1 August 2018
AS
08:40
Aizen Sou
In reply to this message
Won't change anything. Zack will not listen to anyone and keeps doing what he thinks he is right.
S
08:41
Subby
Well it's his project
08:41
Fork it if you want it done your way
OK
08:42
O K
In reply to this message
"his" project
08:42
Aizen is a valuable long term community member
08:42
And his complaints have a lot of merit
S
08:43
Subby
No doubt but being a miner it sounds like his issues revolve around how it's no longer profitable for him to mine this
OK
08:43
O K
Yes, the difficulty retargeting algorithm has been a problem
08:43
It's been a problem for people sending transactions
08:44
It's been a problem for Amoveo's growth
08:44
It's not a new issue
g
08:45
goons
Don’t worry guys my AC went out so my rig will be down for a week or so
08:45
Should help w the difficulty
08:45
YW
OK
08:45
O K
I don't see how that helps 😂 @LGoon
I
08:45
Iridescence
What is the plan to deal with the "stuck" difficulty?
08:46
Sorry I have been away
OK
08:46
O K
In reply to this message
We are supposed to be changing to a more ethereum-like re-targeting algo:
MF
08:46
Mr Flintstone
Zack implemented a range targeting diff adj algo that retargets every block if the exponential weighted block time is outside of that range
I
08:46
Iridescence
I see
08:46
So we will hard fork to the new algo
MF
08:46
Mr Flintstone
I don’t think a fork for it has been merged yet
AS
08:47
Aizen Sou
In reply to this message
Yes, over two months constant 40% hashrates. Pay that from ur own pocket like me and then u have the right to criticise me, idiot.
P
08:48
Peter
In reply to this message
+1
I
08:49
Iridescence
I appreciate the hashpower that miners give to Veo, but mining at a a loss is not sustainable
AS
08:49
Aizen Sou
Every profit loving miners fucker did leave immediately after the difficulty raised, yet I was still staying.
I
08:50
Iridescence
Aizen are you pushing for Zack to merge the new difficulty retargeting code as soon as possible?
OK
08:50
O K
I'd like to see that too @Iridescence
08:50
To me, the worst is that we switch back
AS
08:50
Aizen Sou
And do u even know who created the Amoveo current icon ? It's from us, me and my friend @sebsebzen.
08:51
In reply to this message
I did ask Zack everytime but he just ignored me.
OK
08:51
O K
In reply to this message
No one who has been a part of this community for any length of time is questioning your contributions
+
08:51
++
In reply to this message
I’m new here so don’t understand all that happened but why would there be any delay in decreasing the difficulty of the algo? Is there any incentive not to decrease the difficulty ?
OK
08:51
O K
In reply to this message
We aren't looking to decrease the difficulty directly
I
08:52
Iridescence
Isn't Zack in favour of changing the diff algo too?
OK
08:52
O K
We are looking to change the algorithm that calculates the difficulty
T
08:52
Topab
In reply to this message
By the person that reports
AS
08:53
Aizen Sou
In reply to this message
Thanks for ur kind words, @potat_o . That idiot @subnoize kinda annoyed me. Sorry for my rants again.
I
08:53
Iridescence
I know that Zack is targeting 10 minute blocks, and so if it requires an algo change I'm sure Zack will want to do it
+
08:53
++
In reply to this message
Does it matter how it gets done ? Seems like it is in everyone’s interest to make sure diff algo is very responsive to changes in hash power. Or am I missing something?
I
08:54
Iridescence
In reply to this message
If we keep the same algorithm we will probably end up in the same situation
+
08:55
++
In reply to this message
Again, if not annoying for others can someone catch me up on why isn’t better to have even faster confirms ? Why not copy ethereum or do even faster?
OK
08:56
O K
In reply to this message
It is bad for decentralization
I
08:56
Iridescence
In reply to this message
If I remember correctly, Zack wanted to be as close as Bitcoin as possible to be conservative
OK
08:56
O K
10 min blocks are perfect
08:56
In reply to this message
If it takes 30 seconds to propagate a new block to every mining node
08:56
30 seconds out of a 1 minute block time is a huge handicap for a smaller pool
08:57
30 seconds out of 10 minutes is much less
08:57
But regardless, that's not really the issue at hand
08:57
It's just further conflating the current issue
+
08:58
++
In reply to this message
I heard 44% of bitcoin hashing power is controlled by 1 pool. If they didn’t stay decentralized, why would Amoveo?
08:59
Isn’t ethereum more decentralized ?
OK
09:00
O K
In reply to this message
Sorry, but I feel this is just distracting us from the real issue here
+
09:01
++
In reply to this message
No worries
Z
09:01
Zack
If we do this upgrade to make the difficulty change faster, it will be the most high risk update we have done so far.

It would be better if someone other than me reviewed it.
OK
09:02
O K
Thank you for being candid
09:05
Zack Can you link me to the code please?
[
09:05
[Riki]
The amoveo community has people capable and willing to review.
Z
09:05
Zack
OK
09:05
O K
Thank you
Z
09:05
Zack
I am using the "retarget" branch for this
Aries invited Deleted Account
09:28
Deleted Account
oi m8
Z
09:28
Zack
m8
I
10:07
Iridescence
u wot
10:24
????
10:24
not just 0
10:24
I am trying to udnerstand how the fork works in the code
10:26
You could just write get(1) -> common(4200, 0);
10:26
etc
Z
10:26
Zack
So I can easily change all those numbers at the same time.
sd
10:32
steel dan
Thank you
T
10:46
Topab
I am a supporter of amoveo and believe in its long term success. I am contributing with some hashes. The possibilities to build interesting stuff on top are in my opinion huge. I am not a developer but thinking the way to engage some I know to build. One thing it concerns for the long term is the way the new code is implemented, it seems that Zack has the last word, is there a plan to little by little have other devs involved in decission making? Something like BIP in bitcoin? I do not doubt Zack is very capable but only one person making changes in the protocol seems not sustainable in the long term. Zack bring issues and are discussed in telegram/discord, considering the community opinions which is great but I wonder if a more standarized exists or is planned for the future
Deleted invited Deleted Account
13:28
Deleted Account
Aizen Zack i am quite busy these days, so I am not able to review the change right away, I read it and its too big a change to comment without spending a bit more time. The paradox that we are seeing here is that just now, Amoveo is moving faster than a team (Augur/Æ). This is probably because @zack has been free to implement stuff without discussions and disagreements. And also because Zacks code style is minimalistic and to the point (in particular wrt UI). I feel your pain Aizen as I am mining too, but my belief is that Veo will become more valuable once difficulty stabilizes. The new algo is nice, but we cant rush these things at this point. Once price rises difficulty will rise as well, so mining in the future might not be more profitable wrt VEO mined
13:31
It seems to me that Zack is listening to the community a lot. His comment about mining at loss is correct also, from an economic market point of view. Its a shame if you are offended by hearing the truth so bluntly
13:32
As a supporter of Amoveo I think it would be a shame if you left our community. If you are providing a large sum of hashpower, that has surely helped the project advance
13:47
Deleted Account
@pgonza it seems to me that as the projects matures, changes will become more and more difficult. At some stage we could have a more costly system to oversee changes, like BIP, but I dont believe we have progressed enough yet. If we did that now, Amoveo would probably freeze in its current state
Sergey Troshin invited Sergey Troshin
T
14:21
Topab
It makes sense
Orhan invited Orhan
Shaun invited Shaun
+
22:45
++
In reply to this message
Is the main reason for not changing directly is that it is a major change and could result in things “breaking”?
OK
23:22
O K
In reply to this message
It's because it's a bandaid, not a long term solution
+
23:31
++
In reply to this message
Is changing the algo that calculates the difficulty the bandaid while changing the actual diff algo the long term solution?

And is the main impediment to changing the diff algo is that it is hard to do and may break thing?
OK
23:33
O K
In reply to this message
Changing the difficulty by hand, when it's normally calculated by the algorithm, is a bandaid. It only helps us one time each time we do it.

Changing the algorithm might help prevent this from happening again in the future. Changing the algorithm is risky, which is why we are trying to be cautious
+
23:35
++
In reply to this message
Got it.
Thx.
OK
23:35
O K
🙂
+
23:36
++
In reply to this message
Isn’t there a 20% allocation to a Dev fund that could be used to facilitate more significant and risky changes like this?
OK
23:37
O K
That money is 1) locked and 2) belongs to Zack -- it's his choice how that Veo will be spent

The coding is complete, some of us are reviewing the code or having people review it
23:37
It is a little less than 17%
2 August 2018
00:10
Deleted Account
wtb 30 veo @ 180 usd. pm
DY
00:13
Demi Yilmaz
In reply to this message
If you want to do OTC trade please visit: https://discord.gg/xJQcVaT
#trading channel.

If you'd like to buy sell from exchange visit: https://amoveo.exchange/
Z
00:37
Zack
The new retargetting system as 2 major sub-systems.
1) along with each header, we store the exponential weighted average of the hashrate.
2) using the exponential weighted average of the hashrate, we estimate the block time, and calculate the difficulty for the next block.

I think doing the exponential weighted average of the hashrate could be a mistake.
The problem is that if you put a timestamp on a block to say it was mined 1/10th of a second after the previous block, then it makes the hashrate for that block look massive which can make the difficulty spike.

To avoid this problem, if the hashrate for a block is more than 2x bigger than the exponential weighted average, we round it down to only 2x bigger than the exponential weighted average.

This 2x limitation means the difficulty rises much slower than it could. It takes about 80 blocks instead of 20 blocks for the difficulty to rise after the hashrate increases substantially.

Instead of doing the exponential weighted average of the hashrate, maybe it would be better to do the exponential weighted average of the block-time?
Then the challenge is that the block-time means something different when the difficulty changes.

I tried keeping track of inverse-hashrate instead of hashrate, but the difficulty moved too slowly when I would retarget.
AK
00:50
A K
All other major Blockchains are susceptible to the same timestamp attack, right?
00:50
Maybe we can live with it, too?
+
01:02
++
In reply to this message
Thx
Z
01:08
Zack
The hashrate today is the same as yesterday.
I guess Aizen must have lied about controlling 40% of the hashpower. I think that he only comes here to talk bad about Amoveo.
OK
01:09
O K
In reply to this message
No need to be inflammatory. That's the opposite of good community management. On the contrary several of us actually put effort into keeping Aizen around
01:10
So rather than a lie, you might be seeing the fruits of other's hard work in community management
Z
01:10
Zack
mining at a loss is bad for the network.
OK
01:11
O K
The network grinding to a halt is bad for people holding amoveo, it's bad for all of us
Z
01:16
Zack
That is not a realistic scenario.
OK
01:20
O K
More of us are mining 'at a loss' (i.e. with a high opportunity cost) than you might expect. The current situation, at the very best, is stunting amoveo's growth
PL
01:23
Paww Lee
Mining is hard....why is the price not reflecting that?
OK
01:24
O K
In reply to this message
The situation revolves around the difficulty adjustment algorithm problems, which we are reviewing some changes for
01:24
In reply to this message
If you'd like to help, you can find more info here
01:24
In reply to this message
And here
[
01:32
[Riki]
We are not at a mining pareto optimum currently, i think that is obvious. If the largest miners mine at loss, something is wrong. You dont need much knowledge to conclude that. Having said that, changes (=improvements) to the diff algo should reflect positively for miners while not hurting other interest groups.
01:33
You guys have been talking about diff adjustment for 2 months now. So my question is: is there a good will to change something or do we keep this chat in the domain of academic discussions?
OK
01:34
O K
In reply to this message
I think we're "ready" -- we just need someone of an appropriate skill to review Zack's work
F
01:35
Fića
You said it like its a problem? I also agree with what Riki mentioned
OK
01:35
O K
Personally I think we could launch something broken and still be fine, because we can undo it. Zack may feel uneasy about that kind of action though
01:35
@Simon3456 may have an opinion too
Z
01:35
Zack
if the tokens get hyperinflated, we can't undo it.
MF
01:36
Mr Flintstone
exactly
OK
01:36
O K
Then we should hire an experienced auditor and be done with it
[
01:37
[Riki]
Create a private working group Zack , invite 3 or 10 or 100 people. Ask them 10 questions and publish your conclusion, whatever the outcome is. Telegram chitty chat is also VERY inefficient.
MF
01:37
Mr Flintstone
who designed the ethereum diff adj algo?
01:37
why don’t we ask them for feedback
Z
01:37
Zack
if the tokens are hyperinflated, whoever gets the free money will quickly convert it all to bitcoin.
If we destroy the veo after it has already been sold for bitcoin, then we aren't stopping any attack. We are hurting our loyal supporters.
MF
01:38
Mr Flintstone
In reply to this message
yep, we need to be very careful
Z
01:38
Zack
In reply to this message
ethereum's algorithm is like bitcoin. it takes 2000 blocks to retarget.
01:38
2000 blocks at 10 minutes is how fast we can already retarget.
01:38
we want to go faster.
OK
01:39
O K
I know someone from Ethereum, I will ask him
F
01:39
Fića
In reply to this message
Honestly thats the right mindset on any modern project. The development and deployment of new features should be done in agile matter, doing development of a behemoth project behind closed doores is simply not effective.
Single feature takes months.
Z
01:40
Zack
I didn't find any projects with faster mechanisms.
Which is why I invented a mechanism. It looks like it can retarget in about 40 blocks.
ŽM
01:40
Živojin Mirić
Did you thought about hiring someone for review if that's the only obstacle? It's very impressive that you did all of this on your own in this quality but the project is live and growing it's unsustainable to continue as one man army.
OK
01:42
O K
In reply to this message
We also need to move forward
MF
01:42
Mr Flintstone
so the worry is more about the quick retargeting being exploitable rather than having a block target be a range?
F
01:44
Fića
In reply to this message
Sorry if im simplifying this, but is it possible that you do the implementation in steps. Put up a safe and tested retargeting mechanism that doesnt do record breaking speed but solves the problem at hand. After a successful implementation continue with work on improving the speed of the mechanism even more. Would that be a viable option?
S
01:45
Sebsebzen
@johntromp do you have experience with diff adjustment?
Z
02:18
Zack
How about we use futarchy to decide which retargeting protocol is best.
We can make the decision that is best for the price of VEO.
B
02:27
Ben
Zack aizen controls 40% and i control another 25%
PL
02:30
Paww Lee
Rich AF
MF
02:39
Mr Flintstone
In reply to this message
I like this idea
Justin Buck invited Justin Buck
MF
02:48
Mr Flintstone
we wouldn’t need to wait until the oracle was finalized for this right?
PL
02:52
Paww Lee
In reply to this message
For the sake of decetralisation, you should start sharing.
S
02:52
Sy
how do you share hashrate?
PL
02:52
Paww Lee
😄
02:53
Ah, OK, I thought they have many tokens... 😁
S
02:54
Sy
well...if you own 40% of the net hashrate you are somewhere close to that number aswell
PL
03:22
Paww Lee
In reply to this message
...sooo sharing is caring...
M
03:24
Mike
No one is selling VEO.
PL
03:24
Paww Lee
Should I send my public key? :-) I know from experience that having the majority of a particular token sucks.
03:25
It's like playing with yourself...seriously...
03:27
You give me your coin, I give you my coin... Deal? :-)
OK
03:29
O K
In reply to this message
If the answer is "not right" then I am 100% opposed at this time. We need to get moving
S
03:31
Sy
+1
IP
03:37
I P
basically we need someone to review idea and code behind the new diff retarget algo, right?
OK
03:38
O K
In reply to this message
Yes, an audit basically
Z
04:42
Zack
some people are complaining that txs are being dropped.
I see a tx in my mempool to spend 0.1 veo, ready to be included in 27889.
Lets see if it gets included.
OK
04:43
O K
Based on experience with a couple people I tried to help, I suspect the tx may not be entering the txpool
Z
04:44
Zack
I added a second tx to the tx pool, and I see that it was shared across the network to one of my other nodes.
04:45
In reply to this message
Maybe there is a problem with the light wallet software then
OK
04:46
O K
Does the light wallet use spend/2?
Z
04:47
Zack
I used the new version of the light wallet to make a tx, and it appeared in the tx pool.
04:48
In reply to this message
no. api:spend/2 uses your private key to sign a tx.
04:48
instead it uses the api from the external handler, spend_tx/4
04:49
or create_account_tx/4 depending on the situation.
04:49
I see that my second tx has propagated across the network to my second node.
04:50
I see that my mining pool is working on a potential_block that contains all the txs.
MF
04:53
Mr Flintstone
@potat_o how many different people did you see having this problem today?
OK
04:54
O K
Today: 1
04:54
I experienced myself yesterday with pool
MF
04:54
Mr Flintstone
and sy said the same thing
04:54
I think
04:55
maybe it’s an issue with sy since he’s the one finding the blocks?
04:56
although that wouldn’t explain why we can’t see 0 confirm tx
OK
04:56
O K
Looks like my node gave tx id, but the tx was lost at some point after that
04:57
What happens first, does the node give a tx id before or after the tx is added to the txpool?
Z
04:57
Zack
A block was mined. all of the txs were included.
05:01
In reply to this message
it uses a different thread for calculating the hash and for adding the tx to the pool.
Receiving a hash of your tx is not a guarantee that it got into the pool.
05:03
I think maybe there is no problem with the tx pool, instead some other error happened, and the details of this other error got lost when we played telephone.
05:09
Deleted Account
I know its perhaps overkill, but using the median over many blocks to determine difficulty means any attacker must modify more than half the blocks in a period. Seems like a pretty safe option to me
Deleted invited Deleted Account
Deleted invited Deleted Account
Deleted invited Deleted Account
10:19
Deleted Account
If anyone's interested in breaking out discussions into threads, I set up a discourse board, free trial. If people like it and use it I'm happy to foot the $100/month fee to use their product. https://amoveo.trydiscourse.com/
Z
10:33
Zack
there are channels on discord
10:48
Deleted Account
Channels work a bit differently than threads though. In other communities, like Bitcoin, bitcointalk/forums is used in conjunction with IRC/channels. As an example of where channels may be inadequate, the retargeting algorithm probably deserves it's own channel in discord if the community is to reach a cohesive understanding of the changes being made. But that pattern repeated would lead to an explosion of channels. On the other hand it could go into the mining channel but the substance would likely get lost among the other simultaneous discussions.
S
10:55
Subby
Man I miss old Vbulletin, why is every forum platform trying to be a social media site
11:08
Deleted Account
Yeah it's a little pretty. I just went with the one Zcash uses
S
12:44
Sebsebzen
It also could risk to split the community up between to many platforms
12:44
Reddit (dead), telegram, discord
B
13:02
Ben
discord is great, but it need moderators, like i said like multiple Times, but Zack want to run everything...
H
13:03
Harmony is lifer • $ONE 🦄
How much is per 1 veo now?
AK
13:12
A K
discord and telegram don’t allow for knowledge to be accumulated, refined, updated and shared with the outside world.
13:12
amoveo is still lacking in this regard
13:12
TG is great for building a smaller community, I received many great answers
13:13
however for efficiency sake most of my questions should’ve been posted and answered online already, reddit / discourse allow for that
13:14
V
13:52
Vitaliy
tell me where to find a purse to send my coins to the stock exchange? Thx!
15:44
In reply to this message
Zack I'm curious, why do you want to run everything? or do you really want to run everything?
Denis Voskvitsov invited Denis Voskvitsov
[
20:56
[Riki]
Where do we send our CVs to apply for a summer internship at amoveo
B
21:30
Ben
first get approval from you mom
Z
22:36
Zack
Put your application in a github repository, and PM me a link here on telegram.
3 August 2018
Z
00:01
Zack
I accidentally made an oracle before I finished typing in the question.
I am lucky we lowered the fee for making question type oracles.
MF
00:09
Mr Flintstone
is this the one for price and diff adj speed?
Z
00:11
Zack
https://github.com/zack-bitcoin/amoveo/blob/master/docs/design/futarchy_for_blockchain_updates.md

I am making a futarchy market to decide on if we should do the hard update for retargeting.
MF
00:11
Mr Flintstone
cool, thanks zack
00:11
here is what the oracle question would be
00:11
"X = retarget branch gets merged into mainnet before block height 28000; Y = price of veo around height 29000 is > $300; (X AND Y) OR (!X AND !Y)."
Z
00:12
Zack
once the oracle tx gets included in a block, I will create the market.
00:15
I wonder if any other blockchain has used a futarchy market for a hard update.
People have been talking about this goal for a long time. It is one of the major goals of truthcoin/bitcoin hivemind since it started 4 years ago.
AK
00:31
A K
Price of veo is impossible to define
00:32
At present time. CME has huge rule books re price, we don't
00:32
Why do we need this condition at all?
Z
00:34
Zack
It should still function for futarchy purposes.
MF
00:37
Mr Flintstone
why is the price impossible to define
00:37
we need the price condition so we can understand the impact of the diff adj algo speed on the price of veo
AK
00:40
A K
Price is defined as smth ppl actually are willing to pay (doh)
00:41
So unless you put your money where your mouth is, it's a weak definition
00:41
So we need exchanges / derivatives with rules transparency and volumes/ liquidity
00:42
We don't have them yet
00:43
Price is a market function, not smth to vote on
MF
00:43
Mr Flintstone
we have exchanges
00:43
that have prices
00:43
that we can reference
AK
00:44
A K
Two point if failures, no decentralisation, no audit even
00:44
It's weak
00:44
We can already see real volumes in OTC deviate a lot from exchanges
00:45
With all respect to exchange operators, we are not there yet
OK
00:45
O K
In reply to this message
Did we get an answer to this
Z
00:45
Zack
We can't measure price down to the penny, but I still think there is a >90% chance that the community will be able to agree if it is above or below $300 around that height.
AK
00:46
A K
If I'm not putting my money at stake AND I get to vote on a price
Z
00:46
Zack
In reply to this message
no. we don't need to wait for the oracle to finalize.
AK
00:46
A K
I have no skin in the game
Z
00:46
Zack
there is no voting in Amoveo. voting is not a cryptoeconomically secure protocol.
AK
00:47
A K
PoS is not voting?
OK
00:47
O K
In reply to this message
Great, thanks
Z
00:47
Zack
There is no voting in Amoveo. we use futarchy for the oracle.
AK
00:48
A K
So forking? sorry if I'm using "voting" liberally
Z
00:48
Zack
we use futarchy in combination with nakamoto consensus
00:48
futarchy is governance by the invisible hand of the market.
AK
00:50
A K
Perfect , I love it and veo and that's why I'm here. So you're saying there will be financial consequences for voting against the majority?
MF
00:50
Mr Flintstone
Yes
Z
00:50
Zack
no.
There are financial consequences for lying.
MF
00:50
Mr Flintstone
if you bet incorrectly you lose money
00:51
there is a semantic disconnect right now
Z
00:51
Zack
doesn't matter if the majority is lying.
OK
00:51
O K
Vote is a four letter word here
00:51
Bet
MF
00:51
Mr Flintstone
In reply to this message
😂
AK
00:51
A K
In reply to this message
😂
00:52
And if majority is lying, what invisible had fixes it - miners?
00:53
It seems it will function quite like traditional markets for price discovery, just... without exchanging value. I'll need to think about it 😅
Z
00:53
Zack
tokens on a fork that is lying are less valuable than on the honest fork.
This is why I say it combines Nakamoto consensus with futarchy.
00:54
try reading the documentation.
MF
00:55
Mr Flintstone
I don’t really understand lying or telling the truth in the predictive context
AK
00:55
A K
Thank you for the explanations and honestly I did. Just price discovery is my bread and butter, didn't think before it can happen without exchanging value. I'll think about it.
MF
00:56
Mr Flintstone
In reply to this message
nvm, you meant the oracle
OK
00:56
O K
In reply to this message
I think 'lying' is a strong word. It's easy to be wrong without lying
Z
00:56
Zack
In reply to this message
oracles are not predictive.
If you are participating in an oracle, then you are reporting on the outcome after the fact.
MF
00:56
Mr Flintstone
yeah
Z
00:57
Zack
https://github.com/zack-bitcoin/amoveo/blob/master/docs/design/futarchy_for_blockchain_updates.md
I am still working on this new post to announce the futarchy market.
AK
00:57
A K
To put it another way, oracle on smth like weight of a given object can be objectively measured (cough merchant of Venice cough)
00:57
Can be done by a robot
00:58
Price otoh needs a more thorough definition
00:58
As a matter of fact I'm used to hundreds of pages of definitions ))
Z
00:58
Zack
weight is something that is only locally known, it isn't so easy to put onto an oracle.
Price is globally known, it is the easiest kind of thing to put onto an oracle.
AK
01:00
A K
That is definitely so, another type of problem )
Z
01:02
Zack
As long as the oracle works more than 50% of the time, then we can get useful information out of it for futarchy purposes.
It doesn't matter if it breaks when there is close call, as long as close calls happen <50% of the time.
AK
01:03
A K
We need a meetup )) just after the diff debacle settles pls )
IP
01:06
I P
conference call would be fine for starters
AK
01:07
A K
👌
B
01:11
Ben
In reply to this message
Zack, I just completed reading your design criteria. While I am excited and amazed by what we can do with regards to the Oracle’s on developer assurance, I spotted one aspects where it appears the ‘trust in 3rd parties’ is still entirely centralized and a second aspect where the criteria is largely subjective.

The first is for an issue requiring to be merged. Here the merging is not decentralized at all, but at the discretion of single individual parties who have merging rights. The logical arguments presented on the page is that futarchy solves the inefficiencies with lobbyists being able to bribe developers, but in this case, the ‘merge ownership right’ is just as equally subject to bribery, and so this example should show how that too is solveable. It would be furthermore ideal to have consensus around the taxonomy of references such as ‘merging’ and where must it be merged as well? on a specific git-url? etc.

Secondly, the ambiguity of the use of the word ‘around’ with respect to ‘around block 29000’ is subjective, and may lead to less than optimal conditions for the developer (more of a suggestion on tighter oracle contracts, to avoid misunderstandings between developers and participants in the market.
Z
01:14
Zack
In reply to this message
Miners can only mine on a fork with valuable tokens to sell. Just because someone merges code doesn't mean it will become part of the mainnet.
B
01:14
Ben
Is there an on-chain psuedo-anonymous way for a developer to communicate with an oracle market creator, that they’d like clarity around taxonomy that is subjective or not clear enough to begin working due to subjective perceived risk?
Z
01:14
Zack
Ambiguity has a lot of benefits in this situation.
If you are too specific, then it is easy to make some flash trades and cheat the market.
B
01:15
Ben
In reply to this message
True, but this answer didn’t address whether you can or cannot on-chain address the merging rights centralization risk mentioned in my point.
Z
01:16
Zack
merging is decentralized. anyone can make a github repository and merge whatever code they want.
01:16
merging in the main net is decentralized.
Miners can only mine on the fork with valuable tokens.
Which fork has valuable tokens is determined in a decentralized way.
B
01:17
Ben
so is the ambiguity helpful in this situation as well, with regards to where the ‘merging’ of this oracle contract is referring to?
Z
01:17
Zack
it isn't ambiguous. it says "merged into mainnet".
B
01:17
Ben
which mainnet?
Z
01:17
Zack
there is only one. the one that light nodes syncn with.
B
01:18
Ben
the amoveo mainnet?
01:18
it just says mainnet
Z
01:18
Zack
if the block chain were to fork, then this contract would fork too. Each contract would exist on it's own main net.
B
01:19
Ben
so whichever mainnet a node wants to call their mainnet?
Z
01:20
Zack
if we are in a car, and I say "the car", you can know from our situation that I am reffering to the car that is driving us.

Similarly, when you are in a blockchain, and say "main net", you can know that it is referring to the blockchain we are currently inside of.
B
01:21
Ben
the way i read this, is that whichever mainnet nodes think you are referring to, assuming you know which one they ‘think you are referring to’ then your derivative claim makes perfect sense: If the price of True is higher than the price of False, then that means merging the retarget branch update would increase the odds of Amoveo being worth more than $300.
01:22
as a discovery mechanism of what is probably more valuable to do for the community, this design is useful
Z
01:23
Zack
you can interpret the oracle however you want. Words in english have basically subjective definitions.
You can lose a lot of money if you interpret it differently from the rest of the community.
MF
01:52
Mr Flintstone
the oracle tx should have been included in a block by now right
01:53
yeah, I see a new oracle in 27911
Z
03:03
Zack
yes. I made the market
OK
03:24
O K
In reply to this message
The markets show when I cilck List markets but when I click the button for the market itself nothing happens Zack
Z
03:24
Zack
yes, I am working on a fix for this now.
OK
03:30
O K
I thought you would have tested it first.
03:30
So 1000 blocks from the change, 10 minute blocks, about a week?
03:30
For the price to jump to $300 per Veo?
Z
03:31
Zack
right
03:31
tested what first?
03:32
oh, displaying markets. yes.
OK
03:32
O K
In reply to this message
That's kind of a silly assumption
Z
03:34
Zack
what?
03:44
I fixed it. now you can display both markets.
03:52
In reply to this message
we are not assuming that the price will go to $300.

If situation A would result in a higher price per VEO than situation B, then we can safely assume that situation A has a greater than or equal to probability of VEO being worth more than $300, in comparison to situation B being worth more than $300.
04:01
if (A>B) then (prob(A>300) >= prob(B>300))
Z
04:51
Zack
I made a mistake in the oracle, so I am going to re-make it.
OK
04:52
O K
I almost just placed a bet
04:52
👍
Z
04:53
Zack
if you did place a bet, you can cancel it to get your veo back. as long as no one else matched your bet.
MF
05:04
Mr Flintstone
our veo is valuable because we can do stuff like this, so I hope lots of people participate in this process
OK
05:09
O K
I thought we implemented a default channel delay, it must have gotten overwritten at some point
Z
05:12
Zack
if you want to participate in longer-lasting oracles, you need a longer delay for your channel
Z
07:39
Zack
There is some error stopping me from making bets in the market.
07:41
maybe I need to wait for the new channel tx to be included in a block.
07:53
I put a bet in the market.
As it stands, we will not be doing the update.
It would take a 0.1 Veo bet to change us so that we will do the update.
JL
08:33
Josh L
Can i bet that veo will be worth 0 sats soon?
Z
08:37
Zack
In reply to this message
you could, but it isn't a good idea.
If you won the bet, you would get paid in worthless VEO.
If you lost the bet, you would lose your valuable VEO.
OK
08:38
O K
🔥
Deleted invited Deleted Account
OK
09:20
O K
Zack I am trying to open a channel. I am getting this error:
Z
09:20
Zack
are you synced ?
09:21
how many veo did you put in the channel?
OK
09:21
O K
I have put anything from 0.8 to 1, I have tried several times, and I am synced
09:22
What channel delay and channel life parameters should I use
09:22
I have tried several different ones
Z
09:38
Zack
channel delay should be at least 100. the other should be at least 3000 to participate in this market.
09:39
your light node is connecting to the correct full node that is serving the market?
09:39
near the top it says "channel node ip"
OK
09:40
O K
Yes man, but I definitely wasn't putting a channel delay of 100 years
09:40
I mean, blocks
Z
09:40
Zack
I guess I should put a default in there
OK
09:40
O K
In reply to this message
We did ...
09:41
Mysteriously that change is gone
09:41
It must have been overwritten
S
09:42
Sebsebzen
In reply to this message
years, blocks, all the same here :P
OK
09:43
O K
I also remember making it so the market ID could be copied and pasted
Z
09:43
Zack
it can be. it is written twice.
click the button and when it loads the oracle data, it includes the id
09:44
just above the graph
OK
09:44
O K
The link doesn't even link to the right explorer man
09:44
It goes to 159 from 139
OK
09:45
O K
I'm just saying
09:45
holy fuck
Z
09:45
Zack
what link?
09:45
that sounds like a bad error
OK
09:45
O K
see the available markets here
Z
09:46
Zack
where is that? in the README?
OK
09:46
O K
No, on the wallet, below markets
09:46
It's only visible after channel success
Z
09:48
Zack
yes, I see it. thanks for telling me.
OK
09:50
O K
Alright so it'd still be nice if there was an easy way to match your bet
09:51
I need to bet what 45, true, and anything over 0.1 to match your whole bet
09:51
?
Z
09:55
Zack
you want to match it exactly?
Any bet 45 or greater should match with it.
I think you only need to bet 9 / 110 veo, but I am not certain.
We don't want to leave the market empty, because then it isn't clear if we should do the update or not.
We want the price to be solidly above or below 50%.
09:56
if it is nearly empty, then I guess we would just wait longer for more people to have a chance to participate.
OK
09:56
O K
It threw an error anyway
Z
09:57
Zack
I fixed that link
09:57
making the bet caused an error?
09:58
27919...
Z
09:58
Zack
it says it failed to verify a merkel proof.
09:58
It is because your new channel tx is not in a block yet.
OK
09:59
O K
oh right because the block times are all screwed up
S
10:03
Sebsebzen
if the threshold to betting is high not many will participate
Z
10:03
Zack
you can bet small amounts
S
10:04
Sebsebzen
i mean the technical hurdles
10:04
ok I will try again
10:07
Zack so the price of 300$ per VEO is what decides the bet?
10:07
what about external factors tho
10:07
like Bitcoin dropping
Z
10:09
Zack
there are 2 aspects to the bet. the price of veo, and whether or not the main net does the upgrade.
10:14
@potat_o your new_channel_tx got accepted into a block
OK
10:15
O K
Nice, thanks for the heads up
10:16
Seems a little confusing
10:18
Ok...
Z
10:19
Zack
looks like you made 2 big bets
OK
10:20
O K
Looks like the blue isn't showing up at all on mine
10:20
Which is why I was confused and kept placing bets
Z
10:20
Zack
did you refresh the graph?
OK
10:20
O K
Yup
10:20
It moved
Z
10:20
Zack
which browser?
OK
10:20
O K
LIke, the orange got squashed
10:20
Firefox
10:21
Interestingly, I can see it in the thumbnail
Z
10:21
Zack
there is tons of blue
10:21
maybe you are blind?
OK
10:21
O K
Okay, so with nightshift the blue is not visible
10:22
lmao
Z
10:22
Zack
oh, you have eye protection mode on your computer
OK
10:22
O K
No, now you know
10:22
People who use redshift can't see the markets properly
Z
10:22
Zack
oh, good point
OK
10:22
O K
That's pretty funny
10:22
I'm glad I'm not blind
Z
10:22
Zack
What colors should we try next? this is a quick thing to change
OK
10:23
O K
😂 I'm not sure, we should open a market
Z
10:23
Zack
black white grey
OK
10:23
O K
In reply to this message
Tbh I had this thought yeah
10:23
I really don't know the answer
10:24
I don't know why it happened
10:24
It seems like you should be able to keep the colors and fix the css
10:25
Let me see
Z
10:26
Zack
black yellow white. like amoveo
OK
10:27
O K
In reply to this message
💯I'm sold
Z
10:29
Zack
I updated it. there is no white left because a batch of trades matched.
Z
10:31
Zack
I added some tiny bets to show the white
10:33
I think the channel balance display is flipped.
It says "your balance" for servers balance, and it says "servers balance" for your balance.
or maybe it is messed up in a more complicated way.
MF
10:33
Mr Flintstone
shouldn’t there be no yellow above the white?
10:33
what is going on there
Z
10:33
Zack
this market matches trades in single price batches. it is set to have one batch per 40 blocks.
10:34
bets can overlap until a batch, and then they match.
MF
10:37
Mr Flintstone
it would be nice if you could lock up coins in a tx
10:37
then later, open a channel with someone instantly
10:38
idk if that is possible though because of double spend
OK
10:39
O K
I think you have to change the colors back
10:39
For the description
Z
10:39
Zack
if there is an indirect channel path to the person you want to make a smart contract, you can do it instantly.
You can simultaneously start the process of making a direct channel with that person.
Once the direct channel is ready, you can trustlessly move the contract to the direct path.
OK
10:39
O K
to make sense
Z
10:41
Zack
In reply to this message
thanks, I fixed the description but left the colors yellow and black.
T
10:49
Topab
At what block hight is the next difficulty adjustment?
MF
10:51
Mr Flintstone
27921
10:51
28000*
T
10:52
Topab
👌
S
11:16
Sebsebzen
so what are the next steps?
Z
11:16
Zack
who keeps betting at 1:0 odds? haha
11:17
basically giving me free veo
11:17
In reply to this message
lets give it another day or two for the market to settle.

I also need to program the light node to handle the hard update.
S
11:18
Sebsebzen
nice
OK
11:24
O K
In reply to this message
Maybe the interface should not allow this range
Z
11:27
Zack
there is still 40 blocks left in the batch. each batch is single price.
If more people like me keep undercutting each other to get free money, it will push the price of the batch to a more reasonable level.
So they will likely end up trading at a decent price.
11:28
A lot of firsts for Amoveo today.
Our most popular market ever.
The first time using a market for futarchy for a hard update.
The first time embedding boolean logic inside of an oracle question.
12:08
Deleted Account
I would like to create a market "will price 1 month after ICO will be greater than ICO price for foam"
12:08
How much it will cost?
Z
12:09
Zack
I think it costs about 0.04 veo to make a oracle now. So around $1.
MF
12:09
Mr Flintstone
isn’t it 0.022?
Z
12:09
Zack
could be
12:09
the most expensive part is that you need to put some veo on the server so that people can make channels with your server to do bets.
12:10
you do get compensated for this service.
MF
12:11
Mr Flintstone
1250 sats per veo per block is super high annually
12:11
like 66% interest in veo terms ? annual
12:11
1250*144*365/100mn
Z
12:13
Zack
1 year is 144*365 blocks.
(1.0000125)**(144*365) =~ 1.93
So it is 93% annual interest
MF
12:14
Mr Flintstone
it doesn’t compound
Z
12:14
Zack
so 1250 satoshis per block per veo is a good deal for the user, and a bad deal for the server.
You can edit this fee size in the config file.
MF
12:14
Mr Flintstone
I don’t think
12:14
if I open a 1 year long channel
12:15
Deleted Account
I think it would be great to create a service where for every ICO there is question if price after 1 month will be up - it will extract insider knowledge from the insiders
Z
12:16
Zack
yes, you are right. it doesn't compound for one year long channel.
But if lots of people keep making short-term channels with you, then your profit does compound. Since you have more money available to make more channels.
MF
12:17
Mr Flintstone
In reply to this message
true. We can establish a range of interest Between continuous compounding and non compounding
Z
12:17
Zack
In reply to this message
yes, these kinds of markets will probably be popular. people already interested in cryptocurrency are more likely to care about Amoveo.
MF
12:18
Mr Flintstone
In reply to this message
why would this be good for the user if they have to pay over 60% interest
Z
12:20
Zack
60% annual in the cryptocurrency market is cheap. Less than 5% per month.
12:21
I guess it is all a matter of perspective. Who knows what the market will decide the correct price is.
12:22
Deleted Account
w8 It's possible to earn money for running the market? 60% passive income seems nice & comfy
Z
12:23
Zack
yes. you can make passive income from running a market, or by running a hub for lightning payments.
S
12:28
Subby
How many veo required for a market and how many for a hub? Just for a rough idea
Z
12:28
Zack
if you have 1 veo in your hub, then users can lock up about 1 veo into channels with you.
12:31
Deleted Account
And how routing fees work in LN? Same as in bitcoin?
Z
12:32
Zack
so far we have only talked about a fee for having a channel.
Besides this there is a small fee for making a bet, or for making a payment.

But I expect the majority of the cost will be from having a channel open.
Z
12:56
Zack
I get the feeling someone is trying to draw the anacap flag on the volume graph.
13:02
About $2000 worth of veo are being used for the futarchy market now.
TG
13:58
Toby Ganger
whoever is running qtrade needs to fix their withdrawals...it won't allow any input of under 10 VEO to withdraw..but it also has a max of $2000 worth that you can withdraw...so the coins are stuck on the exchange
A
14:06
Angelos | Coinomi Wallet (I will never PM you first)
is that an "error" to be fixed or is it just their limits?
TG
14:41
Toby Ganger
it's actually a prompt error...I realized that the withdraw actually works..even though the prompt says it won't work
14:42
but yeah...it's a weird glitch in the prompt that won't "allow" a withdrawal of less than 10 VEO...but won't allow a withdrawal of more than $2000 of value...so it will basically say you can't withdraw no matter what you do...but it works...just confuses you
AK
14:48
A K
they switched to mVEO ahead of schedule ;)
TG
15:09
Toby Ganger
haha
15:09
wow someone put in a big buy order
S
16:24
Sy
167 veo buy
16:25
Not Bad, Wish the price was Higher 😅
S
18:00
Sy
about 3 days for next diff...
+
19:17
++
In reply to this message
💯
OK
19:41
O K
In reply to this message
Yes well, this should be done. Zack are you ready to make this change?
ŽM
19:42
Živojin Mirić
In reply to this message
let the Futarchy decide
OK
19:44
O K
In reply to this message
Seems a bit masturbatory considering how much support has been expressed, but ok
ŽM
19:44
Živojin Mirić
I'm joking a bit :)
OK
19:45
O K
In reply to this message
😂
22:23
Zack I think the jury is in
22:45
Deleted Account
When change into mveo?
OK
22:52
O K
In reply to this message
I guess we need to open a market
22:52
ASAP imo
23:15
Deleted Account
@potat_o what does the plot show
OK
23:16
O K
It shows that quite a bit of veo (yellow) has been bet (true) that X = retarget branch gets merged into mainnet before block height 29000; Y = price of veo around height 30000 is > $300; (X AND Y) OR (!X AND !Y)
23:16
vs the grey, false
AK
23:17
A K
i assume only ppl who can understand how this works are here, in this channel
23:17
so the question is, who bet against ))
23:17
Deleted Account
What is on the X axis?
OK
23:17
O K
In reply to this message
Zack
AK
23:17
A K
testing?
OK
23:17
O K
I presume it was some incentive
MF
23:51
Mr Flintstone
In reply to this message
price
23:51
y axis is amount of veo
23:54
the image is an order book
IP
23:56
I P
guys, but i guess it ,males sense to bet against that
MF
23:56
Mr Flintstone
23:56
for example, let’s look at the order that the arrow is pointing at
IP
23:56
I P
to hedge veo price
MF
23:57
Mr Flintstone
this is set at a price of 0.55
23:57
which is the x axis
23:57
and then the amount of veo that can be won at this price is the height on the y axis
23:59
so, let’s say that the price is 0.55 and the amount on the order book is 1 veo. this would mean that you are offering to buy a share of the true outcome for 0.55 veo, which would pay you 1 veo if true is the real world outcome of the oracle question, but only if someone matches your bet
4 August 2018
MF
00:01
Mr Flintstone
the way to match the bet is to take the other side, which would be a bet on false at a price of (1-0.55) = 0.45 and an amount of 1, which would be 1*0.45 = 0.45 veo committed
00:02
@potat_o I think this is why you thought that you were making more bets than you could have, because the “amount” field is the amount you win, not the amount you bet
00:02
I had the same confusion last night, so we updated the channel interface to communicate this
Z
00:28
Zack
I bet against because I wanted the free money from the person who bet at 0:1 odds.
T
01:21
Topab
@Jbreezy0 the example you gave with that order book is the market the retargeting decision? Veo price is in usd, what price units are you using in the example?
S
01:25
Sebsebzen
Yep that probably should have been specified
MF
01:44
Mr Flintstone
In reply to this message
to participate in the market, we are betting with our veo
01:44
the outcome of the oracle is dependent on the usd price of veo and whether or not the diff adj is fast or slow
01:45
so, all the prices I quoted were in veo terms
IP
01:46
I P
In reply to this message
i guess people don't fully understand how bets match on veo and thus you can get free money in such way
S
01:48
Sebsebzen
how to measure usd price of VEO tho
IP
01:50
I P
In reply to this message
yep i guess btc price would've been better in oracle
01:50
usd price is tricky because btc can dump or moon anytime
MF
01:56
Mr Flintstone
In reply to this message
we have exchanges
01:56
also, the ambiguity in the question is helpful
01:57
it’s like a schelling point
01:57
90+% of the time, this should work OK
01:57
the edge cases would be less clear, like if veo keeps bouncing between 295 and 305 or something around the stated block height
01:57
but we have to deal with this possibility in order to effectively use futarchy I think
T
02:25
Topab
In reply to this message
I have never been very good at betting, it's still confusing to me. The bet is in veo, I get this. What is confusing is the diffent explanation of axes by you and @potat_o "X = retarget branch gets merged into mainnet before block height 29000; Y = price of veo around height 30000 is > $300; (X AND Y) OR (!X AND !Y)"
OK
02:26
O K
Sorry, this is not my description of the axes
02:26
That is from the market description, which Zack chose X and Y to be logic variables equal to those explanations
02:27
now we know that was a poor choice, probably p and q would be a better, more classical choice for logic variables
MF
02:31
Mr Flintstone
yeah, agreed on that
Z
02:32
Zack
In reply to this message
yes. X, and Y was a mistake. P and Q would have been better.
02:33
In reply to this message
your confusion is understandable.
This is a very complicated oracle.
02:37
We had always planned to use combinatorial markets for stuff like this.
Hacking the oracle question to accomplish the same goal is a recent discovery.
T
02:42
Topab
In reply to this message
Now getting clearer :)
02:43
In reply to this message
The bet side is then depending on the results of two oracles, is this right?
OK
02:43
O K
There is only one oracle
T
02:44
Topab
I thought it was price of veo and diff adjustment
MF
02:44
Mr Flintstone
that’s right
02:45
but the value of those two are used to create one true or false output
02:45
using AND and OR operators
02:45
like (P and Q) or (not P and not Q)
02:46
this statement resolves to either true or false, but it takes two inputs: veo price > 300, and diff adj is fast
02:46
which themselves are true or false statements
02:48
you can try it yourself: apply different true/false combinations of veo price > 300 and diff adj = fast to the formula and see what the output is
Deleted invited Deleted Account
MF
02:51
Mr Flintstone
In reply to this message
you can call P the veo price Boolean and Q the diff adj Boolean
02:53
Deleted Account
But what does the plot represent, on the X and Y axes? I understand what the oracle text means...
MF
02:54
Mr Flintstone
In reply to this message
was this unhelpful
02:54
apologies if this was unclear, there are probably better ways I can explain it
T
02:55
Topab
I think is clear it's just my poor understanding of trading and betting
02:55
But I still need to look at it more before I can fully understand it
MF
02:56
Mr Flintstone
this can be pretty complicated stuff, but if we break it into its most simple logical components we should all be able to understand
02:56
maybe we just need better explainers
02:56
like in the docs
02:57
In reply to this message
try to interact with the market itself too, that helped me understand it better
Z
03:00
Zack
the yellow-black-grey graph is a price volume chart of the orders in the order book.
left-right is the price being traded at.
up-down is the quantity of veo being traded.

Yellow is for people buying shares of true, grey is for people buying shares of false.
03:01
Deleted Account
No. This is quite simple imho. We just need to come up with some better presentation
03:03
So i could buy «false»shares for ~5 veo to match current bets and then double my money if the statement turns out false
Z
03:04
Zack
right. it would cost around 5 or 7 veo to match all the trades in the market right now.
03:04
Deleted Account
And what we are seeing is that more people thing that the statement is true
03:04
So we can conclude that people think forking is a good idea
MF
03:05
Mr Flintstone
I think stronger evidence that we can conclude that the fork is good is that nobody is matching the bets
03:06
but ya
03:06
Deleted Account
Hmm. How long is the bet opened?
MF
03:06
Mr Flintstone
since 29721
03:06
first batching is at 29761 I believe
03:06
Deleted Account
I mean when does it close
03:07
Batching?
03:09
Interesting thing is that the bets may be used to decide whether to fork, but zack may still win all the money if the reality is not as expected at a later stage
MF
03:10
Mr Flintstone
Zack only wins what he matched
03:10
Deleted Account
Ok. So unnatched bets are returned?
MF
03:10
Mr Flintstone
which is around 0.5 veo right now since people bet on 1:0 odds
03:10
yeah you can take them out at any time
03:10
even now
03:10
Deleted Account
This is pretty cool
MF
03:11
Mr Flintstone
yeah
03:12
Deleted Account
Id say right now Id be willing to bet on false, because even though I would support forking, I believe the condition is also quite likely
MF
03:12
Mr Flintstone
high price slow diff adj would result in false
03:12
as would low price fast diff adj
OK
03:14
O K
If you bet false, you only win if P and Q are opposite conditions
T
03:14
Topab
In reply to this message
But in this case it's about using markets to take better decision, does that mean that diff adjustment fast is a bad decision according to Zack?
03:15
Deleted Account
Yep. But theres more at stake. Perhaps I dont think the price will rise to $300 after forking, but I still believe the price will be higher after a fork compared to no fork. Then i will be inclined to vote yes, to make sure the fork is executed
Z
03:15
Zack
I bet at a price of 0.03.
So I guess that means I think there is at least a 3% chance that it is a bad idea.
OK
03:16
O K
In reply to this message
There is at least 3% chance that it is not as good of an idea as it is portrayed to be in the question
03:16
Not necessarily that it is a *bad* idea
03:18
Deleted Account
Why is the price relevant? You are betting at odds 1:1, and if you believed «false» to be 3% likely, you are acting in a way to lose money on purpose
Z
03:18
Zack
such a goofy oracle, we can't even agree on what it is measuring.

When I do the math, it seems to me that if the outcome of the oracle is "True", then that means the retargeting hard update will positively influence the price.

It is possible I made a mistake.
Humans haven't made many wacky prediction markets before, not much history to learn from.
03:19
In reply to this message
the market lets you bet at any price, not just 1:1.
OK
03:19
O K
In reply to this message
I think your general thought process is correct. However it could be, for instance, that the change could cause the price to remain constant instead of continuing to fall. In that case it is a *good* idea, but the question is *false*
Z
03:22
Zack
In reply to this message
I don't understand.
MF
03:22
Mr Flintstone
In reply to this message
oracle betting is 1:1 only
Z
03:23
Zack
oh, right. the price could be low and then I would win my bets.
OK
03:23
O K
In reply to this message
It is conceivable that without the change, the price would continue to fall, and that with the change the price would stay around 200 USD.

That means making the change is better than not making the change. But it also makes the answer to the question as framed right now false.
Z
03:23
Zack
The market should still accurately measure whether the hard update is good in the beginning, regardless of who wins in the end.
There is luck involved.
03:26
like if we are rolling dice, and I bet big on a 1, and then I win.
There was only 1/6th chance a 1 would come up.
The fact that I won doesn't mean the dice is "broken" or "unfair".
There is supposed to be non-zero odds of every outcome.
OK
03:26
O K
Sure. And to get extreme with it the sun could explode and take us all with it 🤷‍♀️
03:26
It doesn't mean the sun is broken or unfair
Z
03:27
Zack
I would bet you any amount of money that will not happen.
OK
03:27
O K
😂 I'm not taking that bet, you've taught me too well
Z
03:29
Zack
If we made the assumption that the price of veo will stay below $300, that still doesn't tell you which way you should bet.
You want to be on the losing side in this case.
So if you think everyone else will block the update, you should try to make it happen.
If you think everyone else will push the update through, then you should block it.
T
03:30
Topab
In reply to this message
Dice roll is pure luck, effects of decisions are not always predictable but the idea is to be able to get valuable information from people betting because they have some knowledge about the effect of the decision
Z
03:33
Zack
In reply to this message
I guess my example wasn't so good.
I have a hard time explaining these concepts with english.
OK
03:33
O K
In reply to this message
You should try esperanto
03:33
Deleted Account
We should have predicted the future price of veo before wording «$300» in the oracle. Right now you are assuming that $300 is a sensible veo price when the oracle resolves. If people think this is optimistic, then the results will be skewed towards «no fork»
Z
03:34
Zack
In reply to this message
I think we could have used $100 or $500 instead of $300, and it would still be fairly accurate.
03:34
there is probably some price that is a best guess, but in a way we don't want to use that price.
If it is a close call, then it will be confusing to know if the outcome of the oracle is true or false.
OK
03:35
O K
Honestly, since there was a strong community support AND the primary concern was hyperinflation, we could have combined Doing the fork with Price drops below $100
03:35
Then we would have known if it was really as dangerous, without having to know that the price would go up
+
03:35
++
In reply to this message
Does a “bet” just reflect popularity or does it reflect something in the world.
For example, if majority bets price will rise to $300 after forking, does the bet payoff occur only if the actual price rises or does the payoff occur once there are no bets for a period of time?
Z
03:36
Zack
yeah, that is reasonable. maybe $100 would have been better.
03:36
In reply to this message
if you bet correctly you get paid. if you bet wrong, you lose the money you risked.
03:36
that is what "bet" means
+
03:39
++
In reply to this message
I think I understand what bet means, I am asking what “correctly” means. Does correct correspond to the majority bet (like voting with dollars) or does it correspond to something in the world like whether it factually rained on a certain day or whether the price was above a certain level at a specific time.
OK
03:39
O K
In reply to this message
The oracles are decided based on facts + interpretation of question
Z
03:39
Zack
this particular bet has 2 aspects. whether the price of veo is above $300 on a particular day, and whether we do the hard update before a particular block.
+
03:42
++
In reply to this message
Still struggling here. I get the sense that what is fact in the system is based on dollar weighted popularity. Am I wrong?
OK
03:42
O K
In reply to this message
The fact is based on reality, the oracle is decided in the future, after the events have or have not taken place
+
03:43
++
In reply to this message
Who is responsible for getting the “reality” into the oracle? What is the transmission mechanism? Isn’t it just popular opinion of the voters?
OK
03:44
O K
In reply to this message
While also, people's bets before the oracle is decided may give us information about what is expected by those with skin in the game
03:44
In reply to this message
Nakamoto consensus
03:45
Let's do a simple example
+
03:45
++
In reply to this message
Sounds like a voting machine, not a weighing machine.
OK
03:45
O K
There is no voting
+
03:45
++
In reply to this message
Yes. Vote with dollars. Majority rule, regardless of facts. Unless I’m mistaken.
03:46
Dollar weighted voting
OK
03:46
O K
I don't think that is any popular sense of the word vote, but at any rate we avoid the word generally
03:46
Deleted Account
«The first order in the book needs to be as big as the initial liquidity. Every order after that needs to be twice as big as the previous order.» — is this wrong? My understanding was more like: «the minimum order size is doubled for each order»
+
03:46
++
It aggregates opinions with no mechanism to check to reality.
OK
03:47
O K
In reply to this message
I don't know why you keep saying this... there is a mechanism
03:48
Simple example time?
+
03:50
++
In reply to this message
Is there no concern that oracle only reflects dollar weighted popularity rather than truth in the world?
Many unpopular views are often true but a mob mentality would reward falseness.
OK
03:51
O K
In reply to this message
I'm sorry, but I don't think that's an accurate representation of nakamoto consensus. Maybe you should give me a simple example of what you think will happen, since you're not interested in hearing a simple example from me
03:52
Deleted Account
Can anyone point med to a doc explaining the rules of the betting market?
+
03:52
++
In reply to this message
Yes. Sorry if I’m slow.
Give simple example
OK
03:52
O K
Others jump in please if any thoughts —
03:55
Something obviously verifiably true or false: let's say the question is [Some dormant by name] volcano will erupt in Hawaii in August and level more than 15 homes.
+
03:56
++
In reply to this message
Say bet is Trump loses next election.
Say bets are placed. No activity for a while. Would bet close at the majority view? Bet could close before outcome of election is known if there is no activity for a certain number of blocks, right ?
+
03:58
++
In reply to this message
That works too.
What determines whether the oracle says that is true or false? My understanding is it is just the dollar weighted vote. If not, then what?
Z
03:59
Zack
In reply to this message
we don't want there to be too many orders in the on-chain oracle order book simultaneously. It is consuming consensus space to exist.
04:00
In reply to this message
read the documentation first. then come back if you still have questions.
+
04:00
++
In reply to this message
Yep.
This part in particular suggests it is dollar weighted opinion that determines output of oracle, regardless of what truth is in the world.


“The type with the highest unmatched balance determines the output type of the oracle. If the output type of an oracle has not been changed for a certain number of blocks, the oracle can be finalized with that output type and betting ends. Placed bets are locked up until the oracle is finalized, at which point you are paid out according to the final output type. “
OK
04:01
O K
In reply to this message
Okay, so it doesn't happen. That means the answer is false. The miners will mine on the chain where the oracle decided false, because it is the 'honest' chain and the chain that will hold value
04:01
The chain will not hold value if it is not an honest chain
Z
04:01
Zack
In reply to this message
if you fixate on any individual piece of the oracle and ignore everything else, then it will look insecure.

All the pieces complement each other's weaknesses to make a secure oracle.
+
04:03
++
In reply to this message
So it is dollar weighted opinion checked by miner weighted opinion.
OK
04:03
O K
You can call facts opinions if you want, but I'm not really interested in getting that deep into epistemology
04:06
Deleted Account
We discussed previously that if veo was distribute among many hands, a big veo owner could buy an oracle because very few owners were able to match «double that». According to the wording in tje docs that i pasted into my previous message, this seem to be true. But you left me with the impression that many small owners could match such a big veo holder. Im confused...
MF
04:09
Mr Flintstone
In reply to this message
the doubling just refers to orders that would be sitting in the order book
04:10
since we have to do the order book logic on-chain, it is very computationally expensive to have many orders in the book
04:10
if there is an unmatched order of 100 veo on true for example
04:10
I don’t think there’s any restriction on the size of the bet to match it until the 100 veo runs out
04:11
because you’re not adding to the order book
+
04:11
++
In reply to this message
The final arbiter of truth is not actually the betting market and the oracle. It is the opinion of miners backing a given chain. The weighting given to any miner’s opinion is their relative hashing power. So is this just hashing power as the final arbiter of “truth” ?

Taken to an extreme, the machine here is just hash power weighted voting by miners with the actual betting market as 1 of many inputs into the opinion formation process of the miners.
MF
04:11
Mr Flintstone
since you’re taking liquidity out of the order book rather than providing liquidity
04:11
Zack let me know if correct interpretation
OK
04:14
O K
In reply to this message
The final arbiter of truth is what those who demand and use money believe the truth to be, as a chain that is not [perceived to be] honest is worthless, and miners, being profit-motivated, want to mine the profitable chain. The profitable chain is the [perceived-to-be] honest chain
04:15
To say "miners" are the final arbiter of truth, is a misunderstanding I think
Z
04:16
Zack
In reply to this message
it doesn't matter if an attacker has more than half the veo. he still can't cheat the oracle.
And if he tried, he would lose all his Veo.

Betting is only half of the oracle mechanism. It's purpose is to make the situation escalate to the point where we can use Nakamoto consensus to resolve the oracle.
04:17
In reply to this message
no. the miner's opinion does not matter.
Miners can only mine on the fork that has valuable tokens. They have no power over the oracle.
04:17
In reply to this message
you are correct. we only put lower limits on the size of bets if that bet is being added to the order book.
If you are matching an existing bet, there are no size limitations.
+
04:24
++
In reply to this message
I just repeated what you said.

Your comment below
“Okay, so it doesn't happen. That means the answer is false. The miners will mine on the chain where the oracle decided false, because it is the 'honest' chain and the chain that will hold value”
OK
04:25
O K
I never said "It is the opinion of miners backing a given chain," so you did not repeat what I said
04:26
In reply to this message
I think there is some nuance you're missing that I explained here
Deleted invited Deleted Account
MF
04:41
Mr Flintstone
there goes the buy order 😂
04:41
solid 5 btc volume on qtrade today
04:41
man those are gonna be some painful sells
A
04:52
Aries
Deleted invited Deleted Account
I
04:53
Iridescence
In reply to this message
😂
Z
05:40
Zack
Centrally controlled laws are already hard enough to follow. Now people want to make P2P law.

If every community starts making it's own laws, everything will be illegal right away.

So I can get murdered because some people in India are mad at me for eating beef.
05:42
I taught the light node how to sync after we change the difficulty retargeting algorithm.
OK
05:43
O K
In reply to this message
👍
TG
07:19
Toby Ganger
seems like the volume picked up today? someone sold 167 VEO into that buy wall...wow...
MF
07:23
Mr Flintstone
yeah, around 300 veo in volume today
07:23
biggest day yet on the exchanges
TG
07:25
Toby Ganger
seems to be mostly dumping though...sigh
MF
07:27
Mr Flintstone
probably 2/3 market selling yeah
07:27
decent buying on amoveo.exchange actually
TG
07:33
Toby Ganger
i suppose...until this coin becomes more human friendly the price will probably remain pretty stagnant at best
MF
07:34
Mr Flintstone
idk why but I’m feeling very bullish
OK
07:34
O K
Must be the lingering feeling of doing tech support in the discord
07:34
oh wait
MF
07:35
Mr Flintstone
?
Z
07:35
Zack
im waiting.
OK
07:35
O K
In reply to this message
I was encouraging you to get down and dirty remember :b
07:36
In reply to this message
?
07:43
@Jbreezy0 telegram is doing the recording thing on you
MF
07:44
Mr Flintstone
lol
07:44
thanks
07:44
classic telegram app move
OK
07:45
O K
lol happens to me all the time
07:45
scary
MF
07:56
Mr Flintstone
any thoughts on this oracle
07:56
P = (light wallet with MVEO denomination is merged into amoveo repo on zack’s github) and (qtrade.io has MVEO denomination) and (amoveo.exchange has MVEO denomination); Q = price of veo around height 32000 is > $200; (P AND Q) OR (!P AND !Q)
OK
07:56
O K
Too long
07:56
Well, maybe not
07:57
We need to hard fork or do some protective measure for old light wallets so that people don't send too much veo
Z
08:00
Zack
!(P XOR Q) is the same thing, right?
08:02
maybe we should have an oracle question programming competition. To develop cultural norms.
Z
09:10
Zack
P XNOR Q
one operation.
Z
10:05
Zack
The new retargeting algorithm is tuned with the parameters 4 and 20.
The exponential curve is 20. So the most recent block counts for 1/20th of the average.

if a block takes less than 1/4 of the weighted average time, then we count it as if it took exactly 1/4 of the average instead of less.

I suspect that 4:20 is not the ideal pair of parameters, but I am not certain how to determine what would be ideal.
Z
13:10
Zack
I guess it is safe to leave it as 420 for now, and we can do another hard update later if it seems like we can improve on it.
AK
13:26
A K
Agree, there's no magic constants here to be found
13:26
we're not fitting a physical process
Z
13:31
Zack
I think there is at least some math we could do here to come up with better estimates.
For example, if the weighting is 20, and the target ratio has a width of 2x, we should be able to calculate how often the difficulty will move due to randomness.

Having a 4x limit in one direction and not the other means that it will always measure an estimate slightly slower than the actual hashrate.
Maybe we could calculate how much lower this is, and account for it in the protocol.
13:42
I did some more tests with the new retargeting.
It seems like it adjusts difficulty upwards very fast. Within 10 blocks it is mostly adjusted, within 40 it is completely adjusted.

It adjusts difficulty downward too slow.
I waited 100 blocks, and it still isn't finished reducing the difficulty.
MF
13:44
Mr Flintstone
how close is it in 100 blocks
Z
13:46
Zack
turns out it is 110.
So it takes 110 to lower difficulty, or 40 to raise.
13:46
maybe that isn't a bad thing.
AK
13:48
A K
BCH has (had?) an EDA adjusting down faster, than up. 'cos it was vulnerable to all the SHA-256 BTC miners.
13:48
VEO is vulnerable to GPU miners.
13:49
should we have it adjust down faster?
13:49
w/o EDA they could've waited weeks for a block, potentially
MF
13:52
Mr Flintstone
100 blocks isn’t too bad
13:52
for full adjustment
Z
13:53
Zack
One thing that I worry about with this mechanism is that if you have 20% of the mining power, and you shift time stamps around, you can make the difficulty something like 30% higher.

If some minority decides to raise the difficulty, they can.

Maybe we should again look into averaging inverse hashrate instead of averaging hashrate.
S
14:13
Sy
In reply to this message
You are worried 110 isnt fast enough?
14:14
In reply to this message
Zack specificly reduced the down adjustment alot...we could adjust faster with current diff algo aswell
Z
14:17
Zack
I guess if mining pools kept cheating the timestamps to make the difficulty higher, the network would just adjust the block time or block reward to undo the change.

Then if the mining pools switched back to mining normally, it could cause a sudden drop of the difficulty.
BA
14:46
Buzz Aldrin
In reply to this message
You were right, AE put the main net launch on hold but didn't share the reason why they postponed ...
S
15:01
Sy
In reply to this message
for what purpose? it doesnt make any sense to do that
15:01
In reply to this message
we are needing about 18000 blocks right now, 100 is nothing to even think or discuss about
15:38
Deleted Account
Thinking out loud: for a constant C, difficulty D, hashrate H and desired blocktime B, we have: B=C*D/H. Suggested algo: at every 100 blocks, find H’ is median hashrate H’=median(B’/C/D’) for last one week (using miner reported times). B’ and D’ are measured and actual historic values. Now calculate next difficulty to: D_(i+1) = B / C * H’
15:40
This should rise and fall at the same rate, and oscillation period is known (? Must be proven) to be one week. In case of slow/fast blocks, the time step of the algorithm is not affected, only how often D is updated
15:40
We dont need til have ‘sanity’ limuts too high/too low as median function will discard those
15:41
Limits. Sorry
15:45
This is not an exponential filter like we have right now, rather just an estimate of the correct difficulty based on last week’s mining results
15:47
I think that making the algo dependent on time rather than block number is important, as we are seeing now when difficulty is too high, the algorithm slows down due to increased sampling rate bein the block times
15:50
To attack the network you would have to mine 100 blocks with time stamps far into the future. After that the difficulty would be close to zero. So the number 100 is important, and perhaps one could also add a rate limit on changes to D.
15:51
The numbers 100 and one week are arbitrarily chosen of course as an example
16:22
Deleted Account
Make another adjustment: «for the last week looking at timestamps, minimum 100 blocks»
Dogpie invited Dogpie
Z
21:28
Zack
In reply to this message
You can cause an oscillation by making periodic pushes in either direction.
21:32
In reply to this message
What you describe is similar to how we do it now. Except with 100 blocks per period instead of 1000.
This would be a good way to make retargetting faster.

The way I programmed the hard update is even faster at retargetting, and it should be more resistant to oscillations.

The way I programmed the hard update is influenced by ethereum's retargetting.
If you make a block time in the future, then no one will build on it.
If you make a block time in the past, then the difficulty will be higher for you.
So miners are incentivized to use the current time.
21:34
I did some chores for a sick neighbor, and now I got sick too. It will probably be a couple days before I can get much work done again.
22:19
Deleted Account
such a kind person
OK
22:24
O K
In reply to this message
Are we ready to go ahead and merge this?
22:24
In reply to this message
I hope you feel better soon. Attitude is everything
Z
23:16
Zack
In reply to this message
We are probably ready, but we should wait.
I ran some tests while sleeping. To see how fast difficulty can lower in some other situation.

I'll check the results eventually.
Deleted invited Deleted Account
5 August 2018
Z
00:57
Zack
the test I left running crashed right near the beginning.
B
01:13
Ben
bad news ;(
Z
01:14
Zack
not a big deal. we will probably have results in the next hour.
MF
01:21
Mr Flintstone
In reply to this message
hope you feel better
Z
01:34
Zack
We are 12 blocks into the test, we will need about 50 blocks to complete the test.
it looks like after a sudden drop in hashrate, there is a ~8 block transition period where the difficulty stays the same, then the difficulty drops by ~5% with every additional block.

So it would take something like 28 blocks to react after the hashrate dropped by a factor of 2.

My previous estimate of 110 was messed up because I tried running the test too fast, and there were some rounding error with sub-second blocks.
This time I aimed for 5-second blocks, so I think it is more accurate.
AK
01:49
A K
Great!!
Z
01:52
Zack
20 blocks into the test, difficulty has dropped by more than a factor of 2. So this test is a success. We can respond well to sudden drops in the hashrate.

One last thing we should test before the update is the transition from the old method of calculation to the new. We need to make sure it is a smooth transition.
A
01:53
Aries
01:53
Love it when Cobra speaks the truth
Z
01:55
Zack
It took 32 blocks to completely change the difficulty after the drop in hashrate. So it looks like it is fairly symmetric between raising and lowering.
02:01
There was a problem with the transition. Good thing I checked. It should be an easy fix.
А
02:03
Андрюхин
So you changed diff change algorithm ? How will it work ?
Z
02:06
Zack
А
02:08
Андрюхин
just short question: It is not gonna be changed every 1000 blocks, now it's gonna be change depending on hashrate ?
S
02:08
Sebsebzen
Get well soon, Zack
Z
02:09
Zack
In reply to this message
it has always been changing by the hashrate.
Now instead of changing every 1000 blocks, it will change every block.
А
02:12
Андрюхин
good to hear. Thanks. Do you have any simple formula ? like (average blocktime of last n blocks / target blocktime) * diff ?
Z
02:41
Zack
With a retargeting algorithm, first we make an estimate of the block time. Then, based on our estimate, we adjust the difficulty so that the block time will be closer to our target block time.
Previously we estimated block time by waiting 1000 blocks, and then dividing that amount of time by 1000.

The new way is to keep an estimate of the current hashrate.
We use exponential weighted average to estimate the hashrate, the most recent block accounts for 5% of the estimate.
Using our estimate of the current hashrate, we can calculate what the difficulty should be so that our block time will be closer to the target block time.
02:42
======================
I fixed the switch from the old retargeting to the new so that there will not be a spike in difficulty at the transition. it will change smoothly.
I updated the light node for these changes.
04:46
Deleted Account
I would argue there are some important differences between the old algo and my suggestion: 1. The algorithm is based on time rather than a discrete time function based on block times. So it doesnt slow down and speed up due to hashing inconsistensies. 2. Using the median rather than an infinite impulse response function (the exponential decay function with memory), the algo is not dependent on whatever happened > 1 week ago and pretty immune to spikes.
04:46
Even so i have no issues with you wanting to go for the ethereum like solution. Just wanted to hilight some imho rather big differences
Z
05:52
Zack
@tallakt what if b' is 0?
2 blocks in a row with the same time stamp.

The algorithm I proposed looks up each header twice.
Does your proposal look up eah header 100 times?
05:53
I don't understand difference (1).
E
07:02
EdoFazlinovic
In reply to this message
Chill 😳
Crypto Pig invited Crypto Pig
Z
14:14
Zack
http://159.65.120.84:8080/wallet.html
I added a check point to the light node. it syncs in less than a second now.
AK
15:38
A K
When will we update to a new diff algorithm?
15:38
28000 ?
S
15:50
Sy
28100
15:59
Deleted Account
@zack if B’ is zero it will get discarded by the median function. Yes, you would ned access to a weeks worth of H’ values in order to use this algo. I suppose maintaining a ‘deque’ for this is the most straightforward approach. But this is a bad aspect of the algorithm, i agree.
16:03
The difference (1) is that when the algorithm is a discrete time transfer function that has a time step at every n blocks, if the blocks are very slow, ie sampling time increase, the responses are also slowed down. Eg a exponential decay function with sampling time 6 minutes and time constant one week will have time constant 10 weeks if the sampling time was increased to 60 minutes. I believe we are seeing this now
16:04
My algorithm bypasses this effect in large by using blocks for the last week looking at the timestamps, rather that taking the median of N blocks
16:04
:)
K invited K
TG
16:27
Toby Ganger
market keeps falling huh?
E
16:28
EdoFazlinovic
In reply to this message
Chill 😏
TG
16:29
Toby Ganger
definitely under $200 now
E
16:29
EdoFazlinovic
Just buy more. Pump my bags
TG
16:29
Toby Ganger
it's just an observation about short term movements...
16:30
Deleted Account
I can see no mention of batches in the github. Anyone care to explain, point me in the right direction?
18:35
Deleted Account
100Th hashrate now. Totally insane
AK
18:39
A K
100 ??
[
18:39
[Riki]
100???
18:40
Deleted Account
50
18:40
Not 100
[
18:41
[Riki]
I see 18
AK
18:51
A K
Are we stuck on 28007 though?
AK
19:11
A K
40 minutes since last block ?
19:11
Not 50 ths )
DY
19:56
Demi Yilmaz
In reply to this message
Zack this gets stuck on height 28000 and doesnt update more..
IP
20:32
I P
In reply to this message
look at btc and all those massive long liquidations on good news about ICE
20:57
Deleted Account
Zack lookup account in explorer.html is broken. I think the reason is format.js is not loaded.
Z
21:58
Zack
In reply to this message
Thanks for the warning
MF
22:03
Mr Flintstone
yeah wow, looks like we picked up a bunch of hash rate
22:05
Deleted Account
In reply to this message
Almost feels like any pump these days is just a setup for whales to exit
Z
22:10
Zack
looks like we are stuck on 28000 because the checkpoint is too recent, so the retargeting algorithm at 28000 can't access all the historical data.
OK
22:14
O K
Are we?
MF
22:14
Mr Flintstone
I see 19 top height peers at 28020
Z
22:15
Zack
I meant that the light node is stuck at 28000. the full node is fine.
OK
22:17
O K
Aha
MF
22:18
Mr Flintstone
weird, I am able to pull headers up to 28020, though maybe I am using a light node that wasn’t updated
22:19
Yep, that’s why
Z
22:55
Zack
For now I remove checkpointing from the light node.
It seems like it will be easier to set up after 28100 when the retargeting algorithm is different.
22:58
veoscan says you haven't found any of the recent 15 blocks.
I think something is wrong with your mining pool.
OK
22:58
O K
We have been talking, I think veopool.pw is operating normally
Z
22:59
Zack
It looks like someone connected some massive hashpower to amoveopool for just a short period to find those 15 blocks.
22:59
So maybe veopool.pw is fine.
OK
22:59
O K
The hashpower is still there, the current block is an unlucky one
Z
23:00
Zack
great
23:05
veopool.pw found a block. looks like it is working fine.
23:48
Deleted Account
We use mveo now?
S
23:49
Sy
yeah as usual, not sure if unlucky or something wrong
23:50
Deleted Account
everyone running pool and exchanges Pls change into mveo ASAP
23:50
That makes me feel rich
MF
23:51
Mr Flintstone
“Units have been switched from VEO to miliVEO or mVEO. 1 VEO = 1000 mVEO”
23:52
Zack this should be closer to where you send transactions
23:55
maybe even make it a different color ? Like red
23:55
people who use the wallet need to know its mveo as close to 100% of the time as we can get
OK
23:57
O K
In reply to this message
Red is good, you could make the input field red with the message
23:57
Can't miss it then
Z
23:57
Zack
A lot of different places in the wallet are using mVEO.
6 August 2018
MF
00:00
Mr Flintstone
we can put it multiple places
00:00
I feel like it is better to be conservative with this
00:00
as long as it isn’t degrading the functionality of the wallet
Z
00:01
Zack
Feel free to make pull requests
MF
00:01
Mr Flintstone
ok
А
00:43
Андрюхин
So new diff adjustment will start at 28100 block ?
DY
00:45
Demi Yilmaz
In reply to this message
Yes
А
00:45
Андрюхин
thanks
Deleted invited Deleted Account
IP
01:48
I P
do i need to update my node before 28100?
Z
01:53
Zack
you will fall out of sync if you don't update before 28100.
01:53
you need to resync all the old blocks after updating.
TG
02:06
Toby Ganger
does anyone know if the units will change on amoveo exchange and qtrade as well?
02:06
and if that will get passed on to reporting places like Blockfolio and Coinmarketcap?
OK
02:06
O K
In reply to this message
Yes it was discussed to them
02:06
In reply to this message
So yes that is the plan 😄
TG
02:06
Toby Ganger
👍
OK
02:07
O K
Should we fund a rap song about dominant assurance contracts with a dominant assurance contract now?
02:07
😄
TG
02:07
Toby Ganger
as if "welcome to the blockchain" wasn't niche and nerdy enough
OK
02:08
O K
😂
TG
02:09
Toby Ganger
will Blockfolio update units automatically through the API or how does that work? and any potential for Coinmarketcap listing?
MF
02:13
Mr Flintstone
In reply to this message
looks like an application was submitted at one point
02:13
2 weeks ago or so
DY
02:23
Demi Yilmaz
In reply to this message
Cmc is focused on their new api. They’ve been beta testing it for a while andlaunched last week. Its hard to get in contwct with them on anythin other than their new api
TG
02:31
Toby Ganger
so i guess it's wait and see
03:05
Deleted Account
I was hoping the name of mveo would still be veo...
03:06
Is it possible to see how many votes «bad question» is getting?
03:07
Sorry: which bets are placed on «bad question»
Z
03:12
Zack
In reply to this message
From which rapper?
03:13
In reply to this message
You can see the state of the Oracle from the light node.
OK
03:13
O K
In reply to this message
I was talking about Toby, but if he doesn't want the work I think someone else would suffice
03:13
I hear some of the Wu Tang people were at the crypto event at anarchapulco
03:14
We could go even more meta, and make the song about wishing someone would open up a dominant assurance contract to pay them to make a rap song
Z
03:15
Zack
The regulatory fiat blues
03:15
In reply to this message
That's a good idea
OK
03:15
O K
In reply to this message
😂
Z
03:15
Zack
A song funded by dominant assurance contract, about being paid by a dominant assurance contract to write this song.
OK
03:16
O K
💯
Z
03:20
Zack
It could talk about how musicians no longer have to put up with the record industry, iTunes, sound cloud, or lawyers.

How all the customers use torrents to download the song.
03:21
Getting paid by a global audience wihout having to deal with local copywrite laws.
[
03:21
[Riki]
We need a DJ cryptopher columbus
03:22
Deleted Account
Haha, cool idea! Maybe get a whole crew (not just one dude), makes it more legit
Z
03:24
Zack
Easiest strategy is probably to change the lyrics of some existing song that we can get the instrumental version of.
03:26
How about a dominant assurance contract to upload the most expensive textbook as a torrent?
I think it would bring even more attention to these contracts than a rap.
OK
03:29
O K
I like both
Z
03:45
Zack
How can we modify our current market to work as a dominant assurance contract?

I think we should start with with something simple, and then add features based on user experience.

So to start, lets have the entrepreneur, builder, and market maker all be the same person.
We can use a normal binary oracle and binary market, with one additional restriction.
We only accept bets at 9:1 odds. So the market will always say there is a 10% chance that the song will be published for free as a torrent.

This way the market maker who starts with $100 of VEO can earn up to $900 from publishing the song.

You can read about Paul Sztorc's design for blockchain dominant assurance contracts here: http://bitcoinhivemind.com/papers/3_PM_Applications.pdf

His design has more features that we should consider for future iterations.
TG
03:58
Toby Ganger
I'd be down to do the song given the compensation is right...I'd also have to do a lot more technical research
03:59
but i wonder about these sort of things being sort of "preaching to the choir" just from experience with "Welcome To The Blockchain"...the only people who really listen are the ones who already know
Z
04:00
Zack
In reply to this message
you can decide the price for yourself, and you don't even have to tell us what your price is.

But, if you choose to not release the song, you will have to pay the community 1/10th as much money as they raised.

I can help you run a server for this for free. so all the risk and reward is on you.
TG
04:02
Toby Ganger
I can discuss it with some of the producers i work with and see if someone wants to give it a go...
04:02
and if the song doesn't come out well then the community will create an assassination market for me huh?
Z
04:03
Zack
if the song isn't so good, then you will have a hard time raising money this way in the future.
TG
04:04
Toby Ganger
But i'll be balling with my mVeo stash
Z
04:07
Zack
I wonder what Amoveo's first killer use case will be.
Selling music?
DY
04:15
Demi Yilmaz
In reply to this message
It must be solving a 10x problem. Is music 10x?
Z
04:17
Zack
A lot of money is lost to lawyers, copywrite holders, etc.

Free music is more than 10x better than paying for soundcloud or itunes.
04:18
I think it would work even better if we let people write their name somewhere, to show that they participated in funding the song.
DY
04:25
Demi Yilmaz
In reply to this message
True, free music is 10x better for the audience, 10x worse for the artist. Writing names sounds like a gimmick. I believe we need to fund projects related to veo community and see what it can become before trying to get people from different industries to adopt it.

I personally have not inherently understood how exactly it is going to work and I believe this is the case for most people here. We need real life shitty examples of funding before seeking to find product market fit.
Z
04:26
Zack
it might be better for the artist. They don't have to pay lawyers or publishing firms or itunes or soundcloud or youtube.
TG
04:27
Toby Ganger
there is not enough money in music with music being the actual commodity being bought and sold...music is used for brand creation which can then be commodified through merch, licensing, appearances, and leveraging for other paying opportunities
DY
04:27
Demi Yilmaz
In reply to this message
Toby has a point
Jeans invited Jeans
Z
04:50
Zack
15 minutes per block.
MF
05:27
Mr Flintstone
some hash rate just left amoveopool
OK
05:33
O K
Yes, unfortunately the person who joined earlier informed me they were having ventilation issues and might be leaving
05:37
Deleted Account
hi, i would like to know if something like https://github.com/cryppadotta/dotta-license is possible with amoveo smart contract?
Z
07:03
Zack
No. Amoveo doesn't store any on-chain contract state.
07:03
Deleted Account
When halving?
Z
07:04
Zack
In reply to this message
the block reward is controlled by the governance mechanism. We cannot predict what the community will decide to do.
07:06
Deleted Account
How to trigger governance process?
07:06
By hardfork?
J
07:09
Jackie
Forgive me if this already has been mentiond; When the new diff goes live at 28100, will it reset the current diff and hence start from scratch? Or will it continue from the current diff at 34 TH
07:32
In reply to this message
it will continue with current.
EP
07:43
Evans Pan
so will we have a more or less constant 144 blocks a day after 28100?
Z
07:55
Zack
yes. that is the plan.
Deleted invited Deleted Account
EP
08:55
Evans Pan
In reply to this message
Great. You are a better leader now.
09:00
mveo and diff change algo which have been debated, discussed, argued, frustrated, abused for 3 or 4months, finally got updated. Amoveo is moving a big step forward. 👌👌👌👍👍👍
Z
10:18
Zack
Looks like the send all feature in the light wallet is not working. It tries sending too much. I will fix it soon.
T
11:17
Topab
And I can't lookup the amount of the wallet in http://78.46.149.239:8080/explorer.html
OK
11:47
O K
@pgonza that node is not in sync right now
11:47
Checkout veoscan.io
T
13:58
Topab
Ok, thanks
Deleted invited Deleted Account
16:10
Deleted Account
16:11
What's up with this function which open a channel?
17:35
Deleted Account
Veoscan is switched to mVEO.
S
17:39
Sebsebzen
nice!
17:39
@demiculus I hope that mVEO will be the standard on CMC
DY
17:55
Demi Yilmaz
In reply to this message
👍
Z
18:52
Zack
In reply to this message
Use the light node for channels.
HT
20:01
Han Tuzun
Could somebody that were involved in implementing the light wallet mVEO switch have a look at Discord's #general channel please?
Z
21:59
Zack
http://139.59.144.76:8080/explorer.html

Most of the trades have disappeared from this market.
There are no trades above price 50 any more.

Maybe we should consider canceling the difficulty update?
S
22:01
Sy
why? isnt it settled via futarchy?
Z
22:03
Zack
the futarchy market is still live, and it switched from saying "upgrade is best" to saying "undetermined"
22:04
about 80% of the money in the market was removed
OK
22:04
O K
In reply to this message
We already determined that's not what it says
Z
22:05
Zack
look for yourself. most of the trades are gone. there are no shares of True at price > 50.
OK
22:05
O K
In reply to this message
I'll remind you of the discussion
22:05
In reply to this message
And here.
Z
22:05
Zack
Talk is cheap.
OK
22:06
O K
Well, when it's not worded in a way that implies what you're now saying it implies, I'm not so sure
22:06
But at any rate I don't think we are going to unfork our nodes
Z
22:10
Zack
In reply to this message
So you think the futarchy mechanism is broken.
I encourage you to use game theory to show that this is the case.

It seems to me that this market will have a price >50 if the P(price>$300 | we do a hard update) > P(price > $300 | we do not do the hard update)
OK
22:11
O K
In reply to this message
This is already pretty clear.
MF
22:11
Mr Flintstone
the problem is now we are farther from 300 dollars
22:11
so while it may be a good update and drive the price up, it may not increase it over 300 usd
Z
22:11
Zack
If the futarchy mechanism is not broken, and we are actively choosing to do the opposite of what futarchy says we should do... it seems hypocritical.
Why are we building a futarchy tool if we wont use it ourselves?
OK
22:12
O K
In reply to this message
Just open a question that is better
Z
22:12
Zack
In reply to this message
can you show this mathematically?

Can you at least show that my formula is wrong?
OK
22:12
O K
In reply to this message
Here you go
Z
22:12
Zack
It seems to me that this market will have a price >50 if the P(price>$300 | we do a hard update) > P(price > $300 | we do not do the hard update)

Can you prove this formula wrong?
MF
22:13
Mr Flintstone
this is correct
Z
22:14
Zack
If the formula is correct, then that means the futarchy mechanism is telling us that the upgrade will have a non-determined effect on the price.

Additionally, since we started doing this update, the price has fallen significantly on qtrade and amoveo.exchange.
OK
22:14
O K
In reply to this message
False.
22:14
You set a *specific* price
Z
22:14
Zack
Maybe the noisy part of the community likes this upgrade, but it seems the rest of the community does not want it.
OK
22:15
O K
Yeah. There won't be much noise left if you keep talking about the only people left like that
Z
22:16
Zack
The futarchy market will have a price >50 if P(price VEO>$300 | we do a hard update) > P(price VEO > $300 | we do not do the hard update)

You agree with this statement OK? if not, can you use math to show that it is false?
22:16
You know about conditional probability format right?
22:16
P(A|B) = the probability that A will happen, given that B has happened.
MF
22:17
Mr Flintstone
In reply to this message
what if the market has a price > 50 if veo price is 250 in question,
but less than 50 if veo price is 300?
Z
22:20
Zack
In reply to this message
If there was a market to show this was the case, then futarchy would be telling us we should do the update.

The only futarchy market running right now says "undetermined". it isn't clear if the upgrade will make it better or worse.
22:21
Someone added another bet to the market.
Now it says we should do the update.
22:21
Lucky!
MF
22:22
Mr Flintstone
let’s see if someone matches it
OK
22:22
O K
In reply to this message
About as well as you read English
Z
22:24
Zack
In reply to this message
Great!
It is a sign Amoveo is good technology. The people who get involved are already knowledgeable about so many related tools.
S
22:31
Sebsebzen
Dollar predictions in a bear market...
22:33
I hope next thing is to find a futarchy way to fund development
22:33
But price prediction is not good way IMO
Z
22:34
Zack
I don't want to fund development if the development isn't going to increase the price of VEO.
OK
22:35
O K
It's true that in the future it would be better to use a crypto pairing instead of a USD pairing
22:36
$300 is an arbitrary number, why 300, why not $10,000
22:36
Then if the price would only go to 9,999 from the change we shouldn't make the change
Z
22:39
Zack
In reply to this message
I think you still aren't understanding some critical piece of the oracle.

Since you already understand conditional probabilities, maybe it is confusion about the meaning of correlation: https://en.wikipedia.org/wiki/Correlation_and_dependence
22:39
It is also possible I have a flaw in my math.
22:40
the wikipedia article is a little intense.
Correlation is simpler for us because we only have 4 outcomes.
22:41
given the 4 probabilities of the 4 outcomes, we can calculate the correlation this way:
(TT + FF) - (TF + FT)
22:42
I embedded this formula into the oracle, so the price of the market should reflect the correlation of the 2 questions embedded in the oracle.
S
22:42
Sebsebzen
With a small marketcap like VEO, price is depended on too many external factors. Such as BTC price
Z
22:42
Zack
In reply to this message
Why would that matter?
S
22:43
Sebsebzen
Because VEO will be traded in BTC pairing
22:43
*pairings
Z
22:43
Zack
This doesnt make a difference for the effectiveness of Amoveo's futarchy markets.
S
22:43
Sebsebzen
And also general market sentiment will have an impact
Z
22:44
Zack
I don't see how that would matter.
22:44
Random information unrelated to the market.
OK
22:45
O K
In reply to this message
Please explain why in my example we would make the change?
S
22:46
Sebsebzen
Maybe it would be better to ask the question what BTC or ETH rate will VEO achieve if A or B
Z
22:46
Zack
Whether we use $300 vs $10 000 as the over-under, we should still get the same result.

(assuming the update is a good update)
The difference is that with $300 the equilibrium price will be something like 60 or 70, and with $10 000 the equilibrium price will be 50.0001, which is harder to measure.
22:47
In reply to this message
That does not make a difference.
OK
22:47
O K
In reply to this message
👍
S
22:48
Sebsebzen
Maybe I’m missing something. But if BTC drops half in value, wouldn’t that affect the bet?
OK
22:48
O K
In reply to this message
This still seems to be the case, crypto pairings are better than USD parings
Z
22:48
Zack
if the update is good, then
P(VEO > $10 000 | we do the update) >= P(VEO > $10 000 | we do not update)
So if the price will be above 50.

I guess if the market decided there is 0 probability of a $10 000 veo, then the price could be exactly 50.
22:49
In reply to this message
No. That would not matter.
S
22:49
Sebsebzen
So if I’d answer the bet right now. I’d factor in my general bearish sentiment for crypto market and BTC prices
OK
22:49
O K
In reply to this message
Sure it does, because cryptos are priced in BTC and ETH
Z
22:50
Zack
In reply to this message
Can you mathematically show what the difference would be? It seems to me we use the same formula and get the same result whether we measure in BTC or USD.
OK
22:50
O K
That means the bids and asks can remain the same on Amoveo's exchanges, and the USD "price" will go down, even though nothing changed
S
22:50
Sebsebzen
There is some USD denominated otc trading
22:50
But if exchanges go live the crypto pairings will be more important
OK
22:50
O K
In reply to this message
Because it could still correlate with an increase in Veo's price in BTC
S
22:50
Sebsebzen
Also if crypto prices go up, people have more money to spend and invest in VEO
OK
22:51
O K
but if BTC loses half USD value, it would not correlate with the same increase in USD
S
22:51
Sebsebzen
Same if it goes down
Z
22:51
Zack
So you think that the BTC-USD exchange rate is correlated with our decision on whether to update VEO or not?
I highly doubt that.
22:51
Most bitcoin holders aren't paying attention to our hard updates.
OK
22:52
O K
Most crypto people price 'altcoins' in BTC
22:52
fact.