BitShares Forum
Main => General Discussion => Topic started by: R on September 16, 2016, 03:41:05 pm
-
Hey,
In episode #173 of BeyondBitcoin (https://steemit.com/beyondbitcoin/@officialfuzzy/guest-signups-beyond-bitcoin-hangout-173-9-16-16-rsvp-to-friday-s-hangout-refer-projects-for-rewards (https://steemit.com/beyondbitcoin/@officialfuzzy/guest-signups-beyond-bitcoin-hangout-173-9-16-16-rsvp-to-friday-s-hangout-refer-projects-for-rewards)) I raised the following questions:
- Should we increase the quantity of BTS witnesses?
- Should we introduce witness vote degradation?
In BTSX we had 101+ witnesses, now we have 27 witnesses in BTS. Currently, witnesses earn approx $200/month. If in the future BTS increases significantly in value these positions will be earning more. The reason why I ask if we should increase the witness count is primarily because of the current witnesses slow reaction to implement price feeds.
Why I ask if we should introduce witness vote degradation:
- Campaigning for these witnesses positions has become almost non existent at this point (no witnesses in beyondbitcoin hangouts asking for votes for example)
- The witness sub-forum (https://bitsharestalk.org/index.php/board,61.0.html (https://bitsharestalk.org/index.php/board,61.0.html)) has become a ghost town (witness channel on steemit isn't populated with BTS witness posts).
- Witnesses once voted in, unless their node degrades, are unlikely to be replaced by another prospective BTS witness. If we had voting degradation where after 'x' months your placed vote on a witness begins degrading over a period of 'y' months, we would see witnesses who actively campaign potentially becoming witnesses, replacing inactive BTS witnesses.
- Getting current witnesses to implement price feeds for new MPAs is quite difficult, I have been actively campaigning for witnesses to publish price feeds for the 'GRIDCOIN' MPA (https://bitsharestalk.org/index.php/topic,22733.0.html (https://bitsharestalk.org/index.php/topic,22733.0.html)). After 2.5 months and 3 BeyondBitcoin hangout appearances only 2 witnesses have published price feeds for the GRIDCOIN MPA. New witnesses may be more likely to interact with the BTS community.
- By introducing voting degradation, we may increase regular participation in the voting system rather than a 'vote once and forget' mindset right now.
During episode #173, crypto.prometheus raised the issue of due to the current low network voting participation, perhaps an individual armed with $500k they could vote their own witnesses into power - whilst true, on the other hand if an attacker had significant funds they could likely convince an active witness to sell their account (since they only get $200/month, if someone comes along with $10k+ they may be very tempted to take the offer).
Thoughts? No official proposal has been created in the client, just an informal discussion at this point.
Matching Steem post: https://goo.gl/s7xMiT (https://goo.gl/s7xMiT)
Best regards,
CM.
-
I think you have missed the point that CryptoPrometheus was trying to make.
Voting decay means the overall level of votes will go down, due to voter apathy. That means a large BTS holder can vote in not only one witness for himself, but as many as he likes, taking over control of the blockchain.
Note that with 27 active witnesses, buying out 51% of them witnesses will cost at least 14 times as much as buying out one, whereas you currently need less than twice as many BTS to vote 14 in than you need to vote one in.
-
I think you have missed the point that CryptoPrometheus was trying to make.
Voting decay means the overall level of votes will go down, due to voter apathy. That means a large BTS holder can vote in not only one witness for himself, but as many as he likes, taking over control of the blockchain.
Note that with 27 active witnesses, buying out 51% of them witnesses will cost at least 14 times as much as buying out one, whereas you currently need less than twice as many BTS to vote 14 in than you need to vote one in.
I get that voter apathy may be a concern in the future, but at the moment the low level of voting participation in the first place is more of a concern than voter apathy IMO.
At the moment, when you're voted in & your node is running you're very unlikely to be at risk of losing your elected witness position (a potential centralization risk). I don't think that users will become apathetic to voting if their vote degrades over say a 6 month period.
We should be continuously driving users to vote (and re-evaluate their votes after a certain period of time) within the Bitshares network. We should be brainstorming new ideas regarding how we can drive up voter participation in the future, such as holding informal fun polls (favourite colour, community mvp polls, etc).
The Gridcoin network is also experiencing a lack of voting participation - only 1/7th of total vote weight votes which brings the validity of poll results mandate into question.
-
I like the idea of accountability too. It is very sad to see so few witnesses assisting the Gridcoin MPA, it should be easy to do and we will have more and more communities coming onboard so we need an active set of witnesses.
I am watching the Witness report on Steem and I really like the effort. Unfortunately, It requires a lot of time and effort to kick that off. If the report can be posted on Steem, the analyst would get at least a compensation for his work.
Are we missing the right tools for voting or just the reporting?
@crypto123 which two witness are supporting you right now? Lets give them a push in votes!
-
Information is still scattered between here and steem thus may make it difficult for witnesses to keep up to date on information.
-Could there be something built into the GUI when creating a new token that sends a bts with memo to the current witnesses for the feed to be considered?
-Perhaps an additional tab next to worker proposals for this situation to get feeds added by witnesses?
-
I like the idea of accountability too. It is very sad to see so few witnesses assisting the Gridcoin MPA, it should be easy to do and we will have more and more communities coming onboard so we need an active set of witnesses.
I am watching the Witness report on Steem and I really like the effort. Unfortunately, It requires a lot of time and effort to kick that off. If the report can be posted on Steem, the analyst would get at least a compensation for his work.
Are we missing the right tools for voting or just the reporting?
@crypto123 which two witness are supporting you right now? Lets give them a push in votes!
The witnesses that have begun providing price feeds for the Gridcoin MPA are 'delegate.ihashfury' and 'wackou', thanks :)
Here's the witness thread: https://bitsharestalk.org/index.php/topic,22733.0.html
I'll look into another couple steem posts regarding the Gridcoin MPA, our onboarding process/experience thus far and about black swan events for low trading volume cryptocurrencies (looks like the GRIDCOIN MPA is dead for now..).
-
Quick update regarding the 'GRIDCOIN' MPA - The Gridcoin price fluctuated, triggered a black swan state and the asset has been stuck in said state ever since: https://bitsharestalk.org/index.php/topic,22733.msg299091.html#msg299091
It has however opened the door to new developments by PC regarding MPA resiliency to black swan events. Once these new changes are in effect, the Gridcoin community will reactivate the market pegged asset, requiring more witnesses to produce the price feed.
Prior to the black swan event we only had 2 or 3 witnesses out of 25 (27?), whilst changing the witness count may be a dangerous idea the concept of vote weight degradation over time for witnesses and committee members is still sound.
What does everyone think?
-
Quick update regarding the 'GRIDCOIN' MPA - The Gridcoin price fluctuated, triggered a black swan state and the asset has been stuck in said state ever since: https://bitsharestalk.org/index.php/topic,22733.msg299091.html#msg299091
It has however opened the door to new developments by PC regarding MPA resiliency to black swan events. Once these new changes are in effect, the Gridcoin community will reactivate the market pegged asset, requiring more witnesses to produce the price feed.
Prior to the black swan event we only had 2 or 3 witnesses out of 25 (27?), whilst changing the witness count may be a dangerous idea the concept of vote weight degradation over time for witnesses and committee members is still sound.
What does everyone think?
I do think a UIA fits more in Gridcoin's situation, perhaps with a special owner/committee like the STEALTH asset (check https://cryptofresh.com/u/stealth-mgmt , the owner is a multi-sig account, automatically controlled by top 5 holders)
-
I do think a UIA fits more in Gridcoin's situation, perhaps with a special owner/committee like the STEALTH asset (check https://cryptofresh.com/u/stealth-mgmt , the owner is a multi-sig account, automatically controlled by top 5 holders)
I was hoping that we can have a hard fork to make this particular sub-feature of STEALTH be used for any asset .. that would be super super cool!!!
-
I do think a UIA fits more in Gridcoin's situation, perhaps with a special owner/committee like the STEALTH asset (check https://cryptofresh.com/u/stealth-mgmt , the owner is a multi-sig account, automatically controlled by top 5 holders)
I was hoping that we can have a hard fork to make this particular sub-feature of STEALTH be used for any asset .. that would be super super cool!!!
yes, i chose to hide the name of the asset too. i assume that's what u mean.
a bitshares stealth transaction = "unknown" sent n "unknown" to "unknown".
first mentioned here: https://twitter.com/kenCode_de/status/796350090856267776
-
I do think a UIA fits more in Gridcoin's situation, perhaps with a special owner/committee like the STEALTH asset (check https://cryptofresh.com/u/stealth-mgmt , the owner is a multi-sig account, automatically controlled by top 5 holders)
I was hoping that we can have a hard fork to make this particular sub-feature of STEALTH be used for any asset .. that would be super super cool!!!
yes, i chose to hide the name of the asset too. i assume that's what u mean.
a bitshares stealth transaction = "unknown" sent n "unknown" to "unknown".
first mentioned here: https://twitter.com/kenCode_de/status/796350090856267776
Thats not what we mean.
we mean that an asset's issuer is defined as a multisig group of the X biggest shareholders
-
Quick update regarding the 'GRIDCOIN' MPA - The Gridcoin price fluctuated, triggered a black swan state and the asset has been stuck in said state ever since: https://bitsharestalk.org/index.php/topic,22733.msg299091.html#msg299091
It has however opened the door to new developments by PC regarding MPA resiliency to black swan events. Once these new changes are in effect, the Gridcoin community will reactivate the market pegged asset, requiring more witnesses to produce the price feed.
Prior to the black swan event we only had 2 or 3 witnesses out of 25 (27?), whilst changing the witness count may be a dangerous idea the concept of vote weight degradation over time for witnesses and committee members is still sound.
What does everyone think?
I do think a UIA fits more in Gridcoin's situation, perhaps with a special owner/committee like the STEALTH asset (check https://cryptofresh.com/u/stealth-mgmt , the owner is a multi-sig account, automatically controlled by top 5 holders)
We've already got an EBA (OPEN.GRC), issuing an UIA would inflate the GRC coinsupply (unlikely to fly with the GRC network unless we burnt GRC before project-raining an GRC UIA).
We're waiting for the blackswan hardening features to be implemented within BTS early this year, when the changes are implemented we will revive the GRIDCOIN MPA even if it's risky.
This is off topic though, back to the main point - vote degredation!
People are talking about increasing the block rewards for witnesses to encourage more active/productive witnesses: https://bitsharestalk.org/index.php/topic,23702.0.html
Surely degrading votes over say 4-6 months would encourage witnesses to continuously campaign for their positions and open the door for new witnesses to potentially gain power?
Getting witnesses to add price feeds for the GRIDCOIN MPA was like pulling teeth - witnesses should be far more active & inactivity should have consequences, even if their witness node is healthy.
-
(....)
We should be brainstorming new ideas regarding how we can drive up voter participation in the future, such as holding informal fun polls (favourite colour, community mvp polls, etc).
The Gridcoin network is also experiencing a lack of voting participation - only 1/7th of total vote weight votes which brings the validity of poll results mandate into question.
Why not give up the idea of voting.
Maybe we should make the algorithm to vote.
Eg.
A candidate for the witness would have to meet certain requirements.
This is my suggestion:
The maximum possible total score to achieve 1000
The first criterion: proof of geographical dispersion.
The maximum possible score to achieve 300
we want a large scatter in different countries and on different continents
(We need a developers discussion how to achieve it.)
The second criterion: proof of resources (computing power) and net speed
The maximum possible score to achieve 300
Threshold of 50%, if not achieved, the candidate is eliminated completely.
The third criterion: proof of commitment
The candidate must provide prices feeds for MPA
The maximum possible score to achieve 300
The fourth criterion: proof of reputation
The maximum possible score to achieve 100
Points as a result of vote
The first (eg.) 50 candidates will be witnesses, and receives a salary.
The next (eg.) Of 150 candidates will not be witnesses and receives some less pay for it that they are ready to become a node in case of failure of other nodes
-
Quick update regarding the 'GRIDCOIN' MPA - The Gridcoin price fluctuated, triggered a black swan state and the asset has been stuck in said state ever since: https://bitsharestalk.org/index.php/topic,22733.msg299091.html#msg299091
It has however opened the door to new developments by PC regarding MPA resiliency to black swan events. Once these new changes are in effect, the Gridcoin community will reactivate the market pegged asset, requiring more witnesses to produce the price feed.
Prior to the black swan event we only had 2 or 3 witnesses out of 25 (27?), whilst changing the witness count may be a dangerous idea the concept of vote weight degradation over time for witnesses and committee members is still sound.
What does everyone think?
I do think a UIA fits more in Gridcoin's situation, perhaps with a special owner/committee like the STEALTH asset (check https://cryptofresh.com/u/stealth-mgmt , the owner is a multi-sig account, automatically controlled by top 5 holders)
We've already got an EBA (OPEN.GRC), issuing an UIA would inflate the GRC coinsupply (unlikely to fly with the GRC network unless we burnt GRC before project-raining an GRC UIA).
We're waiting for the blackswan hardening features to be implemented within BTS early this year, when the changes are implemented we will revive the GRIDCOIN MPA even if it's risky.
This is off topic though, back to the main point - vote degredation!
People are talking about increasing the block rewards for witnesses to encourage more active/productive witnesses: https://bitsharestalk.org/index.php/topic,23702.0.html
Surely degrading votes over say 4-6 months would encourage witnesses to continuously campaign for their positions and open the door for new witnesses to potentially gain power?
Getting witnesses to add price feeds for the GRIDCOIN MPA was like pulling teeth - witnesses should be far more active & inactivity should have consequences, even if their witness node is healthy.
I think vote degradation may be a problem if there isn't a ton of voter participation to begin with. Also, I think @pc made a good argument against degradation earlier in the thread.
By the way, I wonder if witnesses have been reluctant to offer price feeds for Gridcoin MPA because they think MPA is the wrong way to go. I tend to agree with that. A cryptocuyrrency is not the best use case for an MPA, especially an MPA that will have difficult gaining liquidity. Why not issue a UIA? You could get rid of your blockchain and replace the coins with a committee-controlled UIA. Then you could focus on maintaining/developing your BOINC software rather then futzing around with a blockchain, which has nothing to do with your core competency. Anyway, just my 2 cents.
-
(....)
We should be brainstorming new ideas regarding how we can drive up voter participation in the future, such as holding informal fun polls (favourite colour, community mvp polls, etc).
The Gridcoin network is also experiencing a lack of voting participation - only 1/7th of total vote weight votes which brings the validity of poll results mandate into question.
Why not give up the idea of voting.
Maybe we should make the algorithm to vote.
Eg.
A candidate for the witness would have to meet certain requirements.
This is my suggestion:
The maximum possible total score to achieve 1000
The first criterion: proof of geographical dispersion.
The maximum possible score to achieve 300
we want a large scatter in different countries and on different continents
(We need a developers discussion how to achieve it.)
The second criterion: proof of resources (computing power) and net speed
The maximum possible score to achieve 300
Threshold of 50%, if not achieved, the candidate is eliminated completely.
The third criterion: proof of commitment
The candidate must provide prices feeds for MPA
The maximum possible score to achieve 300
The fourth criterion: proof of reputation
The maximum possible score to achieve 100
Points as a result of vote
The first (eg.) 50 candidates will be witnesses, and receives a salary.
The next (eg.) Of 150 candidates will not be witnesses and receives some less pay for it that they are ready to become a node in case of failure of other nodes
This is really along the lines of a conversation that took place in Telegram yesterday (today for some people). It makes a lot of sense. I think the main thing missing is a bond i.e. locking away a certain amount of BTS in order to qualify as a witness. Some people said it should be minimum of 1M BTS. I'm personally not sure it should be mandatory. But I do think it should at least be part of a scoring system such as the one you laid out. Actually, come to think of it, there should probably be mandatory minimums for all criteria, and anything over the minimum raises the score. Anyway, I made a couple of comments next to some of your criteria (see comments in red below).
By the way, i also like your idea of having some small pay for backup witnesses. I was talking about that in Telegram a week or so ago. Anyway, since you suggest they should get some pay, I assume they would be producing some blocks. If so, what percentage, maybe 5-10%? This would be a good opportunity for back up witnesses (or "witness candidates") to earn some reputation and perhaps prove he can reach some initial scoring thresholds.
The first criterion: proof of geographical dispersion.
The maximum possible score to achieve 300
we want a large scatter in different countries and on different continents
(We need a developers discussion how to achieve it.)
This seems tricky, could leave it out at least initially.
The second criterion: proof of resources (computing power) and net speed
The maximum possible score to achieve 300
Threshold of 50%, if not achieved, the candidate is eliminated completely.
Does this include latency? And what about taking server uptime into account? How feasible is it to measure these metrics? Well, latency seems straightforward. But what about the others?
The third criterion: proof of commitment
The candidate must provide prices feeds for MPA
The maximum possible score to achieve 300
Maybe this should include missed blocks? And perhaps call this proof of ability or proof of work. Proof of commitment should perhaps be locking funds, no?
The fourth criterion: proof of reputation
The maximum possible score to achieve 100
Points as a result of vote
The first (eg.) 50 candidates will be witnesses, and receives a salary.
The next (eg.) Of 150 candidates will not be witnesses and receives some less pay for it that they are ready to become a node in case of failure of other nodes
-
Quick update regarding the 'GRIDCOIN' MPA - The Gridcoin price fluctuated, triggered a black swan state and the asset has been stuck in said state ever since: https://bitsharestalk.org/index.php/topic,22733.msg299091.html#msg299091
It has however opened the door to new developments by PC regarding MPA resiliency to black swan events. Once these new changes are in effect, the Gridcoin community will reactivate the market pegged asset, requiring more witnesses to produce the price feed.
Prior to the black swan event we only had 2 or 3 witnesses out of 25 (27?), whilst changing the witness count may be a dangerous idea the concept of vote weight degradation over time for witnesses and committee members is still sound.
What does everyone think?
I do think a UIA fits more in Gridcoin's situation, perhaps with a special owner/committee like the STEALTH asset (check https://cryptofresh.com/u/stealth-mgmt , the owner is a multi-sig account, automatically controlled by top 5 holders)
We've already got an EBA (OPEN.GRC), issuing an UIA would inflate the GRC coinsupply (unlikely to fly with the GRC network unless we burnt GRC before project-raining an GRC UIA).
We're waiting for the blackswan hardening features to be implemented within BTS early this year, when the changes are implemented we will revive the GRIDCOIN MPA even if it's risky.
This is off topic though, back to the main point - vote degradation!
People are talking about increasing the block rewards for witnesses to encourage more active/productive witnesses: https://bitsharestalk.org/index.php/topic,23702.0.html
Surely degrading votes over say 4-6 months would encourage witnesses to continuously campaign for their positions and open the door for new witnesses to potentially gain power?
Getting witnesses to add price feeds for the GRIDCOIN MPA was like pulling teeth - witnesses should be far more active & inactivity should have consequences, even if their witness node is healthy.
I think vote degradation may be a problem if there isn't a ton of voter participation to begin with. Also, I think @pc made a good argument against degradation earlier in the thread.
By the way, I wonder if witnesses have been reluctant to offer price feeds for Gridcoin MPA because they think MPA is the wrong way to go. I tend to agree with that. A cryptocuyrrency is not the best use case for an MPA, especially an MPA that will have difficult gaining liquidity. Why not issue a UIA? You could get rid of your blockchain and replace the coins with a committee-controlled UIA. Then you could focus on maintaining/developing your BOINC software rather then futzing around with a blockchain, which has nothing to do with your core competency. Anyway, just my 2 cents.
If there's not a ton of voter participation to begin with then surely DPOS is not an effective mechanism for voting witnesses into/out of power? Currently witnesses are voted in and unless they turn malicious/incompetent they never lose their position, thus competition stalls and witnesses contribute the minimum required of their role. By introducing vote degradation(over a long time say 6 months), witnesses would have to at least post an update once in a while to show they're still alive in the community. It'd revitalize the democratic aspect of DPOS.
Should it really be up to witnesses to make decisions that prevent alternative cryptocurrencies from utilizing the full features of bitshares? If it was like pulling teeth for GRC, will all other alternative cryptocurrencies face opposition from these individuals? Surely we want as many assets being traded on the DEX as possible, no? Had we never attempted the MPA, BTS would not be pursuing improvments to blackswan handling: https://github.com/bitshares/bitshares-core/issues/216 This'd have been far worse if it had affected bitUSD.
I've mentioned in previous posts why we won't be pursuing an UIA. Gridcoin is a seperate community & project from BOINC, we do not contribute towards its development and by issuing an UIA for BOINC computation we would be introducing centralization where there is currently decentralization. There will certainly be BOINC UIAs issued in the future (project rain), but GRC will not be migrating to an UIA.
Edit: Relevant github issue: https://github.com/bitshares/bitshares-ui/issues/29
-
(....)
We should be brainstorming new ideas regarding how we can drive up voter participation in the future, such as holding informal fun polls (favourite colour, community mvp polls, etc).
The Gridcoin network is also experiencing a lack of voting participation - only 1/7th of total vote weight votes which brings the validity of poll results mandate into question.
Why not give up the idea of voting.
Maybe we should make the algorithm to vote.
Eg.
A candidate for the witness would have to meet certain requirements.
This is my suggestion:
The maximum possible total score to achieve 1000
The first criterion: proof of geographical dispersion.
The maximum possible score to achieve 300
we want a large scatter in different countries and on different continents
(We need a developers discussion how to achieve it.)
The second criterion: proof of resources (computing power) and net speed
The maximum possible score to achieve 300
Threshold of 50%, if not achieved, the candidate is eliminated completely.
The third criterion: proof of commitment
The candidate must provide prices feeds for MPA
The maximum possible score to achieve 300
The fourth criterion: proof of reputation
The maximum possible score to achieve 100
Points as a result of vote
The first (eg.) 50 candidates will be witnesses, and receives a salary.
The next (eg.) Of 150 candidates will not be witnesses and receives some less pay for it that they are ready to become a node in case of failure of other nodes
This is really along the lines of a conversation that took place in Telegram yesterday (today for some people). It makes a lot of sense. I think the main thing missing is a bond i.e. locking away a certain amount of BTS in order to qualify as a witness. Some people said it should be minimum of 1M BTS. I'm personally not sure it should be mandatory. But I do think it should at least be part of a scoring system such as the one you laid out. Actually, come to think of it, there should probably be mandatory minimums for all criteria, and anything over the minimum raises the score. Anyway, I made a couple of comments next to some of your criteria (see comments in red below).
By the way, i also like your idea of having some small pay for backup witnesses. I was talking about that in Telegram a week or so ago. Anyway, since you suggest they should get some pay, I assume they would be producing some blocks. If so, what percentage, maybe 5-10%? This would be a good opportunity for back up witnesses (or "witness candidates") to earn some reputation and perhaps prove he can reach some initial scoring thresholds.
The first criterion: proof of geographical dispersion.
The maximum possible score to achieve 300
we want a large scatter in different countries and on different continents
(We need a developers discussion how to achieve it.)
This seems tricky, could leave it out at least initially.
The second criterion: proof of resources (computing power) and net speed
The maximum possible score to achieve 300
Threshold of 50%, if not achieved, the candidate is eliminated completely.
Does this include latency? And what about taking server uptime into account? How feasible is it to measure these metrics? Well, latency seems straightforward. But what about the others?
Latency - Yes!
Server uptime - Yes!
How feasible is it to measure these metrics? Well, latency seems straightforward. But what about the others?
Maybe network could send some random( task ) to calculate, once peer random(X days, XX days ). The task should bey predictable within 1s - 10s max seconds (maybe something like pice of "Multicore Super Pi Benchmark" but CPU + RAM )
It could also be that some post-witness-installation, or trustless self test
The third criterion: proof of commitment
The candidate must provide prices feeds for MPA
The maximum possible score to achieve 300
Maybe this should include missed blocks? And perhaps call this proof of ability or proof of work. Proof of commitment should perhaps be locking funds, no?
agree
The fourth criterion: proof of reputation
The maximum possible score to achieve 100
Points as a result of vote
The first (eg.) 50 candidates will be witnesses, and receives a salary.
The next (eg.) Of 150 candidates will not be witnesses and receives some less pay for it that they are ready to become a node in case of failure of other nodes
it's just outline
-
(....)
We should be brainstorming new ideas regarding how we can drive up voter participation in the future, such as holding informal fun polls (favourite colour, community mvp polls, etc).
The Gridcoin network is also experiencing a lack of voting participation - only 1/7th of total vote weight votes which brings the validity of poll results mandate into question.
Why not give up the idea of voting.
Maybe we should make the algorithm to vote.
Eg.
A candidate for the witness would have to meet certain requirements.
This is my suggestion:
The maximum possible total score to achieve 1000
The first criterion: proof of geographical dispersion.
The maximum possible score to achieve 300
we want a large scatter in different countries and on different continents
(We need a developers discussion how to achieve it.)
The second criterion: proof of resources (computing power) and net speed
The maximum possible score to achieve 300
Threshold of 50%, if not achieved, the candidate is eliminated completely.
The third criterion: proof of commitment
The candidate must provide prices feeds for MPA
The maximum possible score to achieve 300
The fourth criterion: proof of reputation
The maximum possible score to achieve 100
Points as a result of vote
The first (eg.) 50 candidates will be witnesses, and receives a salary.
The next (eg.) Of 150 candidates will not be witnesses and receives some less pay for it that they are ready to become a node in case of failure of other nodes
This is really along the lines of a conversation that took place in Telegram yesterday (today for some people). It makes a lot of sense. I think the main thing missing is a bond i.e. locking away a certain amount of BTS in order to qualify as a witness. Some people said it should be minimum of 1M BTS. I'm personally not sure it should be mandatory. But I do think it should at least be part of a scoring system such as the one you laid out. Actually, come to think of it, there should probably be mandatory minimums for all criteria, and anything over the minimum raises the score. Anyway, I made a couple of comments next to some of your criteria (see comments in red below).
By the way, i also like your idea of having some small pay for backup witnesses. I was talking about that in Telegram a week or so ago. Anyway, since you suggest they should get some pay, I assume they would be producing some blocks. If so, what percentage, maybe 5-10%? This would be a good opportunity for back up witnesses (or "witness candidates") to earn some reputation and perhaps prove he can reach some initial scoring thresholds.
The first criterion: proof of geographical dispersion.
The maximum possible score to achieve 300
we want a large scatter in different countries and on different continents
(We need a developers discussion how to achieve it.)
This seems tricky, could leave it out at least initially.
The second criterion: proof of resources (computing power) and net speed
The maximum possible score to achieve 300
Threshold of 50%, if not achieved, the candidate is eliminated completely.
Does this include latency? And what about taking server uptime into account? How feasible is it to measure these metrics? Well, latency seems straightforward. But what about the others?
Latency - Yes!
Server uptime - Yes!
How feasible is it to measure these metrics? Well, latency seems straightforward. But what about the others?
Maybe network could send some random( task ) to calculate, once peer random(X days, XX days ). The task should bey predictable within 1s - 10s max seconds (maybe something like pice of "Multicore Super Pi Benchmark" but CPU + RAM )
It could also be that some post-witness-installation, or trustless self test
The third criterion: proof of commitment
The candidate must provide prices feeds for MPA
The maximum possible score to achieve 300
Maybe this should include missed blocks? And perhaps call this proof of ability or proof of work. Proof of commitment should perhaps be locking funds, no?
agree
The fourth criterion: proof of reputation
The maximum possible score to achieve 100
Points as a result of vote
The first (eg.) 50 candidates will be witnesses, and receives a salary.
The next (eg.) Of 150 candidates will not be witnesses and receives some less pay for it that they are ready to become a node in case of failure of other nodes
it's just outline
Do you want to rewrite your outline with the changes? In the meantime, I'm asking @Chris4210 if he can put this on his radar.
-
There some interesting ideas here, certainly worth considering. I do think it needs better definition tho, especially regarding technical requirements and "diversity", which should also include vendor independence.
Another thing, glad to see info being brought over from the fast moving and hard to keep up with the *DEX Telegram channel.
I'm firmly of the opinion that any conversation of substance needs to be represented here. Telegram is good, useful, just not so much for hashing out such details and proposals. OK for them to start wherever, but I can't keep up with convos there 24x7. It's good for real time, not very good unless you are online and participating AS THE CONVO TAKES PLACE.
We are a global community spanning every timezone. This forum has its' limitations, but geography and time are not among them.
-
@Taconator specifically mentioned this thread on the BitShares specific mumble this morning, but unfortunately no further posts have been made here in almost 2 weeks.
@xeroc has weighed in now that he opposes an increases in witness pay, tho he has not explained why in detail.
He did mention he believes the feed accuracy problem is no longer a problem, but again no data to support that perspective.
I'm beginning to believe there is a lot of reliance on the info from Telegram or other sources as being more influential that this forum. I find that very disturbing. As I've said, Telegram is a r e a l t i m e only platform and it's not good for making decisions and hashing out proposals.
This forum's biggest limitation is its antiquated UI / UX. It is even better than Telegram regarding the multilingual problem, yet Telegram is the most popular platform for discussions.
Is there a typical time of day that people tend to appear to talk on Telegram? I was on it last night & it was totally dead. Perhaps if some of you that use Telegram frequently would chime in here and give us your impression of good times to tune in there where you see the most activity.
-
It seems I failed to make my self clear: For sure, I would love to see more active Witnesses, but given how much people have contributed in the past for free in their spare time, I oppose throwing money at witnesses for just being witnesses and not actually doing anything for the money they get except the "very basic" things. I'd rather see competition among witnesses and fresh blood.
It's not perse that I am against raising the pay, but, I would prefer them to show more efforts FIRST!
-
It seems I failed to make my self clear: For sure, I would love to see more active Witnesses, but given how much people have contributed in the past for free in their spare time, I oppose throwing money at witnesses for just being witnesses and not actually doing anything for the money they get except the "very basic" things. I'd rather see competition among witnesses and fresh blood.
It's not perse that I am against raising the pay, but, I would prefer them to show more efforts FIRST!
@Thom - I agree with you as well, this forum (ugly UI or not) is still very important. I actually find it quite difficult at times to keep up with all the chattiness going on in Telegram. It will suck valuable minutes from your day if you let it.
As for raising Witness pay.. I am against it:
If there are Witnesses running a profitable business in expensive places like Canada and America, then it becomes very obvious that a Witness node can be run pretty much anywhere else on Earth at a profit too. Hence, no payraise is needed for Witnesses at this time.
If your business cannot turn a profit, that should not be the Stakeholder's problem.
-
[As for raising Witness pay.. I am against it:
If there are Witnesses running a profitable business in expensive places like Canada and America, then it becomes very obvious that a Witness node can be run pretty much anywhere else on Earth at a profit too. Hence, no payraise is needed for Witnesses at this time.
If your business cannot turn a profit, that should not be the Stakeholder's problem.
You're kidding right? You seriously think the profits returned from a witness constitutes a business? It may cover the cost of a couple of 4GB VPS accounts but not a dedicated server, let alone 2. Granted right now the time investment is on the minimal end of the spectrum, and a competent IT tech / engineer can easily manage it, but I wouldn't call it much of a business. Even if witness pay was $300 USD a month that's only $3600 a year, hardly a living wage for one person in the west so that hardly qualifies as a sustainable business.
If you want to attract more witnesses you have to raise the stakes. However, I agree with xeroc that we want to attract more qualified witnesses for which we need to establish clear metrics for performance or to test a candidate's abilities.
I thought the way it worked in the pre-graphene days with delegates worked quite well, in that those wanting to become active witnesses must first get in the que by registering a witness account, putting it into service and campaigning for votes. The community could monitor uptime, missed blocks and what value was being proposed in the voting campaign.
For example, given how much importance is now placed on price feeds, any potential candidate needs to describe how they plan to provide them, including:
1) publish frequency
2) data sources
3) self written script or some other trusted source (xeroc, alt, abit, wackou for example)
4) if self written an algorithm description (for example alt's feed script includes evaluating trades)
If you want to create an incentive for a diversity of feed producers we need to define standards for them and provide a stronger incentive to compete. I don't think there much of one now.
BTW, now that I think about it, when a forced settlement for an asset occurs, what feed price is used? There are x number of witnesses producing price feeds, are they ALL averaged or somehow aggregated into a single value? Is it the most recent price feed published?
If we look at the DEX as a business, as a DAC, witnesses are a crucial element that shouldn't be left to altruistic contributions of labor as they have in the past. Right now the cost of a couple of 4GB VPS systems and the time required to keep them maintained is very close to break even. There is little profit. There is not much dynamic range built in to handle an increase in accounts or trading activity.
-
It seems I failed to make my self clear: For sure, I would love to see more active Witnesses, but given how much people have contributed in the past for free in their spare time, I oppose throwing money at witnesses for just being witnesses and not actually doing anything for the money they get except the "very basic" things. I'd rather see competition among witnesses and fresh blood.
It's not perse that I am against raising the pay, but, I would prefer them to show more efforts FIRST!
It's unlikely you will see more competition. Look at what we see in steem. The competition is rather high because the payout is rather reasonable. Nobody gets excited about peanuts.. so those that are left doing it are either people who are filling the void for the sake of the network out of altruism, or people who are in a different economic state where the current pay rate actually is reasonable. Partially explains why more than half of the witnesses now are asian.
I rather see real global competition like what has somewhat been seen in steem. If you want to attract talent, it has to be worth their time and effort. Currently it doesn't look worth it.
-
[As for raising Witness pay.. I am against it:
If there are Witnesses running a profitable business in expensive places like Canada and America, then it becomes very obvious that a Witness node can be run pretty much anywhere else on Earth at a profit too. Hence, no payraise is needed for Witnesses at this time.
If your business cannot turn a profit, that should not be the Stakeholder's problem.
You're kidding right? You seriously think the profits returned from a witness constitutes a business?
...
There is little profit.
For some people, running a Witness node is a business, and might even be their sole source of income. I am also trying to introduce people to Bitshares just like you. Referral program, run a Witness node, trading on the DEx, building a profitable company on the platform, etc.
If a business' profits are not large enough, then that business must find a way to make that venture more profitable. As for Witness nodes, there are awesome server providers in Slovakia, Poland, Brasil, Mexico and everywhere in between, so just shop around a bit. Make the node more profitable.
Paying our Witness node operators more, will not make them any more productive or efficient.
Please don't make their profitability issues the Bitshares Stakeholder's problem, that is not fair.
-
I don't think it's reasonable to compare steem and bitshares witnesses. steem witnesses get paid a ton. But beyond true witness duties, there's a general expectation they will engage in various non-witness activities that benefit the steem ecosystem. The value of their contributions is very subjective, and what you end up with is a situation where each witness is contributing to a different degree. So in some cases, the network may get good value in exchange for the witness pay, and in some cases money is wasted.
Bitshares, on the other hand, has a worker proposal system that enables us to keep the witness duties separate while being able to scrutinize worker/development proposals and track their progress to completion. This is much more sophisticated and sensible. So please, let's maintain the separation.
Having said that, I do think the witness pay is a bit too low for this important role, which includes price feeds as one of the responsibilities. Let's not forget that if we want to be considered a decentralized exchange, our value proposition must be rooted in our offering of MPAs. Otherwise there is NOTHING decentralized about bitshares beyond the core token.
So let's define specifically what we need from witnesses, let's incorporate an objective witness reliability scoring (WRS) mechanism (e.g. server uptime, latency, missed blocks, etc) into the witness selection process (along with the current reputation-based voting), and let's increase the pay modestly in order to a) create a little more competition and b) ensure the pay is sufficient enough that witnesses won't have to cut corners. This is just reasonable and sensible. Let's not be penny wise, pound foolish.
-
For some people, running a Witness node is a business, and might even be their sole source of income. I am also trying to introduce people to Bitshares just like you. Referral program, run a Witness node, trading on the DEx, building a profitable company on the platform, etc.
Yes, I and many others are grateful for your efforts. However, I am extremely confident that you Kencode could not sustain your lifestyle on witness profits, nor could most of those that made BitShares possible. Those you speak of that can are in hugely different economic situations. I'm sure there are many of them that are highly entrepreneurial and competent. If you want to put the reliability of BitShares in the hands of such people that is fine, but you will sacrifice reliability if you don't have standards to which witnesses are measured or those such witnesses can look to for how it's done.
We in the west and others have learned the hard way through experience what it takes to run the network and how to insure decentralization, but it's the "wild wild west" in many respects and there are no solid, well documented standards.
And speaking of standards what of the standards related to freedom? We in the west understand that word and how the status quo financial institutions are inherently contrary to the principles and philosophy that underpin why BitShares and many other digital currency projects were started in the first place. I am not confident that vision would be sustained if western witnesses fade away, which might very well happen if BitShares continues to rely on altruism and on subsidies from outside interests to cover the increasing expenses of running the network.
If a business' profits are not large enough, then that business must find a way to make that venture more profitable. As for Witness nodes, there are awesome server providers in Slovakia, Poland, Brasil, Mexico and everywhere in between, so just shop around a bit. Make the node more profitable.
You can't turn wood into gold, there is no such alchemy. I agree with you in principle, but with the current rate of pay you are limiting who can compete. Unless your labor costs are very low (i.e. non western wages) and you can find reliable cost effective ultra-cheap (no DO or AWS) hosting that fits your needs (language, support, method of payment, reliability, decentralization of locations, etc etc) it's an impossible task. I know, I look for good hosting frequently.
I have limitations that others may not, I require things others may not, such as English speaking support, payment in crypto and flexible locations. Some locations are simply more costly. Try to get a VPS in Australia that excepts crypto payments and you'll see that your choices are quite limited. Same with servers in Iceland. I suspect few of the current 23 witnesses pay much attention to the physical distribution of their nodes. I suspect I'm the only one who even runs 4 nodes. If I must squeeze more out of my existing witness pay to support server upgrades I will have to shut down some nodes.
Not everyone has the same set of skills. I believe you are projecting your abilities onto others with these expectations. You are an example of top tier management with excellent business acumen that most do not possess. I just don't see how the current level of witness pay is a viable business for anyone except those of third world or low ranking economies.
If BitShares could be sustained by competent, freedom loving people from third world or low economic level places around the world whose cost of living and hosting expenses are far less I would be the first to relinquish my witness. I'm not convinced the non-western users of the BitShares community are ready enough to make that happen without sacrificing important core values BitShares was built to establish. It's like asking Thomas Jefferson to have confidence his understanding of freedom would be preserved by a high school graduate raised in the 21st century. Odds are high it won't happen.
Paying our Witness node operators more, will not make them any more productive or efficient.
Please don't make their profitability issues the Bitshares Stakeholder's problem, that is not fair.
Reliability and what BitShares is and the principles it will maintain are very much stakeholder's responsibility. I agree, raising pay does not guarantee higher performance, but odds are very high unless it is raised it will guarantee performance will not improve. You have a year of history that shows that.
-
Do you want to rewrite your outline with the changes? In the meantime, I'm asking @Chris4210 if he can put this on his radar.
These are rewrited propositions:
The maximum possible total score to achieve 1000.
TOP 21 gets 3 BTS
TOP 34 (55 or 89) (https://docs.google.com/spreadsheets/d/1YUeO7hJ35OJHV2DsY72-Ma_8MwVPeb56jMl-Hxf1EvY/edit?usp=sharing) gets 1,5 BTS
The first criterion: proof of geographical dispersion.
The maximum possible score to achieve 300
we want a large scatter in different countries and on different continents
maybe we could use net latency (to serwer X, Y, Z [some form of triangulation] to prevent VPN fraud) or some form of existing geolocation if we can trust it
We need a developers discussion how to achieve this
The second criterion: proof of ability
The maximum possible score to achieve 300
Threshold of 50%, if not achieved, the candidate is eliminated completely.
missed blocks(last month, last week, last day), server uptime
The third criterion: proof of commitment
The candidate must provide prices feeds for MPA
The maximum possible score to achieve 300
probably someone would have to evaluate the quality of the scripts – If the script is new and it is not widely used.
The fourth criterion: proof of reputation
The maximum possible score to achieve 100
Points as a result of vote
The first (eg.) 21 candidates will be witnesses, and receives 100% salary.
The next (eg.) Of 34 (55 or 89) (https://docs.google.com/spreadsheets/d/1YUeO7hJ35OJHV2DsY72-Ma_8MwVPeb56jMl-Hxf1EvY/edit?usp=sharing) candidates will not be witnesses and they will receives some less pay (maybe 50% ) for it that they are ready to become a node in case of failure of other nodes
If 1 of 10 witness is down, next 10 takes his place and recive 100% salary.
Proof of commitment, that is another topic, I see it in this this way:
Proof of commitment (http://"https://bitsharestalk.org/index.php/topic,23706.msg302068.html#msg302068") = sum of ( locking funds in orders + collateral)
-
Do you want to rewrite your outline with the changes? In the meantime, I'm asking @Chris4210 if he can put this on his radar.
These are rewrited propositions:
The maximum possible total score to achieve 1000.
TOP 21 gets 3 BTS
TOP 79 gets 1,5 BTS
The first criterion: proof of geographical dispersion.
The maximum possible score to achieve 300
we want a large scatter in different countries and on different continents
maybe we could use net latency (to serwer X, Y, Z [some form of triangulation] to prevent VPN fraud) or some form of existing geolocation if we can trust it
We need a developers discussion how to achieve this
The second criterion: proof of resources
The maximum possible score to achieve 300
Threshold of 50%, if not achieved, the candidate is eliminated completely.
missed blocks(last month, last week, last day), server uptime
The third criterion: proof of ability
The candidate must provide prices feeds for MPA
The maximum possible score to achieve 300
probably someone would have to evaluate the quality of the scripts – If the script is new and it is not widely used.
The fourth criterion: proof of reputation
The maximum possible score to achieve 100
Points as a result of vote
The first (eg.) 21 candidates will be witnesses, and receives 100% salary.
The next (eg.) Of 79 candidates will not be witnesses and they will receives some less pay (maybe 50% ) for it that they are ready to become a node in case of failure of other nodes
If 1 of 10 witness is down, next 10 takes his place and recive 100% salary.
if such a change would be accepted, I will definitely run one or more the backup nodes in Poland on variety of service providers
Proof of commitment, that is another topic, I see it in this this way:
https://bitsharestalk.org/index.php/topic,23706.msg302068.html#msg302068
-
Do you want to rewrite your outline with the changes? In the meantime, I'm asking @Chris4210 if he can put this on his radar.
These are rewrited propositions:
The maximum possible total score to achieve 1000.
TOP 21 gets 3 BTS
TOP 79 gets 1,5 BTS
The first criterion: proof of geographical dispersion.
The maximum possible score to achieve 300
we want a large scatter in different countries and on different continents
maybe we could use net latency (to serwer X, Y, Z [some form of triangulation] to prevent VPN fraud) or some form of existing geolocation if we can trust it
We need a developers discussion how to achieve this
The second criterion: proof of ability
The maximum possible score to achieve 300
Threshold of 50%, if not achieved, the candidate is eliminated completely.
missed blocks(last month, last week, last day), server uptime
The third criterion: proof of commitment
The candidate must provide prices feeds for MPA
The maximum possible score to achieve 300
probably someone would have to evaluate the quality of the scripts – If the script is new and it is not widely used.
The fourth criterion: proof of reputation
The maximum possible score to achieve 100
Points as a result of vote
The first (eg.) 21 candidates will be witnesses, and receives 100% salary.
The next (eg.) Of 79 candidates will not be witnesses and they will receives some less pay (maybe 50% ) for it that they are ready to become a node in case of failure of other nodes
If 1 of 10 witness is down, next 10 takes his place and recive 100% salary.
if such a change would be accepted, I will definitely run one or more the backup nodes in Poland on variety of service providers
Proof of commitment, that is another topic, I see it in this this way:
Thanks for restarting this thread. Hopefully the idea of more formally applying witness criteria will get more traction.
Looking at your scoring metrics, it looks like you have proof of commitment twice, once for price feeds and once for collateral. I think it makes more sense for collateral. And I think the price feeds can be grouped along with missed blocks under proof of ability. I would also add latency to that. So under proof of ability you could have: a) price feeds, b) missed blocks, and c) latency. What do you think?
It would be great if someone could write some code to measure these metrics and publish them in real time as a resource for proxies and individual voters to assess the quality of witnesses. Perhaps if this works well it could even be built into the blockchain such that if minimum thresholds are not met, then witnesses can automatically be made inactive (or standby) until they meet thresholds and get voted in again.
As for paying standby witnesses, I assume you mean in a scenario where they would randomly get to produce some % of the blocks. In the past I have suggested that it might make sense to let standby witnesses produce perhaps 10% of blocks. This would obviously require a hard fork, but maybe it should be considered. Either way, it would be nice if we could make some progress on applying criteria as discussed above.
-
I corrected my mistake in The third criterion:
it should be "proof of ability"
And I wouldn't group ability (price feeds) and resources (missed blocks(...), server uptime)
Around the globe we could have very different cases. IN some of them, we could have very talented and involved programmers with weak and expensive servers, and in the other hand we could have very good servers and lazy operators.
As for paying standby witnesses, I assume you mean in a scenario where they would randomly get to produce some % of the blocks.
Actually not (let it be as it is easier), They are needed to confirm blocks, not to produce it
and they are needed as a backup – this is very important!
We also create in this way a trained base of administrators and enter the competition.
maybe 21 + 79 is too much, maybe we could use 34 or 55 :
https://docs.google.com/spreadsheets/d/1YUeO7hJ35OJHV2DsY72-Ma_8MwVPeb56jMl-Hxf1EvY/edit?usp=sharing
-
I corrected my mistake in The third criterion:
it should be "proof of ability"
And I wouldn't group ability (price feeds) and resources (missed blocks(...), server uptime)
Around the globe we could have very different cases. IN some of them, we could have very talented and involved programmers with weak and expensive servers, and in the other hand we could have very good servers and lazy operators.
As for paying standby witnesses, I assume you mean in a scenario where they would randomly get to produce some % of the blocks.
Actually not (let it be as it is easier), They are needed to confirm blocks, not to produce it
and they are needed as a backup – this is very important!
We also create in this way a trained base of administrators and enter the competition.
I was just trying to group the metrics that pertain to how good of a job a witness is doing. Seems logical. But it can be grouped any way as long as each individual metric (not each group) has its own score attached.
As for backup witnesses, I agree with the reasons to have backups (to develop pool of competent witnesses). I also agree with incentivizing people to confirm blocks (as opposed to produce them). Actually, if I understand correctly, this is a more important function than people realize. But how do you pay for that? I mean, how do we even know if someone is confirming blocks? Right now we don't. But I've said in the past that it would be nice if the software could be modified such that the account running it could get credit for doing so. If one could earn a return on their stake like this, many people would be willing to run a node and lock up collateral. This could be the Bitshares version of a masternode. The software could even be made to include the upcoming stealth "trusted setup" to enable wide participation in that process. I think we'd get a lot of interest and could generate a lot of buzz with something like this.
-
I corrected my mistake in The third criterion:
it should be "proof of ability"
And I wouldn't group ability (price feeds) and resources (missed blocks(...), server uptime)
Around the globe we could have very different cases. IN some of them, we could have very talented and involved programmers with weak and expensive servers, and in the other hand we could have very good servers and lazy operators.
As for paying standby witnesses, I assume you mean in a scenario where they would randomly get to produce some % of the blocks.
Actually not (let it be as it is easier), They are needed to confirm blocks, not to produce it
and they are needed as a backup – this is very important!
We also create in this way a trained base of administrators and enter the competition.
maybe 21 + 79 is too much, maybe we could use 34 or 55 :
https://docs.google.com/spreadsheets/d/1YUeO7hJ35OJHV2DsY72-Ma_8MwVPeb56jMl-Hxf1EvY/edit?usp=sharing
I was just trying to group the metrics that pertain to how good of a job a witness is doing. Seems logical. But it can be grouped any way as long as each individual metric (not each group) has its own score attached.
As for backup witnesses, I agree with the reasons to have backups (to develop pool of competent witnesses). I also agree with incentivizing people to confirm blocks (as opposed to produce them). Actually, if I understand correctly, this is a more important function than people realize. But how do you pay for that? I mean, how do we even know if someone is confirming blocks? Right now we don't. But I've said in the past that it would be nice if the software could be modified such that the account running it could get credit for doing so. If one could earn a return on their stake like this, many people would be willing to run a node and lock up collateral. This could be the Bitshares version of a masternode. The software could even be made to include the upcoming stealth "trusted setup" to enable wide participation in that process. I think we'd get a lot of interest and could generate a lot of buzz with something like this.
summarizing:
new rating based elections for nodes
21 witness – 100% (3 BTS)
34 (or 55) backup nodes – 50% (~1,5 BTS) ---------- >>> https://docs.google.com/spreadsheets/d/1YUeO7hJ35OJHV2DsY72-Ma_8MwVPeb56jMl-Hxf1EvY/edit?usp=sharing
Please also comment on this post – topic : New fee structure on BitShares
https://bitsharestalk.org/index.php/topic,23706.msg302068.html#msg302068
And could you elaborate "trusted setup", is this has something to do with IPFS? I had to miss it.
-
I looked at @FractalNode's spreadsheet and the numbers on rows 6 & 7 make sense, but don't understand the results on rows 8 & 9. The number of nodes for those rows are larger than row 7, but result is less.
Although I am very much in agreement with your approach, there are a number of things to consider before it could be implemented.
This will require a hardfork. Those more familiar with the code can confirm if any of this can be done without one.
- having payout based on individual witness performance will require a hardfork
- paying standby witnesses will require a hardfork
- allowing a randomly selected standby witness to produce a block and get rewarded is an idea worth considering. If that were implemented, it would certainly require a hardfork. To insure network integrity should be limited to a very small number (1% or 2%) of the total blocks. Only the next N standby witnesses would be chosen for the random drawing, perhaps the top 10 according to voting support.
- using collateral to measure commitment is one way to measure, but not the only one. May not even be the best one. Length of service, participation in forum / Telegram, testnet participation, portion of pay reinvested into witness upgrades, contributions to code, community involvement are IMO at least equal if not a more direct measurement than how much capitol you invest. What good does it do anybody to lockup a large stake of BTS anyway? And has anyone considered the legal ramifications to the witness role? Remember, when graphene was introduced the witness role was compartmentalized to minimize their vulnerability to the regulators. Not saying a collateral affects that, and it is similar to Dash's Masternode architecture, but I'm not a tyrannical bureaucrat either. Something to consider, as is whether existing witnesses would be grandfathered in, given some period of time to produce the collateral and how will it be escrowed or locked.
I also think the 3rd criterion should involve evidence of competency to produce price feeds, perhaps via participation in the testnet and alternatively setting up some type of "feed test oracle" that isn't used for settlements or an any official capacity, but rather might serve as a point of comparison for official feeds. Now this could be done via a private website or modification to wackou's bts_tools web interface to provide feeds on the tools website but not publish them on the blockchain (which is only allowed by active witnesses).
It is also worth considering what could be done without a hardfork, perhaps though various monitoring tools or bots.
I'll have more thoughts later, but for now that''' do.
-
Hardforks can be a pain in the arse for everyone, but there are plenty of other pending issues that will require a hardfork as well, so adding yet another doesn't seem like such a big deal anymore since it would add so many nice new features and fixes to bitshares-core:
https://github.com/bitshares/bitshares-core/issues?q=is%3Aopen+is%3Aissue+label%3Ahardfork
-
I looked at @FractalNode's spreadsheet and the numbers on rows 6 & 7 make sense, but don't understand the results on rows 8 & 9. The number of nodes for those rows are larger than row 7, but result is less.
row 7 & 8 & 9 are simulations
the more (TOP) standby witness you have, less pay they get
- allowing a randomly selected standby witness to produce a block and get rewarded is an idea worth considering. If that were implemented, it would certainly require a hardfork. To insure network integrity should be limited to a very small number (1% or 2%) of the total blocks. Only the next N standby witnesses would be chosen for the random drawing, perhaps the top 10 according to voting support.
Actually I do not support this idea, we could just pay them to be on standby mode – because it is not free.
I think this is unnecessarily complicating matters and it will and degrade performance – probably the average performance of these machines will be lower.
This is the place to learn and reserve just in case
- using collateral to measure commitment is one way to measure, but not the only one. May not even be the best one. Length of service, participation in forum / Telegram, testnet participation, portion of pay reinvested into witness upgrades, contributions to code, community involvement are IMO at least equal if not a more direct measurement than how much capitol you invest.
So I gave this a bad name. The idea is to pay for increasing liquidity, by locking funds under bitASSETS and placing orders to the DEX
Topic: New Bitshares Dividend Idea (https://bitsharestalk.org/index.php/topic,23706.0.html) and my last comment on this, is on page 3 (https://bitsharestalk.org/index.php/topic,23706.msg302068.html#msg302068)
and this is not related to nodes in this topic.
I also think the 3rd criterion should involve evidence of competency to produce price feeds, perhaps via participation in the testnet and alternatively setting up some type of "feed test oracle" that isn't used for settlements or an any official capacity, but rather might serve as a point of comparison for official feeds. Now this could be done via a private website or modification to wackou's bts_tools web interface to provide feeds on the tools website but not publish them on the blockchain (which is only allowed by active witnesses).
I agree and I see your point. Thanks for that!
I think therefore that it should merge 3 and 4 criterion.
And let voters decide, on the basis of the information contained in the Witness subject on the forum.
It may lead us to the creation of the model form
with mandatory fields to fill, such as
how long you provided node services for the test network
o gosh :D
this may kill 1 and 2 criterion
if ask proper questions, like
where is you server ?
What are its parameters ?
but what happens if you suddenly lose 75% of witnesses
by some tsunami or WW3
This is the question to which we seek answers:
How to make automatic switch standby nodes to witnesses?
How to encourage the start-up and maintenance of such nodes?
And I do not like the answer: "that then it will not matter"