started writing up plans for the consensus state datastructures for the perpetual stablecoin hard update. it is at the top of the todo list.
I started writing up plans for the consensus state datastructures for the perpetual stablecoin hard update. it is at the top of the todo list.
10:06
so far, it looks like 1 new merkle tree and 7 new tx types
Deleted invited Deleted Account
Deleted invited Deleted Account
Z
17:48
Zack
What if instead of targetting the value of usd, we made a stablecoin that targetted the price of usd, plus 5% interest monthly?
Or 50% interest monthly?
Or 50% interest monthly?
17:49
I guess people would be less willing to hold the long-veo side of the contract. So the stablecoin would decay in value more quickly.
17:51
By the same logic, if there were lots of demand for long-veo contracts, it seems like the stablecoin could even be rising in value relative to the stability target.
17:53
Do these effects mean it wont be a good stablecoin?
18:07
Could we reduce this effect by making a perpetual stablecoin that is collateralized with another perpetual stablecoin?
Z
19:08
Zack
https://github.com/zack-bitcoin/amoveo-docs/blob/master/blog_posts/voting_in_post_futarchy_world.md
I wrote about why we wont use voting any more, once futarchy wins.
I wrote about why we wont use voting any more, once futarchy wins.
12 January 2021
Deleted invited Deleted Account
13 January 2021
Z
07:52
Zack
I think the swap_tx2 hard update is ready. im going to review it one last time in a day or so before we activate.
I got the single-node tests to run like 20x faster.
I got the single-node tests to run like 20x faster.
07:53
Now I am thinking about the perpetual stablecoin update.
I am guessing that the number of blocks that an auction should last for, it should be customized for each stablecoin.
Do we want each kind of auction to last for the same amount of time?
I am guessing that the number of blocks that an auction should last for, it should be customized for each stablecoin.
Do we want each kind of auction to last for the same amount of time?
07:56
timelimit auctions, we have more anticipation for when they will occur. because it is connected to a block height.
Undercollateralization auctions could basically happen at any time, so they are less anticipated.
So that makes me think timelimit auctions should be shorter.
but during a timelimit auction, the perpetual stablecoin's value is still connected to the finite stablecoin.
during an undercollateralization auction, the perpetual stablecoin is connected to the value of veo, which is probably still falling, since it fell enough to become undercollateralized.
So that makes me think that timelimit auctions can be longer.
Undercollateralization auctions could basically happen at any time, so they are less anticipated.
So that makes me think timelimit auctions should be shorter.
but during a timelimit auction, the perpetual stablecoin's value is still connected to the finite stablecoin.
during an undercollateralization auction, the perpetual stablecoin is connected to the value of veo, which is probably still falling, since it fell enough to become undercollateralized.
So that makes me think that timelimit auctions can be longer.
07:56
I guess ill set up the perpetual stablecoin so that both can be set independently
M Bilal invited M Bilal
14 January 2021
riordant invited riordant
r
01:06
riordant
Hi. Is there a roadmap available for the coming year?
Z
01:31
Zack
There is a todo list in the docs repository
01:37
Right now we are about to do a hard update for a better version of limit orders.
When that activates in 2 weeks, the plan is to integrate it with the AMM swap tool in wallet.html, part of the light node.
Also in the works is the perpetual stablecoin update. It is kind of big, it will probably take a few weeks.
After all that, i guess we will reflect on how things are going and make a new plan for what to do next.
Something i am looking forward to is the single price batch market built in channels.
When that activates in 2 weeks, the plan is to integrate it with the AMM swap tool in wallet.html, part of the light node.
Also in the works is the perpetual stablecoin update. It is kind of big, it will probably take a few weeks.
After all that, i guess we will reflect on how things are going and make a new plan for what to do next.
Something i am looking forward to is the single price batch market built in channels.
J
01:38
Josh
How about "There will not be a drone show surrounding the ECB headquarters in Frankfurt by Jan 31, featuring the Bitcoin logo and the words 'buy bitcoin'"?
Z
01:42
Zack
In reply to this message
Hahaha, sounds great.
Im setting up the limit order tool so you can prevent partial matches.
Which would give the drone operators the opportunity to match the full amount.
Im setting up the limit order tool so you can prevent partial matches.
Which would give the drone operators the opportunity to match the full amount.
B
J
01:58
Josh
I love how there's nothing to lose. If it doesn't happen you get your money back. If it does happen that's even cooler.
01:58
We'd have to specify minimum number of drones, light output, height of the show, duration, size of the logo in sq meters.
Z
01:59
Zack
Someone could attempt and fail. Then you would profit
J
01:59
Josh
Yeah
02:03
this might be easier: https://mappingprojector.com/projectors-of-gobos-and-dynamic-effects/
EA
J
04:58
Josh
If this happened it would be awesome marketing for Amoveo.
ᴅɪꜱᴛʙɪᴛ ⛓️💵 invited ᴅɪꜱᴛʙɪᴛ ⛓️💵
Z
06:31
Zack
Distbit asks about the tradeoffs between LMSR and constant product markets, like in uniswap.
a limitation in lmsr is that you can't add more liquidity to an existing market, and you can't remove liquidity without shutting off the entire market. So they are kind of controlled by whoever launched them.
Constant product markets are nice because anyone can add or remove liquidity at any time. So it is more decentralized.
But there are other tradeoffs. lmsr can share liquidity between lots of different things we are betting on in a way that constnat product cannot. and lmsr are a little more accurate for prediction markets.
constant product is nice because the liquidity provider can profit. with lmsr the liquidity provider always takes a loss.
a good explanation of lmsr is here http://blog.oddhead.com/2006/10/30/implementing-hansons-market-maker/
a limitation in lmsr is that you can't add more liquidity to an existing market, and you can't remove liquidity without shutting off the entire market. So they are kind of controlled by whoever launched them.
Constant product markets are nice because anyone can add or remove liquidity at any time. So it is more decentralized.
But there are other tradeoffs. lmsr can share liquidity between lots of different things we are betting on in a way that constnat product cannot. and lmsr are a little more accurate for prediction markets.
constant product is nice because the liquidity provider can profit. with lmsr the liquidity provider always takes a loss.
a good explanation of lmsr is here http://blog.oddhead.com/2006/10/30/implementing-hansons-market-maker/
ᴅ
06:31
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Interesting thanks zack I have a bit of reading to do on this
06:32
Out of interest besides OB, are there any other blogs you would recommend that discuss pms/futurarchy
Z
06:34
Zack
Distbit also asked about creation of shares.
In amoveo, we create shares basically the same as how it works in truthcoin.
There are contracts. The contract is betting on something, for example "will trump win the election".
so it has 2 shares types. "trump wins" and "trump loses".
And the contract has an input currency. like maybe we are betting in Bitcoin, or we could be betting in some shares that were output from a different market.
If we are betting in bitcoin, that means you can use the contract to turn bitcoin into the 2 share types. and you can combine the 2 share type back into bitcoin.
In amoveo, we create shares basically the same as how it works in truthcoin.
There are contracts. The contract is betting on something, for example "will trump win the election".
so it has 2 shares types. "trump wins" and "trump loses".
And the contract has an input currency. like maybe we are betting in Bitcoin, or we could be betting in some shares that were output from a different market.
If we are betting in bitcoin, that means you can use the contract to turn bitcoin into the 2 share types. and you can combine the 2 share type back into bitcoin.
06:35
in amoveo the tx type for making the complete set of shares is "contract_use_tx". you can read about the tx types in the documentation https://github.com/zack-bitcoin/amoveo-docs/blob/master/design/transaction_types.md
ᴅ
06:37
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Can you create a share pair/group without having to buy all of them and then sell the ones you don't want? Is there a matching mechanism that allows you to only put up the capital for the share(s) that you want to own and have someone else put up the capital for the rest?
Z
06:37
Zack
In reply to this message
bitcoinhivemind.com and truthcoin.info are great.
Robin Hanson was the original person talking about futarchy. He has some nice papers.
Robin Hanson was the original person talking about futarchy. He has some nice papers.
ᴅ
06:38
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Great thanks ok, am already subbed to both of those rss feeds
06:38
I am surprised there are so few people who are deeply passionate about pms
Z
06:39
Zack
In reply to this message
you could participate in a market that lets you swap one of your currencies for the kind of currency that you want to buy.
the swap tab in Amoveo's light wallet automatically finds the combination of markets to give you the best possible price for turning any share type you already have into the share type that you want.
It also does contract_buy_txs if that is the optimal way for you to make the exchange. http://159.89.87.58:8080/wallet.html
the swap tab in Amoveo's light wallet automatically finds the combination of markets to give you the best possible price for turning any share type you already have into the share type that you want.
It also does contract_buy_txs if that is the optimal way for you to make the exchange. http://159.89.87.58:8080/wallet.html
06:42
I used a gradient descent algorithm to calculate the optimal combinations
ᴅ
06:42
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Hmm
06:42
Ok I am missing something
06:43
But maybe I should just read, I am sure all of my question will probably be answered that way
06:43
I find this very interesting, very cool that you have went and actually built all of this
Z
06:44
Zack
we also have limit orders. you could post an order to buy share type #1 in exchange for your veo.
and someone else who wants share type #2 could match with you yes.
and someone else who wants share type #2 could match with you yes.
06:44
we are working towards integrating the limit orders with that constant-product swap tool in the wallet.html page
ᴅ
06:45
So limit orders only exist for creating shares, as opposed to simply buying/trading shares
06:45
Since buying/trading uses constant product
Z
06:48
Zack
the limit orders are handled by swap_tx. swap_tx simply exchange one kind of currency for another between 2 accounts.
we have a tx type called multi-tx. you can put other tx types into it, and it works like flash minting in ethereum.
You can temporarily have negative balances, as long as all the balances end up positive.
So you can accept a limit order, and include that in a multi-tx along with a contract_use_tx to buy the shares that you need to complete the order. They both occur atomically.
creating new shares, or combining shares back to the currency, that is handled by contract_use_tx.
the constant product markets are handled with market_swap_tx. you can also put the market_swap_tx inside of multi_tx.
we have a tx type called multi-tx. you can put other tx types into it, and it works like flash minting in ethereum.
You can temporarily have negative balances, as long as all the balances end up positive.
So you can accept a limit order, and include that in a multi-tx along with a contract_use_tx to buy the shares that you need to complete the order. They both occur atomically.
creating new shares, or combining shares back to the currency, that is handled by contract_use_tx.
the constant product markets are handled with market_swap_tx. you can also put the market_swap_tx inside of multi_tx.
06:49
constant product markets and limit orders are the 2 ways to exchange one kind of currency for another.
06:50
Here are contracts that are live on Amoveo now http://159.89.87.58:8080/contracts_list.html
06:55
It looks like we are about ready to start the activation of hard update #44
ᴅ
07:15
ᴅɪꜱᴛʙɪᴛ ⛓️💵
In reply to this message
Ah nice yeah I was thinking it could be solved that way, that is cool that your protocol allows for such mechanisms.
Z
07:18
Zack
veo is the native currency.
you can also use the multi-tx to buy the veo you need to pay the tx fee. So even if you own zero veo, you can still make txs.
And unlike ethereum, if the tx fails, like by running out of gas for example, you don't have to pay any fee.
The tx isn't included.
multi-tx are limited to only contain txs that take finite time to verify, nothing turing complete.
you can also use the multi-tx to buy the veo you need to pay the tx fee. So even if you own zero veo, you can still make txs.
And unlike ethereum, if the tx fails, like by running out of gas for example, you don't have to pay any fee.
The tx isn't included.
multi-tx are limited to only contain txs that take finite time to verify, nothing turing complete.
Z
07:49
Zack
This is the announcement for hard update #44.
it will activate on block 149729, which is around the 28th of January.
The purpose of this update:
* turn off the old channel types that we no longer support.
* add a new tx type swap_tx2, meant to replace swap_tx. It allows for partially matching limit orders, and it gives a better user experience for canceling unmatched limit orders.
Here is documentation for the normal process for updating https://github.com/zack-bitcoin/amoveo-docs/blob/master/getting-started/updating.md
it will activate on block 149729, which is around the 28th of January.
The purpose of this update:
* turn off the old channel types that we no longer support.
* add a new tx type swap_tx2, meant to replace swap_tx. It allows for partially matching limit orders, and it gives a better user experience for canceling unmatched limit orders.
Here is documentation for the normal process for updating https://github.com/zack-bitcoin/amoveo-docs/blob/master/getting-started/updating.md
s
17:19
sanket
Z
17:24
Zack
In reply to this message
I don't see how this could help Amoveo, or any other project. It looks like a scam.
17:25
Any project that talks about airdrops is usually a scam.
J
18:18
Josh
@zack_amoveo what do you think of this new stablecoin? https://medium.com/fei-protocol/introducing-fei-protocol-2db79bd7a82b
Z
18:44
Zack
"The price function will start low to reward early adopters for purchasing FEI"
so, it is a pyramid scheme.
"Arbitrageurs can buy on the bonding curve and sell on secondary markets if the price is higher elsewhere."
the existence of arbitrage means value is being wasted.
Bonding curves are a terrible way to create a stablecoin. Every time the price of the collateral moves, the miners can extract a ton of value from users of that stablecoin.
It involves an ICO, re-branded as "IDO".
There is no reason to raise money to create a cryptocurrency collateralized stablecoin.
There is no reason to over-reward early participants.
Consider basic principles from finance.
If one person wants a stablecoin collateralized with cryptocurrency, meaning they want less risk against the cryptocurrency price volatility, then they need to sell that risk to someone else.
Risk cannot be created or destroyed, it can only be transferred to someone else.
So if one person is holding a stablecoin, that means they are paying someone else to hold extra cryptocurrency price volatility risk.
Any supposed stablecoin plan needs to explain who is holding the extra risk, and why they are incentivized to do this.
Pretending that the risk is disappearing, that is an impossible fantasy. People trying to sell impossible fantasies are scammers.
The idea that a basic risk transfer contract should somehow involve raising $250 million, and paying out over-rewards to early contributors. it is ridiculous. This is an obvious scam.
so, it is a pyramid scheme.
"Arbitrageurs can buy on the bonding curve and sell on secondary markets if the price is higher elsewhere."
the existence of arbitrage means value is being wasted.
Bonding curves are a terrible way to create a stablecoin. Every time the price of the collateral moves, the miners can extract a ton of value from users of that stablecoin.
It involves an ICO, re-branded as "IDO".
There is no reason to raise money to create a cryptocurrency collateralized stablecoin.
There is no reason to over-reward early participants.
Consider basic principles from finance.
If one person wants a stablecoin collateralized with cryptocurrency, meaning they want less risk against the cryptocurrency price volatility, then they need to sell that risk to someone else.
Risk cannot be created or destroyed, it can only be transferred to someone else.
So if one person is holding a stablecoin, that means they are paying someone else to hold extra cryptocurrency price volatility risk.
Any supposed stablecoin plan needs to explain who is holding the extra risk, and why they are incentivized to do this.
Pretending that the risk is disappearing, that is an impossible fantasy. People trying to sell impossible fantasies are scammers.
The idea that a basic risk transfer contract should somehow involve raising $250 million, and paying out over-rewards to early contributors. it is ridiculous. This is an obvious scam.
18:47
Finance isn't complicated. It is just trading risk.
The only reason to act like it is too complicated to understand is because they don't want you to understand what is actually happening.
The only reason to act like it is too complicated to understand is because they don't want you to understand what is actually happening.
19:00
It is like how a magician works. They get your attention in one place, so you don't notice what they are doing elsewhere.
15 January 2021
Z
02:03
Zack
I started making the new merkle tree for storing stablecoin info.
J
06:09
Josh
Is there a special merkle tree just for stablecoins?
Z
06:09
Zack
there will be
06:10
the trees are all hashed together into a single root for the block header, so it is kind of like different branches of one big tree
06:11
https://github.com/zack-bitcoin/amoveo-docs/blob/master/design/trees.md here is the documentation for the trees now
J
07:40
Josh
Is that the "contracts" merkle tree?
Z
07:41
Zack
In reply to this message
https://github.com/zack-bitcoin/amoveo-docs/blob/master//white_paper.md
Have you seen the smart contract section in the new whitepaper?
Have you seen the smart contract section in the new whitepaper?
07:42
it has links to more details as well
07:43
In Amoveo we have abstracted the idea of truthcoin oracles into 2 parts.
One very specifically asks a yes/no question, and makes that data available. we call it an "oracle".
the other, called a "contract", does the rest. It is also able to do basic computation, and the oracle data is available during that computation.
the result of the computation decides how to divide up the collateral inside the contract between between the different share types.
One very specifically asks a yes/no question, and makes that data available. we call it an "oracle".
the other, called a "contract", does the rest. It is also able to do basic computation, and the oracle data is available during that computation.
the result of the computation decides how to divide up the collateral inside the contract between between the different share types.
07:44
This also allows us to make the smart contract before the oracle, which allows us to do last-minute optimizations in the oracle text.
Сергей Краснодар. Россия invited Сергей Краснодар. Россия
Z
08:12
Zack
I started making the stablecoin_new_tx.
For creating new perpetual stablecoins.
It looks like it will be 8 tx types now.
For creating new perpetual stablecoins.
It looks like it will be 8 tx types now.
Deleted invited Deleted Account
Deleted invited Deleted Account
J
17:53
Josh
Cool, I'll read up.
Z
18:24
Zack
Are stablecoins always collateralized by veo?
Do we want to build a stablecoin on top of a stablecoin?
Do we want to build a stablecoin on top of a stablecoin?
J
18:26
Josh
We could build baskets of stablecoins. But probably not the highest priority.
Z
18:29
Zack
the stablecoin is going to be influenced by the demand for long-veo. leveraged veo is the other side of the contract.
If there is high demand for leveraged veo, the stablecoin will do better than the underlying target.
If there is low demand for leveraged veo, it will do worse.
I wonder if we can reduce these fluctuations by layering perpetual stablecoins
If there is high demand for leveraged veo, the stablecoin will do better than the underlying target.
If there is low demand for leveraged veo, it will do worse.
I wonder if we can reduce these fluctuations by layering perpetual stablecoins
J
18:31
Josh
But it will always involve veo right?
Z
18:39
Zack
yeah. veo is the base collateral under everything
18:40
maybe there is some way to show that layering perpetual stablecoins this way isn't helpful. that would allow me to use a simpler design.
J
19:20
Josh
My intuition is that veo is the base collateral so anything you do on top of that won't help get rid of veo volatility. As you said, it has to go somewhere.
19:22
Is it possible to do a 2 way peg with another coin?
Z
19:29
Zack
In reply to this message
the idea is that the layer-1 perpetual stablecoin could be the collateral for a finite contract.
type 2 of that finite contract would be a bet that the demand for long-veo is increasing.
type 1 would be the collateral for the layer-2 perpetual stablecoin.
type 2 of that finite contract would be a bet that the demand for long-veo is increasing.
type 1 would be the collateral for the layer-2 perpetual stablecoin.
19:29
In reply to this message
yeah it is possible. im not sure if it is something we would want to do though.
J
Z
19:35
Zack
In reply to this message
because the finite stablecoin's value is influenced by the demand for leveraged veo.
19:36
In reply to this message
added complexity. less reason to hold veo. a distraction from building other things.
J
19:39
Josh
In reply to this message
I still don't get it. If the demand for veo isn't sufficient, layer 1 will have a problem anyway and that will influence layer 2.
Z
19:39
Zack
demand for veo and demand for leveraged veo are different things.
19:40
the layer 1 perpetual can handle changing prices in veo fine.
the problem is that leveraged veo and veo change relative to each other.
the problem is that leveraged veo and veo change relative to each other.
J
19:44
Josh
So this is a way to avoid needing leveraged veo?
Z
19:46
Zack
no.
if someone is holding stablecoin, the extra veo risk needs to go somewhere.
if someone is holding stablecoin, the extra veo risk needs to go somewhere.
19:46
this is a way to have a stablecoin that is less influenced by fluctuations in demand for leveraged veo.
J
19:49
Josh
So where does that risk go?
Z
20:38
Zack
In the layer 2 finite contract. We split up the layer 1 perpetual stablecoins into layer 2 finite stablecoins and layer 2 type 2.
Layer 2 type 2 holds that risk.
Layer 2 type 2 holds that risk.
J
20:41
Josh
Who would want the layer 2 finite stablecoin?
Z
21:03
Zack
Layer 2 finite stablecoin is the collateral for the layer 2 perpetual stablecoin.
Z
21:44
Zack
the stablecoin_new_tx is a little tricky.
We want to generate the finite stablecoin contract to create collateral for the stablecoin_new_tx.
But to be compatible with the memoryless full node design, we need to know the contract ID of the finite stablecoin contract in order to build this tx.
So, the CID probably shouldn't depend on which block the stablecoin_new_tx gets included in, otherwise we wouldn't be able to hold it in the tx pool normally.
The CID is deterministically related to the code of the smart contract.
So we can't hard code that block height into the contract.
We want to generate the finite stablecoin contract to create collateral for the stablecoin_new_tx.
But to be compatible with the memoryless full node design, we need to know the contract ID of the finite stablecoin contract in order to build this tx.
So, the CID probably shouldn't depend on which block the stablecoin_new_tx gets included in, otherwise we wouldn't be able to hold it in the tx pool normally.
The CID is deterministically related to the code of the smart contract.
So we can't hard code that block height into the contract.
21:44
I wonder if it is better to build the finite stablecoin contract first, and have the stablecoin_new_tx reference that existing contract.
ᴅ
23:07
ᴅɪꜱᴛʙɪᴛ ⛓️💵
I saw an interesting design for a stablecoin a while back whereby you convert your say 1 dollar of collateral into:
1) A token that is worth a maximum of 1 dollar as long as the collateral is worth >$1 but is worth whatever the collateral is worth if the collateral is worth <$1 at the expiration date
2)A "call option token" that is worth whatever the difference of the value of the collateral is at expiration to $1 if the collateral is worth >$1 and worth nothing if it is below $1
1) A token that is worth a maximum of 1 dollar as long as the collateral is worth >$1 but is worth whatever the collateral is worth if the collateral is worth <$1 at the expiration date
2)A "call option token" that is worth whatever the difference of the value of the collateral is at expiration to $1 if the collateral is worth >$1 and worth nothing if it is below $1
23:07
You then sell the call option onto the market
23:07
I.e transfer the risk to someone else
23:08
And now you are left with the $1 token as well as the sale price of the call option
23:09
And then you basically keep refreshing these token pairs when they expire
23:09
Something like that
23:09
It is a pretty interesting idea imo as it seems to remove the need for over collateralization
23:09
Although I may be missing something, it was a while ago that I read about it
23:11
I may also be incorrectly describing it or missing a detail
Z
23:12
Zack
That is the stablecoin design we have right now.
the drawback is that it has an expiration. So it can't be used for very long-term bets, and it requires coming online occasionally to update it.
the stablecoin tab in the wallet.html page of the light node allows creating that kind of contract.
the drawback is that it has an expiration. So it can't be used for very long-term bets, and it requires coming online occasionally to update it.
the stablecoin tab in the wallet.html page of the light node allows creating that kind of contract.
23:12
we want to use these finite-stablecoins as collateral for the perpetual stablecoins.
ᴅ
23:12
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Ah interesting ok I am surprised I remembered the details then lol
23:13
I thought I had made an error in explaining it
23:13
So does that mean that your stablecoin design doesn't require overcollaterlisation?
Z
23:16
Zack
Now we are talking about the finite stablecoin.
You can set the margin to whatever you want.
But as you said, if the price of veo drops below the margin, then the stablecoin starts tracking the price of veo instead of usd or btc or whatever.
You can set the margin to whatever you want.
But as you said, if the price of veo drops below the margin, then the stablecoin starts tracking the price of veo instead of usd or btc or whatever.
Elvin Guzman invited Elvin Guzman
Z
23:16
Zack
This is how stablecoins work in the truthcoin design as well
ᴅ
23:17
ᴅɪꜱᴛʙɪᴛ ⛓️💵
In reply to this message
Wouldn't it still be offset in the positive direction by the price that the call option was sold for?
23:18
Like if the collateral falls below $1 then your stablecoin is still above $1 until the collateral falls below $1 minus the price that the call option was sold for
23:18
Right?
Z
23:20
Zack
Its not a call option.
An option means you have the option to buy or sell at a price in the future. It is an assymetric derivative that usually has trust in one direction.
I think the technical term for our derivative is a forwards contract. Basically, the long-veo side is worth all the leftover veo, once the stablecoin side is paid the value of their stablecoins in veo.
An option means you have the option to buy or sell at a price in the future. It is an assymetric derivative that usually has trust in one direction.
I think the technical term for our derivative is a forwards contract. Basically, the long-veo side is worth all the leftover veo, once the stablecoin side is paid the value of their stablecoins in veo.
ᴅ
23:20
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Hmm let me think
Z
23:20
Zack
So the long-veo side is like holding leveraged veo
ᴅ
23:21
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Isn't it equivalent to a synthetically settled call option?
23:21
It seems to me like it behaves the same way
Z
23:22
Zack
Options are popular today because trust is in one direction. So a large centrally trusted institution can be enforced by the law.
And its customers are untrusted.
And its customers are untrusted.
ᴅ
23:22
You can't be margin called with an option (unless you are the seller ofc)
Z
23:23
Zack
Forwards contracts are better suited for blockchains. We dont have any trust, and the users are symmetric. So a symmetric derivative makes more sense for us.
Any kind of derivative is just a way to trade risk, so they arent so different from each other.
Any kind of derivative is just a way to trade risk, so they arent so different from each other.
ᴅ
23:24
ᴅɪꜱᴛʙɪᴛ ⛓️💵
What is the difference exactly between a forward contract and a call option? I was watching a lecture on this a couple of weeks ago, and to me it seemed as though a forwards contract was a less-standardised type of futures contract
Z
23:24
Zack
In reply to this message
Margin calls only happen if you use debt to create the derivative. Blockchains dont have debt.
When our contracts pass the margin, then one side of the contract is worth zero. And the other side is worth all the veo collateral.
When our contracts pass the margin, then one side of the contract is worth zero. And the other side is worth all the veo collateral.
23:26
In reply to this message
In amoveo contracts, the usd stablecoin side that you buy for $10, it stays worth $10 of veo, as long as veo is above the margin.
The other side of the contract, it is worth all the extra veo collateral that doesnt go to the stablecoin holders.
The other side of the contract, it is worth all the extra veo collateral that doesnt go to the stablecoin holders.
ᴅ
23:26
ᴅɪꜱᴛʙɪᴛ ⛓️💵
So are you saying the only difference between this leveraged token and a call option is that this token is constructed in a symmetric manner rather than an assymetric manner as call options are usually created?
Z
23:28
Zack
An option gives one participant the option to buy or sell at a future date, and the other is obliged to do it. In a forwards contract they both have an obligation to make the trade at a future date.
ᴅ
23:28
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Hmm ok.
23:28
Maybe a better way of putting it is:
23:29
Would this token be priced the same way as a call option with the same qty and settlement date, if it were to be traded?
23:31
If the answer is no then I clearly need to read more and am fine with ending this conversation here as I have a lot of amoveo docs on my reading backlog👌
Z
23:32
Zack
Call options can get margin called right? So that is an extra risk that doesnt exist with stablecoin contracts.
23:33
In reply to this message
This is a complete definition of amoveo stablecoin contracts. Im not sure why you want to pull in legacy terminology.
ᴅ
23:33
ᴅɪꜱᴛʙɪᴛ ⛓️💵
In reply to this message
Not if you are sure that the underwriter has the collateral in the denomination of the contract, which for the sake of this conversation I am assuming they do and are guaranteed to settle according to the contract terms
23:34
In reply to this message
Mainly because I like to mentally link related concepts as much as possible to facilitate more efficient transfer of ideas and concepts between different applications
23:36
And am also interested in understanding the differences between these two assets if there are any insofar as ones that would cause them to be priced differently (I completely appreciate the fact that the way in which they are generated is very different)
23:37
If it is true that there is a difference in how the market would price them than an equivalent call option, then that signals to me that I have a gap in my understanding of either what they are or what call options are
Z
23:55
Zack
In reply to this message
even if your broker is rich, they aren't going to hold an out-of-the-money contract for their client.
23:57
In reply to this message
amoveo smart contracts are turing complete.
The current stablecoin contract we are using is one example of what is possible.
We can program in whatever risk profile the users want.
The current stablecoin contract we are using is one example of what is possible.
We can program in whatever risk profile the users want.
23:58
Maybe it helps to think of Amoveo contracts as cash settled bets, where all the collateral is locked up for the duration of the bet, and you can't risk any more than what you had locked into the collateral.
J
16 January 2021
ᴅ
00:12
Ok cool yeah I get it 👍
Z
00:14
Zack
00:14
I think the risk profile is currently something like this
00:14
The 2 share types need to add up to veo.
MF
00:23
Mr Flintstone
hard to hand draw but the price in usd is more nonlinear i think
00:24
because longveo’s value priced in veo is linear but in usd it isnt cuz the veo is moving in usd as well
00:25
actually not sure
Z
00:26
Zack
Say the margin price is $20 per veo.
So if we compare the price of long-veo when veo is $30 vs $40. I think it doubles.
And comparing $40 to $60, it doubles again.
I think it could be non-linear if we change the units.
So if we compare the price of long-veo when veo is $30 vs $40. I think it doubles.
And comparing $40 to $60, it doubles again.
I think it could be non-linear if we change the units.
ᴅ
00:33
In reply to this message
Yeah the value of the long-veo token is the value of the collateral minus the stablecoin value if it is positive, else $0
00:36
So if the value of veo doubles, the value of the collateral doubles and the value of the long-veo token increases by a % that approaches 100% as the stablecoin value becomes a smaller percentage of the total collateral value
00:36
But the relationship between the value of long-veo and the price of veo is a linear one
00:39
So your drawing is correct
00:40
Why do you think that overcollateralised stablecoins have become so popular on ethereum as opposed to this design?
00:40
Couldn't this design be packaged up in such a way to make it much more tollerant to risk/black swan events in the price of the collateral?
Z
01:02
Zack
In reply to this message
I think for a couple reasons.
They cant do an ico for this.
It has an expiration date.
They cant do an ico for this.
It has an expiration date.
01:03
In reply to this message
Thats what im working on now. A perpetual stablecoin collateralized with these finite stablecoins, and when the price approaches the margin, or when it is near expiry, there is an auction to swap out the finite stablecoin collateral for a new one.
ᴅ
01:13
ᴅɪꜱᴛʙɪᴛ ⛓️💵
In reply to this message
Couldn't they just create a platform thing that has a value extraction/rent seeking token that is paid dividends from fees taken on stablecoin txs?
01:13
I mean this already exists in the form of lien protocol
01:14
I just don't know how popular it is
01:15
In reply to this message
Hmm I would be interested in better understanding how this would work
01:16
To me it seems that if the price of the collateral goes down you are already in the red relative to your initial investment...
01:16
Since the stablecoin token isn't actually worth $1
01:17
It is, when created, worth the difference between $1 and the price of the long veo token
01:17
So by creating both assets they are still, when combined, only worth $1 (how much it cost you to create them)
01:18
So if you sell the call option and then veo declines
01:18
Aren't you already in the red?
01:18
I must be missing something
Z
01:19
Zack
In reply to this message
lets say the stablecoin was created by locking 1000 veo into the contract.
So there are 1000 units of stablecoin, and 1000 units of long-veo.
If the price of veo is falling, eventually the long-veo is worth zero.
So to alert the blockchain that this is happening, we allow anyone to use 950 veo to buy all the finite stablecoins backing the perpetual stablecoin.
At that point the perpetual stablecoin is temporarily backed by veo.
We do an auction to turn this veo into a new finite-stablecoin contract to back the perpetual stablecoins.
So there are 1000 units of stablecoin, and 1000 units of long-veo.
If the price of veo is falling, eventually the long-veo is worth zero.
So to alert the blockchain that this is happening, we allow anyone to use 950 veo to buy all the finite stablecoins backing the perpetual stablecoin.
At that point the perpetual stablecoin is temporarily backed by veo.
We do an auction to turn this veo into a new finite-stablecoin contract to back the perpetual stablecoins.
01:20
ive got notes about my planned implementation at the top of the todo list https://github.com/zack-bitcoin/amoveo-docs/blob/master/todo.md
01:21
In reply to this message
both assets together are worth the same amount of veo. That veo could be worth a different amount of USD.
ᴅ
01:28
ᴅɪꜱᴛʙɪᴛ ⛓️💵
In order to generate $10 worth of stablecoins using this mechanism, how many dollars worth of collateral do I need to deposit?
01:28
Just to clarify
Z
01:31
Zack
In reply to this message
It depends what the margin price is.
If it starts at exactly the margin price, you only need $10 of veo.
If you want veo to be able to fall 2x without crossing the margin, then you need at least $20 of veo.
When you generate the stablecoins, you simultaneously receive the long-veo on the other side of the contract.
If it starts at exactly the margin price, you only need $10 of veo.
If you want veo to be able to fall 2x without crossing the margin, then you need at least $20 of veo.
When you generate the stablecoins, you simultaneously receive the long-veo on the other side of the contract.
ᴅ
01:32
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Got it
01:32
Ok it makes sense now
01:33
If you put $10 of collateral, you will have to instantly liquidate though right?
01:33
Like it wouldn't work
Z
01:33
Zack
Oh, thats another reason these contracts arent so popular as stablecoins yet.
There are so many different margins, and they arent fungible if they have different margins.
There are so many different margins, and they arent fungible if they have different margins.
ᴅ
01:34
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Hmm interesting
Z
01:35
Zack
In reply to this message
Idk what liquidate means in this context.
The long-veo side would be worth zero.
If the price of veo fell, the stablecoin would fall along with veo. The long-veo side is still worth zero.
But if the price of veo rose above the margin, then the long-veo would have value again.
The long-veo side would be worth zero.
If the price of veo fell, the stablecoin would fall along with veo. The long-veo side is still worth zero.
But if the price of veo rose above the margin, then the long-veo would have value again.
ᴅ
01:36
ᴅɪꜱᴛʙɪᴛ ⛓️💵
I should probably wait until tomorrow morning to continue this discussion, it is 4:30 am here
01:36
Which probably isn't making this any easier
01:37
Thanks for explaining all of this though 👍
Z
01:40
Zack
Im happy to explain.
Answering questions is how a lot of the docs get written.
Answering questions is how a lot of the docs get written.
ᴅ
01:48
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Appreciate it Zack 👌 I am glad I am not the only one enjoying these discussions
J
02:05
Josh
I'm enjoying them as a lurker.
MF
02:12
Mr Flintstone
In reply to this message
i think for the perpetual scheme you might need a liquidation provision in the oracle language that locks in the finite stablecoin’s value at 1:1 in veo and sends the longveo to near 0 once the collateral limit price is passed. that way, veo denominated arb players are willing to buy the finite stablecoin for veo immediately
Meaning once the longveo goes to 0 it stays there even if veo price rises back above the collateral limit price
Meaning once the longveo goes to 0 it stays there even if veo price rises back above the collateral limit price
Z
06:46
Zack
In reply to this message
This makes sense.
I think this decision is happening inside the smart contract, so we dont need to worry about this at the tx type level.
Im setting up the perpetual stablecoin to have a hash of a smart contract embedded in it, that will be used to build the finite stablecoin contracts.
I think this decision is happening inside the smart contract, so we dont need to worry about this at the tx type level.
Im setting up the perpetual stablecoin to have a hash of a smart contract embedded in it, that will be used to build the finite stablecoin contracts.
Gary Wang invited Gary Wang
Z
17:07
Zack
https://github.com/zack-bitcoin/amoveo-docs/blob/master/basics/futarchy_efficiency.md
I think i solved the problem of capital efficiency in futarchy markets where we are making predictions about the impact of decisions that are very improbable.
Futarchy is stronger in communities that believe in it.
I think i solved the problem of capital efficiency in futarchy markets where we are making predictions about the impact of decisions that are very improbable.
Futarchy is stronger in communities that believe in it.
G
21:31
Gregory
It workes with every fugazzi actually. Look tether)
17 January 2021
Deleted invited Deleted Account
Z
07:24
Zack
In reply to this message
So my plan is that the tx should have a "expiration" in it for when the next finite stablecoin should expire.
And the tx type should do some check that this expiration is near enough to when the next period would end.
But how near should it be? within 10%?
should that be a governance variable? or another number embedded in the stablecoin object?
And the tx type should do some check that this expiration is near enough to when the next period would end.
But how near should it be? within 10%?
should that be a governance variable? or another number embedded in the stablecoin object?
x
11:28
it already has a well designed wallet, but the design hasn't been added to light node
11:33
Desktop wallets are updated to version 2.7!
Oracles support with scalar ones, fixes and improvements including upcoming Ledger support.
Download links:
— Windows: https://myveowallet.com/desktop/win/latest (portable EXE)
— OS X: https://myveowallet.com/desktop/osx/latest (PKG)
— Linux: https://myveowallet.com/desktop/linux/latest (AppImage)
Under Linux you can also use Snap:
Oracles support with scalar ones, fixes and improvements including upcoming Ledger support.
Download links:
— Windows: https://myveowallet.com/desktop/win/latest (portable EXE)
— OS X: https://myveowallet.com/desktop/osx/latest (PKG)
— Linux: https://myveowallet.com/desktop/linux/latest (AppImage)
Under Linux you can also use Snap:
snap install amoveo-wallet
11:33
just saw this, so it has desktop version too
Z
16:49
Zack
That wallet only works for spending, right?
The light node i build has tons more features, and works anywhere a browser does.
The light node i build has tons more features, and works anywhere a browser does.
x
17:07
x
I havent tried the wallet
Z
x
17:21
x
by welll design i mean it looks good
18 January 2021
B
02:19
Ben
I does not only look good, you actually understand what you are doing :)
x
x
14:14
x
or maybe if someone wrap the light node into an android app & chrome extention, that will be much safer than then using a domain name
14:15
it's too inconvenient for average user to open
Z
17:09
Zack
opening a url is as easy as clicking on it.
Every time I try to install an app, my phone is all "You need to delete 2 other apps to have enough space for this new app"
Every time I try to install an app, my phone is all "You need to delete 2 other apps to have enough space for this new app"
17:10
if you are using a significant amount of cryptocurrency, you need to be using cold storage.
x
20:50
x
In reply to this message
light node is actualy ok solution.
Still, 1 new users don't know light node is actually the wallet, so the download link of the light node is not easy to find for average users: https://github.com/zack-bitcoin/light-node-amoveo/archive/master.zip
2 and it won't be very clear to to them which file can open the wallet after it's unziped
Still, 1 new users don't know light node is actually the wallet, so the download link of the light node is not easy to find for average users: https://github.com/zack-bitcoin/light-node-amoveo/archive/master.zip
2 and it won't be very clear to to them which file can open the wallet after it's unziped
19 January 2021
Deleted invited Deleted Account
Deleted invited Deleted Account
T
13:23
Topab
Zack, have you come across this whitepaper and do you think it could be useful in Amoveo? I was wondering that because there will exist different markets, oracles, subcurrencies, ... the approach in this paper could be useful to allow for further scalability https://github.com/holochain/holochain-proto/blob/whitepaper/holochain.pdf
Z
16:18
Zack
In reply to this message
This is a sharding proposal.
It looks like there is a ton of needless complexity in their documentation.
When i see needless complexity, i have to think that they did not actually solve anything useful, and they dont want their investors to realize.
A sharding proposal should be focused on the speaker listener fault equivalence, which is the data availability problem. I dont see it mentioned in their documentation.
For example, if you look at optimistic rollup proposals, data availability is solved by having everyone have access to the data, and having a low cost format for verifying it is correct.
Zk rollups solve data availability by making short proofs of the datas existence, and everyone has access to the proofs.
Drivechain solves data availability by blocking withdraws from any sidechains that dont make all the data available.
I think that holochains documentation isnt optimized to explain the solution to any real problem, rather it is optimized to prevent anyone from understanding it.
It looks like there is a ton of needless complexity in their documentation.
When i see needless complexity, i have to think that they did not actually solve anything useful, and they dont want their investors to realize.
A sharding proposal should be focused on the speaker listener fault equivalence, which is the data availability problem. I dont see it mentioned in their documentation.
For example, if you look at optimistic rollup proposals, data availability is solved by having everyone have access to the data, and having a low cost format for verifying it is correct.
Zk rollups solve data availability by making short proofs of the datas existence, and everyone has access to the proofs.
Drivechain solves data availability by blocking withdraws from any sidechains that dont make all the data available.
I think that holochains documentation isnt optimized to explain the solution to any real problem, rather it is optimized to prevent anyone from understanding it.
ᴅ
16:25
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Yeah I read the Holochain "green paper" a couple of days ago coincidentally
16:25
I found it to be incomprehensible
16:25
I thought maybe it would make more sense if I read their whitepaper (the one posted above) beforehand
16:26
Btw @zack_amoveo what would you recommend for best understanding the importance of the data availability problem and how it relates to sharing/horizontal/layer 2 scaling solutions?
16:27
I am really interested in better understanding this principle as it seems quite integral to the problem
Z
16:29
Zack
https://vitalik.ca/general/2017/07/16/triangle_of_harm.html vitalik talks about it in this post
ᴅ
16:31
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Thanks 👍
ᴅ
17:43
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Do you have any articles applying this concept to any technologies in particular?
17:47
I am yet to read this article but have added it to my reading list and it would probably, for pedagogical purposes, be good to have some examples of it being applied as well if you know of anyway 👍
J
18:29
Josh
Zk rollups is really interesting tech.
ᴅ
18:36
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Do you have any recommendations for intros/explanations of it?
18:36
It is something I have been meaning to learn about for a while
20 January 2021
Ilmu Somebody invited Ilmu Somebody
IS
01:06
Ilmu Somebody
Edit: re zkrollups:
You collect a bunch of transactions and collapse them into the netto delta and make a zero knowledge proof that you did this correctly, then you submit to the chain the proof and the delta needed to correct the on-chain state
You collect a bunch of transactions and collapse them into the netto delta and make a zero knowledge proof that you did this correctly, then you submit to the chain the proof and the delta needed to correct the on-chain state
01:10
Edit: separate msg:
Zack, I came here from reading your criticism of avalanche, it's true that you can denial of service with a relatively small part of the validator pool and you can also do this kind of "soft influence" by being stubborn in the gossiping thing but from what I can tell these platforms aren't developing in the monero direction (anonymous "cash" transactions with decentralized validation and censorship resistance) but rather a collaborative AWS where the validators are known and you just implement 100% slashing if someone does something wrong (not right away, first you gather inertia and then you slowly make things more draconian), to me the strongest criticism of avax is that the main net is owned by some company and everyone is forced to validate it...
Zack, I came here from reading your criticism of avalanche, it's true that you can denial of service with a relatively small part of the validator pool and you can also do this kind of "soft influence" by being stubborn in the gossiping thing but from what I can tell these platforms aren't developing in the monero direction (anonymous "cash" transactions with decentralized validation and censorship resistance) but rather a collaborative AWS where the validators are known and you just implement 100% slashing if someone does something wrong (not right away, first you gather inertia and then you slowly make things more draconian), to me the strongest criticism of avax is that the main net is owned by some company and everyone is forced to validate it...
01:12
so I guess my question is; can you give me a quick rundown of how you classify the different projects and tell me whether you think security or expedience is more likely to win out in the end?
You don't have to, I'm currently lurking through the docs on the github and I just became aware of this project so maybe the questions are answered somewhere in there..
You don't have to, I'm currently lurking through the docs on the github and I just became aware of this project so maybe the questions are answered somewhere in there..
01:13
(I did read through a couple of months of the telegram chat here and it looks like the idea with your project is to reify a lot of finance understanding to make participation in a decentralized economy more user friendly)
J
01:50
Welcome @IlmuKalm, I know you from the Statebox channel 🙂
IS
01:51
Now more through your code in lamassu than from statebox but both :p
01:52
wow I'm really loving the documentation, super concise
01:53
it's super cool how everything is kept in balance in this project but https://github.com/zack-bitcoin/amoveo-docs/blob/master/design/governance.md this looks like it will bottleneck on the UX/UI
Deleted invited Deleted Account
Z
04:32
Zack
In reply to this message
https://github.com/zack-bitcoin/amoveo-docs/blob/master/design/consensus_efficiency.md
It seems to me that higher security, more scalability, and lower tx fees, these are all different ways of measuring the same underlying factor.
So security and expedience are really the same thing.
It seems to me that higher security, more scalability, and lower tx fees, these are all different ways of measuring the same underlying factor.
So security and expedience are really the same thing.
IS
04:45
Ilmu Somebody
Yeah sorry I'm still reading will have more thoughtful comments in the future'
04:54
From what I can tell this model also presupposes a certain consensus type (namely nakamoto)?
Z
04:57
Zack
In reply to this message
thanks for the heads up. I fixed it just now, so if you refresh the page, it should work.
IS
04:57
Ilmu Somebody
Thank you
Z
04:58
Zack
In reply to this message
it doesn't. in particular, check out the trust theory page. it compares other consensus mechanisms to nakamoto
IS
04:58
Ilmu Somebody
Yep, I'll read everything.
T
09:01
Topab
In reply to this message
Thanks Zack for taking the time to read. It really sound that what they describe in the paper is a different way of doing sharding, but it sounds to me like it is efficient. The protocol enables each node to keep a record of all their transactions and only when they join an app parts of their information gets published and those in the same app can certify the vearcity of that information. Nodes that are not in the app do not need to keep any of that information. Anyway, I do not want to go deep into it, just that I thought it could be interesting for Amoveo in some way given there will be many different markets, oracles and other parts that only some participants will be interested so I was thinking it could be useful to look into such approach to avoid replicating information in the blockchain. Is Amoveo doing sharding?
Z
13:29
Zack
In reply to this message
It looks like a scam.
If it is efficient, then how do they efficiently solve speaker listener equivalency faults in the case of data availability problems?
Amoveo uses the memoryless full node strategy. It is a kind of optimistic rollup.
If it is efficient, then how do they efficiently solve speaker listener equivalency faults in the case of data availability problems?
Amoveo uses the memoryless full node strategy. It is a kind of optimistic rollup.
IS
18:33
Ilmu Somebody
do you guys accept truthcoins arguments?
18:34
it's weird, cryptography people go "we have a forever 22kb blockchain using better cryptography" operating system people go "we have a more efficient communication pattern allowing 2 second finality" economists go "we can bribe someone to destroy your projects, nothing beats bitcoin"
18:35
meanwhile some 3 dudes in china own 58% of the hashrate, finality takes hours, the transaction fees are immense, everything about it is terrible and yet this is supposed to be the best ever thing because MR = MC
18:38
I can't help but think that there is a mistaken assumption about the motivations of people. The only actually good argument is that oracles are the only problem worth solving in these programmable platforms and they indeed can only be solved using economics.
Z
18:43
Zack
Everyone wants to think that the tools they spent years studying are the tools needed to make the best blockchain.
If all you have is a hammer, everything looks like a nail.
I tend to think truthcoin is right.
If all you have is a hammer, everything looks like a nail.
I tend to think truthcoin is right.
IS
18:48
Ilmu Somebody
The thing is, they are right that their tools are applicable (all factions) but there are a lot of concerns to balance and it's easy to overlook some of them
18:49
and it seems to me that all of the calculations of the economics are thrown off if the system has permissioned validation, i.e. 100% slashing
18:50
but then you are just building a decentralized AWS and your concerns are no longer mechanism design but just computer science
18:52
from what I can tell; the only project that is trying to solve the whole thing is this one (surprisingly! I just heard of it yesterday.. maybe there are others hiding in the woodworks), what I question is this "the market can stay irrational for longer than you can stay solvent" I mean ethereum is clearly useless from day 1 (by the kind of logic being applied by econ) but still it's worth quite a lot and it's hard to call it a complete failure.
Z
IS
18:56
Ilmu Somebody
truthcoin on ethereum: "it has no oracles, it has no use case"
Z
18:57
Zack
In reply to this message
Game theory still applies in that case.
If there is a coalition that can rob you, you need to pay high enough fees to them so it is not in their interest to rob you. The long term profits from fees needs to exceed the short term profits of abusing the users.
And a coalition could be bribed.
If there is a coalition that can rob you, you need to pay high enough fees to them so it is not in their interest to rob you. The long term profits from fees needs to exceed the short term profits of abusing the users.
And a coalition could be bribed.
IS
18:59
Ilmu Somebody
sure, but my point is that you can be a lot more brutish in your econ if you have access to violence to offset the temptation of robbery
Z
18:59
Zack
In reply to this message
Truthcoin on truthcoin: oracle parasites inside ethereum make it impossible to build a secure oracle in truthcoin.
Surely the ability to do cheap experiments without launching a new blockchain is worth something.
If it is worth it to make sidechains/new blockchains, then it must also be worth it to make cheaper experiments using eth contracts.
Surely the ability to do cheap experiments without launching a new blockchain is worth something.
If it is worth it to make sidechains/new blockchains, then it must also be worth it to make cheaper experiments using eth contracts.
IS
19:00
Ilmu Somebody
okay fair.
Z
19:00
Zack
In reply to this message
That violence becomes another player in the game theory.
If a violent actor can rob users, it needs to be paid enough fees so it is not in its interest to rob users.
If a violent actor can rob users, it needs to be paid enough fees so it is not in its interest to rob users.
19:01
A blockchain immune to the violence doesnt need to pay that tax, and so should eventually win.
IS
19:01
Ilmu Somebody
yeah you (and truthcoin) are right I guess, I just wonder how long it takes for the reality to converge
ᴅ
19:11
He referenced an article he wrote on the topic and I haven't read it yet but it is on my reading list
19:11
So is it true that the summary is that he thinks tc is vulnerable to eth parasite contracts?
19:11
And he has no solution?
Z
19:13
Zack
The vulnerability is because his current design requires the truthcoin oracle to extract trading fees.
That is why he had to make such a complicated system of channels for truthcoin. To be sure that trading fees are still paid, even if you use channels to trade.
Amoveo's oracle has no fees, and so is immune to parasites.
That is why he had to make such a complicated system of channels for truthcoin. To be sure that trading fees are still paid, even if you use channels to trade.
Amoveo's oracle has no fees, and so is immune to parasites.
IS
19:15
Ilmu Somebody
yeah the design where you keep raising the stakes against liars and splitting timelines so people follow the timeline they believe is clever
Z
19:15
Zack
In reply to this message
https://twitter.com/zack_bitcoin/status/1331185562259165184?s=20 here I commented on a relevant tweet he made.
ᴅ
19:15
ᴅɪꜱᴛʙɪᴛ ⛓️💵
In reply to this message
Ah yeah you have mentioned this to me before. Don't worry I will get around to reading all of your documentation soon so I will understand how this works then 👌
Z
19:16
Zack
In reply to this message
Paul thinks it will cause the blockchain to keep forking, and there will be lots of tiny low-value forks.
IS
19:16
Ilmu Somebody
sure, but that's fine! what you've invented is a political system with a nice equilibrium
19:17
delusion is something people are entitled to
Z
19:17
Zack
I think the Amoveo oracle, or something very similar, will end up being the most popular design
ᴅ
Z
19:17
Zack
PoW gives strong incentives to have one chain
19:18
In reply to this message
we had an extended discussion in this telegram channel before. I have it recorded in the docs somewhere.
19:18
it was over a year ago I think
IS
ᴅ
19:18
ᴅɪꜱᴛʙɪᴛ ⛓️💵
I find I have so many questions for Paul about his ideas but as anyone reading dcinsiders closely over the last few days would have noticed, I seem to be very bad at explaining them well enough to make him want to answer them... Hopefully this situation improves over time as I find him to be interesting to talk to
Z
19:19
Zack
If there were two forks using the same ASIC for example, the smaller side is vulerable to attacks from miners on the bigger side
19:20
In reply to this message
Paul has never much liked live discussions. He prefers thinking longer, and then making blog posts.
IS
19:21
Ilmu Somebody
okay slight off-topic, but for the past few years I have been partial to the kind of economic theory jane jacobs has: i.e. currencies are load-balancers and having large national currencies creates stagflation in counties within the economic area
ᴅ
19:22
ᴅɪꜱᴛʙɪᴛ ⛓️💵
In reply to this message
Hmm yeah I think I just need to gain a better background in all of the pre req econ that is necessary to better understand paul's articles and then I will have less need to discuss them with him
Z
IS
19:22
Ilmu Somebody
paul believes the euro-area is a "natural economic zone" or whatever the term was (I don't have it open on this pc) but I believe it was essentially a reactionary move to create a big enough economy so that the dollar wouldn't just eat the world as the single source of truth
19:24
from my perspective; the goal is to decentralize: i.e. break the world up into as many (hyper-)locally governed regions as makes sense and have them act as individuals in second-order politics (so they can compose in different ways to debate cross-regional concerns)
19:24
I have basically from the beginning been opposed to the blockchain vision of a top-down currency
19:25
what bothers me about it is that the security of the network is the most important thing in the world (from the perspective of the network) so survival of humanity or of the planet are second order concerns
19:26
the way I've been seeing it is that as you compose small currencies (via exchange rates or similar contracts) the global currency would be emergent
ᴅ
19:26
ᴅɪꜱᴛʙɪᴛ ⛓️💵
In reply to this message
The network doesn't have a perspective, it is made up of individuals who have perspectives.
IS
19:27
Ilmu Somebody
In reply to this message
sure but the externalities (such as energy use) are not accounted for within the system, it only cares about hashrates
Z
19:28
Zack
Money is for solving the coincidence of wants problem, an entirely human concern.
Security of the money system is important as long as it is helping to solve the coincidence of wants more efficiently.
Security of the money system is important as long as it is helping to solve the coincidence of wants more efficiently.
19:29
Using lots of energy helps to solve coincidence of wants more efficiently, because converting value from energy form into currency form is the most efficient way to grow the supply of currency, so we can have more liquidity to solve more coincidence of wants problems.
IS
19:29
Ilmu Somebody
idk, have you read "seeing like a state"? it's kind of like trying to come up with a information technology that gives us enough conceptual space to model the natural systems without sacrificing the consistency of globalization
19:29
(btw that last comment was not yet an answer to what you said zack)
19:32
okay, so you want to solve the coincidence of wants by making a bunch of markets within amoveo for the different "wants" but the more we rely on this the more veo is worth, correct?
Z
19:33
Zack
derivatives are for solving coincidence of wants between different moments in time
19:33
Amoveo is a platform for derivatives
19:33
bitcoin already solves coincidence of wants in the same moment in time pretty well.
Amoveo is focused on solving coincidence of wants between different moments in time.
Amoveo is focused on solving coincidence of wants between different moments in time.
IS
19:40
Ilmu Somebody
I guess the distinction I'm trying to draw is between a medium of exchange and an expression of power
19:47
ah I don't know enough jargon, I'll try again after an indeterminate amount of time
21 January 2021
J
01:03
Josh
What do you mean by an expression of power?
01:04
Individual people having lots of money seems to be an inefficiency in this "coincidence of wants between different moments in time".
01:05
Money flows are getting stuck at different points in the graph because there's lots of friction in the system.
son burn invited son burn
IS
03:08
Ilmu Somebody
I mean political power, the one you gain by people respecting you
03:08
IF you have a great economic system then these things are supposed to coincide (meritocracy)
Edit: but I don't believe it should..
Edit: but I don't believe it should..
03:10
I see a shadow of what my idea is about in the split between bitcoin and truthcoin (where the oracles are the priests and the plutocrats own the bitcoin)
03:10
but I want to find a better way to communicate this structure before I try to defend this argument
(I have an outdated and rambly exposition of this kind of thing on rishi (dot) is, it's called 'digital identity')
(I have an outdated and rambly exposition of this kind of thing on rishi (dot) is, it's called 'digital identity')
Z
J
04:49
Josh
I was thinking about a more structural reason. If we don't have efficiency ways to solve coincidence of wants over time, some individual people will wind up earning a lot of money.
04:50
But things would probably be more efficient if the system was better at moving resources around where needed and we didn't need one person to make all the right decisions.
04:50
I'm comparing american capitalism and Elon Musk versus futarchy.
IS
05:00
Ilmu Somebody
In reply to this message
Yes. Having a secure store of value is the whole problem imo
05:01
It's why I don't like central currencies
Z
05:01
Zack
In reply to this message
If some people are better at predicting the future needs of humanity, then the system is more efficient if they have more of the money. because they can use more money for their good predictions.
IS
05:02
Ilmu Somebody
How do you slash them when they defect?
J
05:03
Josh
I agree with that but maybe that person is just better than the current broken system and in a better system the resource would be spent better by more people
05:03
resources
05:04
At some point even Elon Musk doesn't have enough CPU cycles to spend all that money well
mx
05:04
mr x
blockchain itself should spend money using hashrate futarchy
Z
05:09
Zack
In reply to this message
if you make bad bets in a prediction market, you lose your money. Losing bets don't get paid.
05:10
In reply to this message
People tend to specialize.
Someone who is really good at predicting the weather could make good bets for petroleum consumption, and certain crops.
Other people who are good at international politics might do better at predicting impact from changes in trade policy.
Maybe a biologist could predict when the banana fungus will wipe out the bananas.
Someone who is really good at predicting the weather could make good bets for petroleum consumption, and certain crops.
Other people who are good at international politics might do better at predicting impact from changes in trade policy.
Maybe a biologist could predict when the banana fungus will wipe out the bananas.
IS
05:10
Ilmu Somebody
I think I don't understand prediction markets then, how do you avoid self-fulfilling prophecies that are self fulfilling because the one who made it is so influencial that they can move the room.
Z
05:11
Zack
In reply to this message
that isn't necessarily a bad thing.
Like, that is how we avoid assassination markets.
Can you give a more specific example of when you think it could be a problem?
Like, that is how we avoid assassination markets.
Can you give a more specific example of when you think it could be a problem?
IS
05:12
Ilmu Somebody
Hitler?
05:12
Idk
05:13
The X are the problem, let's kill them and it works cause the problem was overpopulation
05:13
Sth like that
Z
05:14
Zack
this example isn't clear enough for me to understand
05:14
how can you bet on whether "X are the problem"?
05:14
You can bet on things that eventually become common knowledge. you can bet on the future price of bread. you can bet on the future exchange rate of currencies.
05:15
you can bet on whether an event will come to pass by a certain date
IS
05:15
Ilmu Somebody
But you are also making people work to have the bet resolve in whatever manner they'd prefer
05:16
So once that culture is in place then the prophets wield quite a lot of power and influence
Z
05:16
Zack
In reply to this message
that is how we use amoveo to fund public goods as well. I don't see how this could be a problem.
IS
05:17
Ilmu Somebody
'I bet all the jews will be dead in 10 years'
05:17
And I put up the wealth of several lifetimes
05:17
And all the influence of being the number one prophet
Z
05:18
Zack
I guess Jews would be willing to bet against that, since if they lose, they are dead anyway.
05:18
but if they survive, they get to take hitler's money. so that would be nice
IS
05:19
Ilmu Somebody
I think what I am saying hinges on people caring more about following a leader than the economic dimension
05:20
So it doesn't really apply, out of scope (for this project)
05:22
The cognitive capacity argument also falls flat by the same logic, you'd end up with something closer to top poker / chess players ruling these markets rather than a hitler
05:25
How valuable would veo be if all the world uses this platform for trading risk? Would it ever be risky to hold veo for the already rich? Just by changing the governance params to thin out the people who aren't playing?
Z
05:27
Zack
The global market for derivatives is pretty big
mx
05:27
mr x
How do you know who arent playing?
05:27
What does that mean?
IS
05:28
Ilmu Somebody
You wouldn't target them I'm saying you'd just start printing a bunch of veo to thin out rich complacent hodlers
05:28
I don't see how it could be risky to hold it otherwise
mx
05:28
mr x
How do you know there are rich complacent holders?
IS
05:28
Ilmu Somebody
Idk, I am just expressing a georgist concern
Z
05:29
Zack
If futarchy says that increasing the block reward would be beneficial for the price of veo, then the block reward would increase
05:29
but if the price of veo is increasing, that is beneficial for veo holders
IS
05:30
Ilmu Somebody
So you have no way to diffuse plutarchy
mx
Z
05:31
Zack
In reply to this message
someone must be paying for the new veo to be produced from miners.
05:31
if the rich get richer, but they also end up holding a smaller fraction of the whole pie, that isn't exactly a plutocracy.
05:32
it's more like a rising tide lifts all boats situation
IS
05:33
Ilmu Somebody
Sure, if the rich are getting richer slower than everyone else then that would worl
05:33
Work*
05:33
But it would need to be accessible to start mining
Z
05:34
Zack
Mining is done by specialists, but they need to sell almost all of their new veo to pay their costs of hardware and electricity.
Someone else is buying the newly mined veo.
Someone else is buying the newly mined veo.
IS
05:35
Ilmu Somebody
Ah yea now that you mention it you had developers also included in rewards and the on platform futarchy can play with the params so mining is not the only distribution mechanism
05:35
Yeah this project is definitely interesting
05:36
Gonna mine it 😁
M
07:03
Micka
is there some kind of roadmap for veo?
I
07:04
Instinct
Z
Zack 14.01.2021 01:31:31
There is a todo list in the docs repository
I
07:04
Instinct
Z
Zack 14.01.2021 01:32:31
M
07:06
Micka
thanks. But I was looking for some kind of date goals too
Z
07:46
Zack
Right now im working on the perpetual suncurrency hard update.
Mixing the amm swap tool with the limit order tool is coming up soon as well.
After that, i guess we will reflect on how things are going and make the next plan. Maybe it will be time to do some marketing.
Mixing the amm swap tool with the limit order tool is coming up soon as well.
After that, i guess we will reflect on how things are going and make the next plan. Maybe it will be time to do some marketing.
J
08:07
Josh
In reply to this message
Another good thing about betting is it makes things explicit. I think most germans didn't really want to kill all the jews, they just wanted their jobs and self respect back. I don't think they'd be willing to put down a lot of money betting for jews to die. I also really doubt many people believed something like that would actually happen.
EA
14:31
Eric Arsenault
Making a nice interface would be good to do before marketing
Z
15:57
Zack
In reply to this message
https://twitter.com/jadler0/status/1352038344717447170?s=19
John adler has studied optimistic roll up a lot. He wrote some nice blog posts about it, and he was in a podcast that explained well.
He also compares it to zkroll up in interesting ways.
John adler has studied optimistic roll up a lot. He wrote some nice blog posts about it, and he was in a podcast that explained well.
He also compares it to zkroll up in interesting ways.
ᴅ
16:07
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Thanks so much, I highly value having another good information source
16:07
Can you link to his podcast/blog on rollups?
16:07
I can't seem to find either readily available on his Twitter page
Z
ᴅ
16:15
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Damn that blog looks great
16:16
Just added all of the articles to my reading list
16:16
It is a good sign when I find every article on a blog interesting enough to add to my reading list 😎
Z
16:21
Zack
My personal experience with building blockchains has been that reading and writing to consensus state is the bottleneck by far. Thats why amoveo had the memoryless full node design before "memoryless full node" was a named thing.
It was a hard decision to make amoveo txs 10x bigger, but today it seems like this kind of design is the normal perspective of how scaling needs to work.
So it was refreshing to read something about blockchain scalability that focuses on solving the actual bottleneck.
It was a hard decision to make amoveo txs 10x bigger, but today it seems like this kind of design is the normal perspective of how scaling needs to work.
So it was refreshing to read something about blockchain scalability that focuses on solving the actual bottleneck.
ᴅ
16:22
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Btw @zack_amoveo have you considered mirroring/replicating your amoveo github docs repo onto a medium or other rss compatible feed? I'd definitely subscribe to it via rss but github doesn't support that for repositories
Z
16:24
Zack
Can rss connect to the twitter feed?
ᴅ
16:43
ᴅɪꜱᴛʙɪᴛ ⛓️💵
What do you mean by that?
Z
16:44
Zack
I post links to twitter when a blog post is more polished
ᴅ
17:39
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Ah I see
17:39
Is that automated?
x
17:41
x
i think iftttt/zapier... can do that
ᴅ
17:54
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Hmm yeah probably
J
J
19:11
Josh
This is the key insight of rollups: "It turns out that history is much, much cheaper than state (as in, orders of magnitude cheaper)."
J
19:30
Josh
This is also interesting
ᴅ
IS
22:01
Ilmu Somebody
the 1000 member idea is a good one.
Z
IS
22:38
ahh maybe it's telegram making a distinction between a super group (or whatever it's called) and a group
22:38
but I think 1000 is a good upper limit for a discussion group tbh
22 January 2021
MF
00:22
Mr Flintstone
the key for the LP for a lot of these markets is in bricked hardware now so free money ahoy:
159.89.87.58:8080/contracts_list.html
159.89.87.58:8080/contracts_list.html
Z
MF
00:22
Mr Flintstone
but you would need to wait for oracle settlement
00:23
yeah RIP that VEO
Z
00:23
Zack
you can get the hard drive out of computers and read it
MF
00:23
Mr Flintstone
one day it will perhaps be worth it
Z
00:25
Zack
I have this external hard drive.
I opened it up, and it was a totally standard laptop hard drive inside. I can swap it out for other laptop hard drives and plug them in like a usb.
Another strategy, you can put that hard drive into a working computer, and boot up from a linux flash drive, then read whatever from that hard drive.
I opened it up, and it was a totally standard laptop hard drive inside. I can swap it out for other laptop hard drives and plug them in like a usb.
Another strategy, you can put that hard drive into a working computer, and boot up from a linux flash drive, then read whatever from that hard drive.
00:26
it is pretty easy to make linux flash drives.
unetbootin is available for windows/mac/linux.
ubuntu is easy to download.
unetbootin is available for windows/mac/linux.
ubuntu is easy to download.
MF
00:27
Mr Flintstone
yeah, thanks for the suggestions. its not a huge deal, just wanted to let ppl know lol
Z
01:01
Zack
I undid a few commits to the master branch on github so that we could keep the perpetual stablecoin stuff in it's own branch for now, and we don't pollute the master branch with doing and undoing these changes.
If you try to do a
You can get back onto the current master branch like this:
Alternatively, you can always make a fresh clone of amoveo in a new empty directory with
Thank you for your patience.
If you try to do a
git pull, it may fail.You can get back onto the current master branch like this:
git fetch origin
git reset --hard origin/master
Alternatively, you can always make a fresh clone of amoveo in a new empty directory with
git clone https://github.com/zack-bitcoin/amoveo.git .
Thank you for your patience.
M
01:01
Minieep21
someone getting fomo from reading the past 24hrs of chat 😂
R
02:29
Rom
Zack spoke about marketing, hope is back 👍. It will also be necessary to review the aesthetics of the platform and the website
x
02:31
x
i believe it just need a chrome extention that wrap the light node in it, design is not even needed. just like a shortcurt ,after it's installed on a browser,
02:40
or maybe not chrome extention, but a standalone app which should also be easy
mx
02:40
mr x
electron or chrome extension
x
02:41
x
right, that way, we also never rely on zack's IP
IS
b
19:36
bjorn
Hi, where can we buy or sell VEO?
b
19:39
bjorn
Thx
Adrien invited Adrien
23 January 2021
J
01:01
This post on rollups is really good
EA
01:33
Eric Arsenault
Statebox?
J
MF
02:03
Mr Flintstone
any good summary of this concept without attending a course? not exactly sure what this means
J
02:35
Josh
It's about writing programs using a construct called Petri nets which are like state machines that can also do operations in parallel.
02:36
they map that to mathematical concepts via category theory.
Z
03:01
Zack
In reply to this message
Their analysis of sidechains is very interesting. The conclusions they come to about data availability could be really important, and I haven't seen it explained before.
They make a good case for UTXO based optimistic rollup.
They make a good case for UTXO based optimistic rollup.
J
03:04
Josh
"However, accounts-based execution models will have much lower scalability than Ethereum today.
This is because fraud proofs for accounts-based execution models require serializing the state after every transaction (or greatly increasing the size of transactions, rendering it strictly inferior to simply using the UTXO data model to begin with)."
This is because fraud proofs for accounts-based execution models require serializing the state after every transaction (or greatly increasing the size of transactions, rendering it strictly inferior to simply using the UTXO data model to begin with)."
03:05
The "strictly inferior" part is interesting. Does this apply to Amoveo?
IS
03:18
Ilmu Somebody
In reply to this message
Visual programming language for smart contracts to enable anyone to use the internet as a counterparty
Z
03:28
Zack
Amoveo is using memoryless full node scaling.
It isn't based on fraud proofs.
Each block includes with it the merkle proofs for all the consensus state that you need to know about to verify that block. So that is a kind of validity proof.
It is similar to rollups in that full nodes don't need to read or write any state to verify the blocks. And it is similar to rollup in that some archival node is keeping a linear record of the entire history which can be used for re-syncs. We both take advantage of the fact that recording a linear history without doing random read/writes to that history is orders of magnitude faster than a database that is being read/written to.
Keep in mind that the pure UTXO model only works for owning and spending tokens. We cant put uniswap markets, or oracles into it. we can't process limit orders. We cant have turing complete contracts based on results from multiple oracles. We can't enforce single-price-batch off-chain markets.
Yes, Amoveo does have to serialize the final state of accounts/markets/oracles/etc after updating them in a block, in order to calculate the new state root after processing that block. But this serialization doesn't involve reading or writing to the hard drive, so it is very fast.
I think the memoryless full node model has some decentralization advantage over optimistic rollup UTXO.
Amoveo can process the blocks in any order. including in parallel, or in reverse order. This makes it a lot cheaper to launch a new full node. It makes it a lot cheaper to set up a mining pool, and quickly recover if your full node crashes for any reason. You can start from an empty linux server, install Amoveo, and start mining within a couple minutes.
I guess you could think of Amoveo as similar to zk-rollups, because it is based on validity proofs instead of fraud proofs.
But we aren't zero-knowledge. It is a full merkle proof every time. So we aren't using any weird unknown cryptography.
They make a good case that optimistic rollup with UTXO could be the best we have so far for doing normal payments, but I think it is not the best solution for a platform for financial derivatives.
It isn't based on fraud proofs.
Each block includes with it the merkle proofs for all the consensus state that you need to know about to verify that block. So that is a kind of validity proof.
It is similar to rollups in that full nodes don't need to read or write any state to verify the blocks. And it is similar to rollup in that some archival node is keeping a linear record of the entire history which can be used for re-syncs. We both take advantage of the fact that recording a linear history without doing random read/writes to that history is orders of magnitude faster than a database that is being read/written to.
Keep in mind that the pure UTXO model only works for owning and spending tokens. We cant put uniswap markets, or oracles into it. we can't process limit orders. We cant have turing complete contracts based on results from multiple oracles. We can't enforce single-price-batch off-chain markets.
Yes, Amoveo does have to serialize the final state of accounts/markets/oracles/etc after updating them in a block, in order to calculate the new state root after processing that block. But this serialization doesn't involve reading or writing to the hard drive, so it is very fast.
I think the memoryless full node model has some decentralization advantage over optimistic rollup UTXO.
Amoveo can process the blocks in any order. including in parallel, or in reverse order. This makes it a lot cheaper to launch a new full node. It makes it a lot cheaper to set up a mining pool, and quickly recover if your full node crashes for any reason. You can start from an empty linux server, install Amoveo, and start mining within a couple minutes.
I guess you could think of Amoveo as similar to zk-rollups, because it is based on validity proofs instead of fraud proofs.
But we aren't zero-knowledge. It is a full merkle proof every time. So we aren't using any weird unknown cryptography.
They make a good case that optimistic rollup with UTXO could be the best we have so far for doing normal payments, but I think it is not the best solution for a platform for financial derivatives.
03:37
In reply to this message
It looks like statebox is trying to solve a problem that no one has.
I see Ethereum as a playground to experiment with what is possible in blockchains.
Because it is turing complete everywhere, it is extremely flexible. but that flexibility comes with slower speeds, higher fees, and lots of safety concerns.
But I think once we realize which kinds of contracts people actually want to use, we will prefer more limited kinds of blockchains that just do what people want, and nothing more.
Reducing capabilities gives us speed, lower fees, and better security guarantees.
I think the statebox guys are working with the broken assumption that turing completeness everywhere is how it will always be done, and they are trying to build tools to make that world cheaper and safer. They think many people will be writing turing complete contracts all the time.
My expectation is that there will be a very small handful of turing complete contracts that are almost never changed. That these contracts will be so sandboxed from the rest of the system, that it is easy to verify security.
That when individuals want to customize what they are doing, it will either be by changing the configuration of the handful of contracts that they use, or by mixing together different tx types inside of atomic flash-mints.
I see Ethereum as a playground to experiment with what is possible in blockchains.
Because it is turing complete everywhere, it is extremely flexible. but that flexibility comes with slower speeds, higher fees, and lots of safety concerns.
But I think once we realize which kinds of contracts people actually want to use, we will prefer more limited kinds of blockchains that just do what people want, and nothing more.
Reducing capabilities gives us speed, lower fees, and better security guarantees.
I think the statebox guys are working with the broken assumption that turing completeness everywhere is how it will always be done, and they are trying to build tools to make that world cheaper and safer. They think many people will be writing turing complete contracts all the time.
My expectation is that there will be a very small handful of turing complete contracts that are almost never changed. That these contracts will be so sandboxed from the rest of the system, that it is easy to verify security.
That when individuals want to customize what they are doing, it will either be by changing the configuration of the handful of contracts that they use, or by mixing together different tx types inside of atomic flash-mints.
IS
03:43
Ilmu Somebody
I agree w.r.t. non-turing completeness
03:43
But statebox is based on petri nets which are not turing complete
03:44
Rather they are the model that emerges when you have communicating state machines
03:44
(so many states can be inhabited at once)
Z
03:48
Zack
This is about the most complicated contract that I can imagine people would want:
I expect contracts like this to make up less than 0.1% of the txs, and to never be a bottleneck.
I don't see a way to use parallel execution to meaningfully change the speed that this contract would execute.
Look up the result of oracle X.
verify that the oracle question matches the required format.
verify that the oracle returned "True".
Parse a number from the oracle question.
Split up the value from this contract based on that number that was embedded in that oracle question.
I expect contracts like this to make up less than 0.1% of the txs, and to never be a bottleneck.
I don't see a way to use parallel execution to meaningfully change the speed that this contract would execute.
03:49
https://github.com/zack-bitcoin/amoveo/blob/perpetual_subcurrencies/apps/amoveo_core/priv/scalar.fs
Here is that contract written in Chalang, the language we are currently using for Amoveo.
Here is that contract written in Chalang, the language we are currently using for Amoveo.
IS
03:52
Ilmu Somebody
The state machines that are communicating could be on different computers and across trust boundaries
Z
03:52
Zack
In reply to this message
so, this isn't related to cryptocurrency then?
This is a place to discuss Amoveo and related concepts.
This is a place to discuss Amoveo and related concepts.
IS
03:52
Ilmu Somebody
The petri net can be designed visually (kind of like a state machine) but the regions different people are responsible for denoted and then you compile this into something like these specialized blockchains
03:53
Ah sorry I don't mean to go offtopic
03:54
It's kinda hard to nail down statebox, it's about a whole new approach to computing
03:54
But whatever
J
04:42
Josh
Yeah, I was just pointing out that there are some guys from the statebox channel coming in.
IS
05:14
Ilmu Somebody
In reply to this message
I suppose the reason you'd never need a more complicated contract is because you place all of the evaluation burden on the oracle (which is why I think this design is very clever).
Do you then see some idea like having layers of "miners" on top of the oracles to automate some kinds of oracle bets and allow for constructing protocols for interacting with business conducted off chain?
I have been thinking about the types of miners that would be possible to build on this platform since I started to understand the idea, if you have any documents to link to or thoughts on this then I'd like to know more.
Do you then see some idea like having layers of "miners" on top of the oracles to automate some kinds of oracle bets and allow for constructing protocols for interacting with business conducted off chain?
I have been thinking about the types of miners that would be possible to build on this platform since I started to understand the idea, if you have any documents to link to or thoughts on this then I'd like to know more.
ᴅ
Z
05:26
Zack
In reply to this message
The design for perpetual stablecoins i am working on, it involves periodic auctions to swap out the collateral for the perpetual.
Someone needs to be holding the inverse of the perpetual stablecoin, which will probably end up being 5x leveraged veo.
And they may want a system to automatically participate in these auctions and cash out their expired contracts.
Maybe that is the kind of thing you are thinking of by "miner".
Someone needs to be holding the inverse of the perpetual stablecoin, which will probably end up being 5x leveraged veo.
And they may want a system to automatically participate in these auctions and cash out their expired contracts.
Maybe that is the kind of thing you are thinking of by "miner".
05:27
Yeah, you are right. We try to offload computational complexity to the oracle when possible.
IS
05:35
Ilmu Somebody
Okay so you can keep confidence in the stability of the coin by challenging the truthfulness periodically.
I was thinking how to build something at web-speed and have bandwidth, insurance (against spearphishing via request responses) and social reputation .. all of which is backed by a series of economical games being played on the chain (you put up collateral when you service others and they pay - bridge at restaurant pattern - you can apply it recursively for progressively bigger claims)
A miner would be sever providing bandwidth in the social network
I was thinking how to build something at web-speed and have bandwidth, insurance (against spearphishing via request responses) and social reputation .. all of which is backed by a series of economical games being played on the chain (you put up collateral when you service others and they pay - bridge at restaurant pattern - you can apply it recursively for progressively bigger claims)
A miner would be sever providing bandwidth in the social network
05:38
In reply to this message
But yes the automated participation in auctions is precisely what I mean
Z
IS
05:50
Ilmu Somebody
Yes? And I'm talking about automated businesses that need to be valued automatically based on how well they behave in maintaining the internet
06:04
Names,money and oracles need nakomoto consensus, I'm thinking how to do dns+routing using your oracles.
06:04
In a dynamic/open p2p network
Z
06:09
Zack
You can bet on the value of a publicly traded business by making a derivative to bet on the price of its shares.
Dns is not an application of derivatives.
Dns is not an application of derivatives.
IS
06:11
Ilmu Somebody
Yes sorry, to clarify; the dns part is fully off chain and something I am thinking about on my own, the betting structures is what I am interested in and seems I got my answers; bridge and restaurant, stablecoin..
06:13
Trust is the application of derivatives (as I see it) and 'how?' is what I'm still wrapping my head arouns.
06:14
(you see, I am quite interested in building a web of trust)
Z
06:14
Zack
A "derivative" is when you bet on some information that will become publicly available later.
In the context of blockchain, the word "trust" is a synonym for "insecure".
We are trying to build trustless products.
In the context of blockchain, the word "trust" is a synonym for "insecure".
We are trying to build trustless products.
06:15
Betting on the result of a football game is a kind of derivative. Betting on the price of gold is a kind of derivative.
IS
06:17
Ilmu Somebody
You are making a way to build products that we can have great confidence in and others will want to build the confidence in their product using them.
Z
06:22
Zack
You could use a prediction market to bet on the odds of whether an airline will have a fatal crash landing in the next year. If the odds are very low, that could give people confidence to fly with that airline.
IS
06:23
Ilmu Somebody
Yes and you can bet on a hash being accessible on IPFS for the next year or that it won't be accessible
06:25
You can bet that there is a digital signature by <key> for some content address
06:25
By some date
06:25
You can bet that there is no bug in amoveo patch <some hash>
Z
06:28
Zack
In reply to this message
I think the way to do this is with futarchy.
We would try to show that the update will have a positive impact on the price.
We would try to show that the update will have a positive impact on the price.
IS
06:29
Ilmu Somebody
The way I see it you can build confidence that the platform is bug free by making a weekly bet, the more users the bigger the bet
06:31
So from both sides, the bets build confidence in where you are and the futarchy leads the way to other places
Z
06:42
Zack
If you know that an update would negatively impact the price of veo, you could profit by betting that it would.
IS
07:24
Ilmu Somebody
so yes it only thinks of deltas when they happen it doesn't look at the long tail of old code
07:24
I see something like a pijul patch based system where you upgrade your confidence in the patches that compose the system based on how big the bets that they are bug free are
07:24
then anyone can submit new patches using the futarchy
07:25
but you'd want to be able to aggregate bets nicely, so you can pick any subset in the set inclusion lattice of patches in a project
07:28
can you track who is betting how much?
ᴅ
08:15
ᴅɪꜱᴛʙɪᴛ ⛓️💵
In reply to this message
Or it gives terrorists an incentive to bet that one will occur and make it so...
08:17
In reply to this message
Wouldn't this be a bad bet since if you bet and you influence the price, the update will not go through so you will lose your capital? Ah nvm I suppose you would also buy shares to hedge against the update not occuring.
This is similar to my other question to sztorc ha
This is similar to my other question to sztorc ha
J
08:19
they'd have to make it look like an accident, which might be hard
ᴅ
08:19
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Yeah true, that would definitely increase the difficulty of the attack
08:19
Not impossible ofc but more difficult
MF
08:24
Mr Flintstone
In reply to this message
markets with lots of liquidity always provide incentives to do horrible things
08:25
any market
Z
08:28
Zack
Markets also give a warning of what will happen, and let us prepare.
MF
08:28
Mr Flintstone
plus, if they buy shares of yes, the probability of the plane crash would increase so conceivably people would not fly
ᴅ
08:29
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Yeah I suppose it depends on how liquid it is and how much capital he wants to put up
A
ᴅ
08:31
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Hasn't Paul written something about how assassination markets aren't likely though? Idk I am yet to read it.
Either way I don't think what we are discussing is a major issue. It will be interesting to see what kind of mitigations are invented in the future to address such questions
Either way I don't think what we are discussing is a major issue. It will be interesting to see what kind of mitigations are invented in the future to address such questions
Z
08:31
Zack
A good terrorist plot probably involves multiple actors.
We can make markets for individual flights, to bribe some of the terrorists to reveal the plot of the rest.
We can make markets for individual flights, to bribe some of the terrorists to reveal the plot of the rest.
08:32
https://en.m.wikipedia.org/wiki/Policy_Analysis_Market
Here is some historical context for using prediction markets to prevent terrorism.
Here is some historical context for using prediction markets to prevent terrorism.
ᴅ
08:34
Do you think pms can be used to fund public goods?
08:34
If so that is extremely interesting
Z
08:34
It is in the truthcoin docs
ᴅ
08:35
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Hmm
08:35
I was thinking even better than dacs
08:35
I know I should read the tc docs, and I am approaching them on my reading list
08:35
(which has 1.3k articles in it ;)
08:35
But before I get to that
Z
08:36
Zack
I wrote about them in amoveos docs too
ᴅ
08:36
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Ah ok cool well I will definitely end up reading both of those
08:38
But before I get to that, may I ask if your and paul's writings discuss the idea of pms funding public goods via a mechanism whereby someone who is able to build a "lighthouse" (even though a lighthouse isn't really a public good) can't attract people willing to fund it, so instead funds it by betting that it won't be built on a pm on whether it will be built
08:38
And then paying for the development with the profit he makes from selling the shares at $1 and buying them at 1c
08:39
If that idea is discussed, which I am sure it is, then I will just wait and read it
Z
ᴅ
08:39
ᴅɪꜱᴛʙɪᴛ ⛓️💵
If not I'd be interested in hearing whether you think it is flawed/makes sense?
Z
08:40
Zack
Who bets that it wont be built?
ᴅ
08:40
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Speculators
Z
08:41
Zack
The usual idea of a dac is that people bet it wont be built as a form of insurance against the possibility that it wont be built.
ᴅ
08:42
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Ofc the price of the yes shares would factor in the probability that someone takes the initiative and goes and builds it by buying yes shares ;)
08:42
Damn there are so many interesting ideas in the pm-idea-universe
Z
08:42
Zack
Dac are optimized vs normal markets to prevent speculators from betting that it will get built and free riding off someone else building it
ᴅ
08:43
ᴅɪꜱᴛʙɪᴛ ⛓️💵
Ah I see ok yeah I need to read more
08:43
Very interesting
Deleted invited Deleted Account
24 January 2021
S
21:13
Sebsebzen
is ExanTech releasing any UI updates anytime soon?
DV
S
21:34
Sebsebzen
Is it due to funding, or waiting for stable release candidate?
21:34
Just interested
25 January 2021
MF
02:18
Mr Flintstone
@denis_voskvitsov how much would need to be in the below contract to build something? just curious
159.89.87.58:8080/contract_explorer.html?cid=fwwL387sbGiDeTfPa73fEcHNdKkoafJdMeYipkboChE=
159.89.87.58:8080/contract_explorer.html?cid=fwwL387sbGiDeTfPa73fEcHNdKkoafJdMeYipkboChE=
DV
02:20
Denis Voskvitsov
I'm afraid I can't say it right now
we're deeply engaged in another blockchain consulting project at the moment
we're deeply engaged in another blockchain consulting project at the moment
B
03:58
Beer
-.-
M
N
R
19:06
Rom
Big dump
R
22:42
Rom
In reply to this message
I don't think it can go any lower... i hope... otherwise the project is dead !
IS
23:23
Ilmu Somebody
the project isn't dead till there is no one maintaining it (devs + miners)
J
IS
23:25
Ilmu Somebody
price has a lot to do with availability I'd think, people just trade where they trade and if they can't access the token from there then it may as well not exist
x
23:29
x
it has been a stablecoin for a few weeks. USD price is fine.
26 January 2021
D
00:00
Devender
Wondering what's the mining cost to price ratio now?
00:01
And POW difficulty change like bitcoin or via futurachy?
Z
00:02
Zack
In reply to this message
we have a custom difficulty adjustment algorithm, it is a lot faster than bitcoin's 2000 block average.
00:02
the block reward and block time are set by governance futarchy.
D
00:03
Devender
👍👍
mx
00:32
mr x
blocktime already governance variable?
Z
00:33
Zack
Yes. We have never changed it.
mx
00:34
mr x
Alright. Does it autoadjust blockrewards with it?
Z
00:35
Zack
Im pretty sure it doesnt. We would need to adjust them both at once.
mx
00:36
mr x
right
MF
00:46
Mr Flintstone
i think there was a plan to be able to change all the rate variables at once
00:46
along with block time
00:46
at some point. like a new governance variable
Z
00:50
Zack
https://github.com/zack-bitcoin/amoveo/blob/master/apps/amoveo_core/src/consensus/txs/coinbase_tx.erl
Looks like we just use the one block reward for now.
No scaling based on blocktime
Looks like we just use the one block reward for now.
No scaling based on blocktime
Z
06:33
Zack
I'm still working on the perpetual stablecoin implementation. It is hard for me to keep track of everything that is going on for actually building the smart contract.
So I decided to go abstract and write some documentation about how exactly Amoveo smart contracts are being created, and I realized there are 4 very independent languages here. 4 turing complete contracts that all sort of feed into each other to enable an Amoveo smart contract.
https://github.com/zack-bitcoin/amoveo-docs/blob/master/basics/smart_contract_creation.md
I think with this model of how smart contracts are being created, it will be easier to create contracts, and know which parts of the computation should happen when and where.
So I decided to go abstract and write some documentation about how exactly Amoveo smart contracts are being created, and I realized there are 4 very independent languages here. 4 turing complete contracts that all sort of feed into each other to enable an Amoveo smart contract.
https://github.com/zack-bitcoin/amoveo-docs/blob/master/basics/smart_contract_creation.md
I think with this model of how smart contracts are being created, it will be easier to create contracts, and know which parts of the computation should happen when and where.
ChonkyPanda invited ChonkyPanda
Unihax0r invited Unihax0r
Higgo invited Higgo
Casper 👀 DeFi Search 🔥 invited Casper 👀 DeFi Search 🔥
banth invited banth
C
15:07
Quick question, is marketing veo on your to do list also? Just wondering
M
C
15:14
ChonkyPanda
Thank you sir
rico invited rico
Deleted invited Deleted Account
IS
23:13
Ilmu Somebody
In reply to this message
You can use the futarchy to ask for money to implement some advertising strategy
D
IS
23:15
Ilmu Somebody
Supposedly it'll get easier with time and afaict amoveo has no competitor yet
Z
23:21
Zack
In reply to this message
Check out the futarchy tab in the wallet.html page of the light node.
D
23:24
Yeah found it.
x
23:32
x
looks great.
27 January 2021
B
00:20
Ben
error 500 somebody killed it 😛
x
00:26
x
probably we should all use offline version : https://github.com/zack-bitcoin/light-node-amoveo/archive/master.zip
Z
B
00:39
Ben
works again!
MF
00:40
Mr Flintstone
In reply to this message
maybe should pre-populate the fields with a block reward futarchy or something. examples make everything easier to understand
Z
00:40
Zack
Always worked. The other link is to a non existent page
MF
00:41
Mr Flintstone
Then it would be easier to substitute for the decision of your choosing i think
Z
00:41
Zack
In reply to this message
Makes sense.
I was also thinking communities other than amoveo could use this futarchy tool to make decisions.
I was also thinking communities other than amoveo could use this futarchy tool to make decisions.
Z
01:31
Zack
what page?
does it show the correct pubkey?
does it show the correct pubkey?
x
01:46
x
it shows now
Z
05:27
so him and I are going to bet on this game
OK
05:29
O K
OH SH-
05:29
gl hf
05:30
How many spectators are in there right now @zack_amoveo?
Z
05:31
Zack
looks like none? this is my first time using this server
05:31
it has a one move per day timer
OK
05:31
O K
Oh, you're playing correspondence. Cool. I was about to pop some popcorn for a live game 🙂
r
05:32
rico
We can do live too
Z
05:43
Zack
welcome rico. he is the guy I am playing against