BitShares Forum
Main => Stakeholder Proposals => Topic started by: xeroc on July 12, 2017, 11:28:55 am
-
(http://www.bitshares.foundation/img/logo.png)
Hello community,
it is our pleasure to let everyone know that ChainSquad has now created
the worker 201707-bsip18 for Peter Conrad.
This worker is endorsed and supported by the BitShares Blockchain Foundation!
All the details can be found on this constantly updated page:
http://www.bitshares.foundation/workers/2017-07-peter-conrad
Please consider voting for it!
Steem discussion: http://steem.link/yJByx
-
Voted. :-)
-
Fully supported. This is awesome!
-
Awesome!! This has been a long time coming, and was (in my eyes) really a pain point and a weakness of the BTS blockchain. Glad to see this is finally getting fixed, and glad to see that pc is on it! +5%
-
how many ~hours and people are we talking about?
-
Awesome!! This has been a long time coming, and was (in my eyes) really a pain point and a weakness of the BTS blockchain. Glad to see this is finally getting fixed, and glad to see that pc is on it! +5%
technically, @pc is a contractor on the bitshares foundation ... that's not to say he could and should become a board member too, but i don't want to mix up to many things at the same time ..
how many ~hours and people are we talking about?
Currently, we are three - the foundation is running for almost a year or so .. but we want to extend the team and board ..
This is probably going to happen over August ..
-
Awesome!! This has been a long time coming, and was (in my eyes) really a pain point and a weakness of the BTS blockchain. Glad to see this is finally getting fixed, and glad to see that pc is on it! +5%
technically, @pc is a contractor on the bitshares foundation ... that's not to say he could and should become a board member too, but i don't want to mix up to many things at the same time ..
how many ~hours and people are we talking about?
Currently, we are three - the foundation is running for almost a year or so .. but we want to extend the team and board ..
This is probably going to happen over August ..
so 3 are working full time on the bsip?
-
bsips are only one thing to work on .. it just happend that the first worker is for a particular bsip ..
-
Voted 8)
-
bsips are only one thing to work on .. it just happend that the first worker is for a particular bsip ..
but this thread is about this one specific BSIP, isn't it? My questions are only about BSIP 18
-
bsips are only one thing to work on .. it just happend that the first worker is for a particular bsip ..
but this thread is about this one specific BSIP, isn't it? My questions are only about BSIP 18
Oh .. peter is implementing it .. once done chainsquad will take care for auditing and code quality .. as well as managing the worker through an escrow .. soon, two more people will be involved in escrow
Does that answer your question?
-
This will be a great worker for the community.
Voted for this.
-
bsips are only one thing to work on .. it just happend that the first worker is for a particular bsip ..
but this thread is about this one specific BSIP, isn't it? My questions are only about BSIP 18
Oh .. peter is implementing it .. once done chainsquad will take care for auditing and code quality .. as well as managing the worker through an escrow .. soon, two more people will be involved in escrow
Does that answer your question?
yes, thank you
-
voted, @pc is great!
-
I'm in! Thanks everybody! :-)
-
congratulations! it is great to have you :) all the best. anything i can do for you please just let me know.
-
Status update
I have just created two PRs on github, completing the first two milestones from my proposal:
- https://github.com/bitshares/bitshares-core/pull/340 - this is the initial implementation of the features discussed in the proposal. Unit tests are included (and pass, obviously). The code is able to replay the current chain. I have also let the hardfork pass on a local node, without problems (but also without consequences, of course). I urge developers and witnesses to review this PR!
- https://github.com/bitshares/bitshares-core/pull/339 - since my worker proposal includes support during the hardfork, and (hopefully unnecessary) debugging should any problems arise, and due to experience gained from the recent blockchain halt, this PR intends to improve startup time after a node crashes. This should speed up debugging and restart significantly, should it become necessary. This change is not consensus-related and can be applied immediately by node operators who feel brave enough. :-)
As of today, the worker has accumulated about 88k BTS, which would at today's price on CMC cover about half of my requested pay.
Next steps:
- wait for review of PRs
- make adjustments, if any are deemed necessary
- integrate into testnet after PR has been accepted
-
Final report
The hardfork containing the BSIP-18 implementation paid by this worker was activated today. So far, everything seems to be working as expected.
* People are now able to settle bitAssets that are in black swan state. The first settlement happened soon after the hardfork time, in block http://bitshares-explorer.io:9009/#/blocks/22475777 .
* Some witnesses have reactivated price feeds for most of the bitAssets that were in black swan state. The following assets have been revived instantly after a sufficient number of feeds became available: CASH.USD, GRIDCOIN, HKD, NZD, RUB, SEK, SGD and TUSD. KRW will probably follow soon, it currently has 6 of 7 required feeders.
* GOLOS has enough feeders, but is still waiting for collateral bids.
* It seems to be consensus among witnesses that holders of GOLD and SILVER should be given some time to settle, before price feeds are reactivated. At the current BTS price, SILVER would be revived instantly, while GOLD would require additional collateral.
Unfortunately it seems that @alt hasn't upgraded, as he previously announced. His witness node dropped out immediately after the fork. Please remove your votes.
My work on this has now been completed. I will request payment from the BitShares Foundation.
-
Unfortunately it seems that @alt hasn't upgraded, as he previously announced. His witness node dropped out immediately after the fork. Please remove your votes.
How did it happen guys that you didn't work out the fork such that it does not piss off anybody? Alt was a very good witness and he shut down btsbots together with his witness node, which is a lot of pity...
-
Final report
The hardfork containing the BSIP-18 implementation paid by this worker was activated today. So far, everything seems to be working as expected.
* People are now able to settle bitAssets that are in black swan state. The first settlement happened soon after the hardfork time, in block http://bitshares-explorer.io:9009/#/blocks/22475777 .
* Some witnesses have reactivated price feeds for most of the bitAssets that were in black swan state. The following assets have been revived instantly after a sufficient number of feeds became available: CASH.USD, GRIDCOIN, HKD, NZD, RUB, SEK, SGD and TUSD. KRW will probably follow soon, it currently has 6 of 7 required feeders.
* GOLOS has enough feeders, but is still waiting for collateral bids.
* It seems to be consensus among witnesses that holders of GOLD and SILVER should be given some time to settle, before price feeds are reactivated. At the current BTS price, SILVER would be revived instantly, while GOLD would require additional collateral.
Unfortunately it seems that @alt hasn't upgraded, as he previously announced. His witness node dropped out immediately after the fork. Please remove your votes.
My work on this has now been completed. I will request payment from the BitShares Foundation.
congrats on the successful HF!
-
How did it happen guys that you didn't work out the fork such that it does not piss off anybody? Alt was a very good witness and he shut down btsbots together with his witness node, which is a lot of pity...
We came up with a detailed description of the proposed changes 6 months ago. The BSIP-18 proposal was discussed here in the forum, all questions that came up were answered, all concerns addressed (AFAIK). Then a worker was created and was quickly voted in, meaning the proposal has been approved by a majority of the shareholders. The proposal was implemented as planned, and was extensively tested in testnet. All witnesses were supposed to participate in testing, were repeatedly asked to do so, and had lots of time to do it. Then we created a release and announced the hardfork, 5 weeks before the scheduled time. We even postponed the release for another 10 days after discovering a bug.
AFAICS we did everything exactly right.
Then, a mere 5 days ago, this message appeared:
btsbots.com will not support the fork because of the unreasonable trade logic.
No explanation what was meant with "unreasonable trade logic". Apparently there were discussions in various other channels. I have heard about claims that the release contained "malicious code injection", and that your "funds are not safe on BitShares". Again, without any explanations what exactly the problem might be.
I have been told that several chinese-speaking community members talked to @alt - it seems that the "unreasonable trade logic" is that holders of SILVER can settle their bitassets against the settlement_fund. I don't see how that is unreasonable. Quite the opposite, I think this is the way how things are intended to work. Nobody else who I talked to really understands alt's point of view, or supports his position.
I'd like to point out that at "T - 5 days" you can't simply stop such a hardfork. Almost everybody who runs a full node will most likely already have upgraded at that point. We've announced the release 5 weeks before the planned date, precisely because it takes time to get the information out to an unknown number of people.
How did it happen guys that you didn't work out the fork such that it does not piss off anybody?
Please tell me what we could have done better. Maybe it will help next time.
-
Please tell me what we could have done better. Maybe it will help next time.
It looks like you did pretty much everything correctly from organizational perspective (can't judge how reasonable is technical implementation). Very sad that miscommunication with alt has happened (I don't blame anybody by no means).
-
pretty sure alt is old enough to raise concerns within 5 months, so no idea what's the issue here. I'm following most development on bitshares, and this one was executed very good in my opinion
-
His account is trading very actively now, which means that he does not find trading logic to be too unreasonable.
-
look at market SILVER/BTS
what a joke
-
There are no price feeds for silver yet. I guess, markets does what it whats.
-
the force settle shouldn't be allowed before the feed price worked.
then SILVER would not become such a Joke.
who made this silly logic in the core?
and who take responsibility for it?
no, in fact the share holders think this logic is right.
they have done another awesome worker.
congratulations to the Great Bitshares community.
-
the force settle shouldn't be allowed before the feed price worked.
then SILVER would not become such a Joke.
who made this silly logic in the core?
and who take responsibility for it?
no, in fact the share holders think this logic is right.
they have done another awesome worker.
congratulations to the Great Bitshares community.
What are you implying? Did someone get away with an improper trade/settlement and make "free" money? Can you share some proof of any issues?
-
Guys, let's calm down and discuss the issue a civilized way. If mistakes were made, lets's identify them clearly and work out a solution on how to correct them.
-
just don't trade it until the market is fully working? I still don't understand the issue here
-
we don't need to fix it because it already past.
or I should say we can't fix it because it already happen.
when we got enough price feed, the SILVER would be settlement at the price feed.
the point is there is a flaw before the price feed acitve,
but our share holders and core team don't think it's a flaw.
they will screwed up again.
what ever you think it's a flaw or not,
I don't care now. I have lost my pacient to argue with anybody about these things.
Guys, let's calm down and discuss the issue a civilized way. If mistakes were made, lets's identify them clearly and work out a solution on how to correct them.
-
the point is there is a flaw before the price feed acitve,
but our share holders and core team don't think it's a flaw.
they will screwed up again.
If this is a flaw, this has to be corrected, such that when the next black swan happens, we don't have the same problem. @pc, what do you think?
-
If this is a flaw, this has to be corrected, such that when the next black swan happens, we don't have the same problem. @pc, what do you think?
I would call this "flaw" by what it is: "free markets". There is one "issue" and that is, BTS price changed quite a bit since most of the settled assets have black swanned.
That means that those that held on their long (e.g. bitSILVER) can now (finally) again settle into BTS. That settlement happens at the price of black swan (as has been the promise of BitShares from day one).
However, a bug prevented those longs from settling if they didn't settle within 24h after the black swan. Now, finally, longs can settle. Actual price of silver changed, settlement price did not.
Also worth noting, there are **NO SHORTERS** on bitSILVER anymore. After a black swan, all collateral is held by committee-account.
-
So, it looks like nobody else see any problems with the current black swan revival logic. At least nobody among the biggest proxies. In this case, DPOS rules require to leave it like it is, for good or for bad. This is fair, isn't it?
-
If this is a flaw, this has to be corrected, such that when the next black swan happens, we don't have the same problem. @pc, what do you think?
SILVER has been an extreme case (one of several actually, but the only one with significant volume), because BTS has gone up quite a lot since it black swanned.
This *should* not happen again, because in the future witnesses will simply continue to publish a price feed, and with a feed the new mechanisms that were implemented with BSIP-18 will revive such bitassets long before the situation becomes so extreme.
-
let me tell you what had happend to SILVER first.
1. based BSIP18, when the fork happend, the SILVER holders can ask for a settlement at blackswan price immediately.
it's about 4000BTS/SILVER.(but the real price is about 1000BTS/SILVER, we have about 9000 SILVER available in the market.)
this will not stop only when we have enough efficient feed price from witness(maybe 7 or more).
2. about 5 days before the fork, I know this rule from somebody, I think it's unreasonable, and talk to others. but others think this rule is right.
after some discuss, I am disapoint and give up.
3. somebody who have know this rules begin buy SILVER from market, the SILVER price up to 3000 BTS/SILVER
4. when the fork happen, people sold SILVER, got BTS at price 4000 BTS/SILVER.
5. the witness act very slowly, after 24hours later, we got enough feed price from them. and the blood stop
6. somebody said nobody lose anyting at this accident. because these BTS which have sold belong to the system. so all BTS holders lose.
here is some question to the community:
1. do you think this is a flaw? or it's a right logic? personally I think it's very easy to avoid this from the code, the settlement shouldn't be allowd unless we got enough price feeds.
2. did you descipe all the rules changed because of BSIP18 really clearly? did you descipe this rule?where have you discuss these?
3. did the share holders really know these rule? and who reviewed these rules? who make the decition to change rule like this?
4. who will take responsibility for the fault?
5. how do you avoid next mess up?
-
You might want to add that the original promise of bitassets was that in the event of a global settlement/black swan, those that are long on that asset can settle it into collateral at black swan price.
A bug prevented them from being able to do that, so that they were stuck with their long position.
That said, BSIP18 did *NOT CHANGE* any rules but fixed the behavior of the code to match the intention of all documentation around bitassets.
Discussion here:
https://github.com/bitshares/bsips/blob/master/bsip-0018.md#bugfix-mpas-that-have-seen-a-global-settlement-cannot-be-settled-after-the-price-feed-expires
https://github.com/cryptonomex/graphene/issues/664#issuecomment-254056746
https://bitsharestalk.org/index.php/topic,24322.0.html
http://steem.link/yJByx
and above in this thread
If you need someone to blame, then blame me.
Is this a loss for the ecosystem? I don't think so. Why? Because we have revived bitassets which IMHO is a huge signal.
Did any trader lose? No, because they have been "settled" at black swan time
Did any long lose? No. They actually gained independent of the black swan price, because they can finally settle their long as was promised
Did BTS holders pay the price, yes of course. The network had the bug that prevented longs from setteling to begin with.
Also, if it wasn't for the "bug", someone who was long on SILVER and forgot about it for 3 months would now see the same with BSIP18 as without BSIP18, with the only difference that he is now happier because he **CAN** settle into BTS.
-
a train was broken and can't move.
you pay somebody repair it, so it moved, and hit to a shit car.
you said the train should move always, the shit car shoud have be hit if the train no broken.
we can't blame the repai man, we should thanks to the reapir man.
how stupid are you? or how shameless are you? why don't you wait for the shit car pass then start the train?
you are the speake man of Bitshares? or the speak man of workers?
what a fuck defend.
you only know spend money, never take any responsibility. bull shit
You might want to add that the original promise of bitassets was that in the event of a global settlement/black swan, those that are long on that asset can settle it into collateral at black swan price.
A bug prevented them from being able to do that, so that they were stuck with their long position.
That said, BSIP18 did *NOT CHANGE* any rules but fixed the behavior of the code to match the intention of all documentation around bitassets.
Discussion here:
https://github.com/bitshares/bsips/blob/master/bsip-0018.md#bugfix-mpas-that-have-seen-a-global-settlement-cannot-be-settled-after-the-price-feed-expires
https://github.com/cryptonomex/graphene/issues/664#issuecomment-254056746
https://bitsharestalk.org/index.php/topic,24322.0.html
http://steem.link/yJByx
and above in this thread
If you need someone to blame, then blame me.
Is this a loss for the ecosystem? I don't think so. Why? Because we have revived bitassets which IMHO is a huge signal.
Did any trader lose? No, because they have been "settled" at black swan time
Did any long lose? No. They actually gained independent of the black swan price, because they can finally settle their long as was promised
Did BTS holders pay the price, yes of course. The network had the bug that prevented longs from setteling to begin with.
Also, if it wasn't for the "bug", someone who was long on SILVER and forgot about it for 3 months would now see the same with BSIP18 as without BSIP18, with the only difference that he is now happier because he **CAN** settle into BTS.
-
keep it civil here.
-
It may be just me, but i still don't get what the problem is ?!
-
1. do you think this is a flaw? or it's a right logic?
I think it is the right logic. When a black swan happens, the peg is broken, and holders of the bitasset can settle their holdings at the price of the least collateralized short. This is the defined behaviour of our smart coins, see for example https://bitshares.org/technology/price-stable-cryptocurrencies/ :
If this revaluation happens faster than the short positions can be forced to cover, then all SmartCoins are liquidated at the exchange rate of the least collateralized short position. This is similar to an insolvent bank converting its deposits to equity.
The fact that the backing collateral increased significantly in value since the black swan is irrelevant. With the black swan, the collateral put into the settlement fund effectively belongs to the holders of the smartcoins.
2. did you descipe all the rules changed because of BSIP18 really clearly? did you descipe this rule?where have you discuss these?
Yes, see BSIP-18 (emphasis added):
When a market-pegged asset undergoes a global settlement, [,,,] the outstanding debt (the BitAsset long positions) are still collateralized through the settlement pool at the fixed settlement price. It is even possible that the value of the collateral exceeds the nominal value of the MPA significantly.
Bugfix: MPAs that have seen a global settlement cannot be settled after the price feed expires
It has turned out that force-settling an MPA requires a valid price feed even when the MPA has a settlement_price set. This is clearly a bug, since in that case the settlement price is independent from the price feed. Furthermore, publishing price feeds is no longer possible after a global settlement, so the time when settlement is possible at all is limited to the expiration period of the price feed of the MPA.
This bug will be fixed. See https://github.com/cryptonomex/graphene/issues/664#issuecomment-254056746 for a discussion.
BSIP-18 was discussed here: https://bitsharestalk.org/index.php/topic,24322.0.html .
3. did the share holders really know these rule? and who reviewed these rules? who make the decition to change rule like this?
The shareholders have approved of these rules in the usual process, i. e. through a worker proposal. I can only assume that everyone involved did their due diligence.
4. who will take responsibility for the fault?
5. how do you avoid next mess up?
There is neither fault nor mess up.
-
I think the principle rule should be make SILVER worth 1 SIVLER.
you make SILVER worth x4, it's the same wrong to make it /4,
but obviously the share holders don't care about it.
this is why I quit.
enjoy your fun.
-
I think the principle rule should be make SILVER worth 1 SIVLER.
you make SILVER worth x4, it's the same wrong to make it /4,
but obviously the share holders don't care about it.
this is why I quit.
enjoy your fun.
Not every share holder understands all the technical aspects around the BitShares project. Not everyone will agree on all paths forward however the intention of this BSIP was positive for BitShares. Companies make decisions like this all the time, trade off's, opportunity costs, are all part of growing an ecosystem.
Why quit when BitShares and the blockchain space is finally coming alive? You have been a dedicated member to the community, we all appreciate your help and feedback.