Author Topic: Announcement on BSIP42 relevant actions  (Read 3848 times)

0 Members and 1 Guest are viewing this topic.

Offline Thom

Re: Announcement on BSIP42 relevant actions
« Reply #30 on: October 03, 2018, 12:26:05 pm »

BitShares has been around and around trying to come up with a SmartCoin algorithm that strikes that balance but I fear due to differences in economic principles there will always be a strong tension by one camp or another to change the rules to their advantage. Remember, before the promise of redeem-ability through forced settlement people were creating sock puppet accounts and draining the reserve pool. That problem is no more.


Interesting, I wasn't here back then. Why did people have to make sock accounts to get their collateral back? Were SmartCoins backed by the reserve pool?

It was dubbed "yield harvesting". I just couldn't think of the name for it.
Injustice anywhere is a threat to justice everywhere - MLK |  Verbaltech2 Witness Reports: https://bitsharestalk.org/index.php/topic,23902.0.html

Offline Peryn

  • Newbie
  • *
  • Posts: 3
    • View Profile
Re: Announcement on BSIP42 relevant actions
« Reply #31 on: October 03, 2018, 01:38:35 pm »
Good day to all.
with current collateral greater> 1
current 1.75. bts will always be cheaper over time, striving for parity bitusd = bts * 1.75 / usd = 1/1
what we see on the graph starting in 2014

Offline bitcrab

  • Committee member
  • Hero Member
  • *
  • Posts: 1067
    • View Profile
  • BitShares: bitcrab
  • GitHub: bitcrab
Re: Announcement on BSIP42 relevant actions
« Reply #32 on: October 03, 2018, 01:42:08 pm »
Thanks @xeroc, I great appreciate your efforts to find the consensus.

First, let’s define the vision of smartcoin,  in my view, it is 1. enough accurate pegging and 2. enough liquidity. based on these smartcoin will seek wildly adoption comparing to USDT.
I think this is already where people may have a different vision.
For me (and others) the core value proposition of bitassets was:
* always fully collateralized
* settlement available to obtain collateral at a "fair" price.

With respect to settlement, the community has already moved from a
settlement at parity 1:1 with the introduction of the settlement
offset to incentivize shorters to increase liquidity.

My feeling is that those optimization criteria (tight pegging & liquidity
vs. backing and settlement) may not work out together, at least not in
the short term.

So I think we have some difference in defining smartcoin.

In my idea, smartcoin is one solution of the “stable coin” concept which are discussed a lot this year, it is designed as a token that has stable value and is easy to get and transfer when one need it, that’s why I emphasize pegging accuracy and liquidity.

It seems in your idea, smartcoin is more like something that can help people to store value, to hedge the risk in cryptocurrency market.

Let me tell 2 stories.

1.AEX and bigone has listed bitCNY and use bitCNY as the base currency in their exchange, but several months ago they removed bitCNY as base currency because “bitCNY is in serious shortage and very high premium, difficult to get, users are confused and very unsatisfactory”.

2.in August 1971, U.S. President Richard Nixon announced to remove the dollar peg to gold, which had been fixed at $35 an ounce. the Bretton Woods System collapsed.

I’d like to say, before we discussed BSIP42, bitCNY is at a similar dilemma as USD in 1971 - the supply of gold is limited, but the demand to USD increased rapidly, a big conflict comes and US cannot maintain the $35 an ounce gold price.

bitCNY(surely also bitUSD and bitEUR) is potential to get widely adopted, but before BSIP42 the supply of biCNY depend on the price of BTS, that’s why each time bear market comes, bitCNY was squeezed out quickly and lead to very serious shortage and very high premium, at the same time, continuous margin calling bring many users to nightmare.

Maybe in initial design smartcoin is more like just one asset to store value and hedge risk. But if we hope smart coin be widely used in many CEXs as base currency, and in international payments, we have to review and update the design to guarantee the pegging accuracy and liquidity, otherwise smart coins can only  be used by a small community and have no chance to let the whole world share the great innovation of Bitshares community.

force settlement enable smart coin holders to forcibly buy BTS from debt position owners, so here price play a key role, now we set forcesettlement offset and quantity limit to protect debt position owners, now in almost all the cases, to directly buy from DEX is a better way for smartcoin holders to get BTS.

bitCNY has a 5% force settlement offset for long time and this does not impact the value of bitCNY, the value of smart coin do not rely that much on force settlement. if we can not guarantee tight peg+liquidity+fair price force settlement, I believe we should give up the last one.

We both agree on bitassets having a supply side problem and that
we should allow shorters to easier borrow bitassets. According to
my optimization criteria, I still believe we must not allow bitassets
to become undercollateralized (overall - we may find a way to allow
individual positions to go CR < 1)

I agree it's OK to not allow MCR<1 in the MCR based solution.
I agree that the overall collateral ratio need to be >1, the problem is, how to handle some single margin call orders with CR<1? I think it's bad to let these single orders trigger global settlement process. can we just leave these orders there and wait they be eaten or disappear?

After BSIP applied to bitUSD the premium fell drastically, we hope we can keep this trend continue until to a close to 0 premium.
Well, arguable, that is expected behavior because we are "forcing" the marketing
to trade according to parity, we do not provide incentive for the market to
move to parity on its own. This difference is what does sit well with the Western
community.
I believe there are incentive - actually the “negative feedback” make this logic appears: if there are demands on smart coins (premium>0), there will be incentive to borrow smart coins and buy BTS, this is true incentive that we had ignored, it means if we can make more demands on smartcoins, we can help the BTS price to go up.

@Jerry, how about the both of us come to a compromise in that we allow the witnesses
to decide how to go forward and whether or not they want to run BSIP42 on bitUSD, or
not. I think, as a proxy, we shouldn't threaten them with the removal of our votes
just because they support(or not support) a BSIP that actually gives the freedom of
choice to them (the witnesses). With that said, I will retract my statement of removing
votes from BSIP2-bitUSD witnesses and would like to keep rational and
constructive discussion like this one going.

at this moment I'd like to stop "threatening" witnesses - frankly speaking I don't think this can be called "threaten", this is just to express opinion of a voter, it's always not easy to push such kind of update, so sometimes voting power is needed, I don't promise to give up the right of a proxy.




Offline abit

  • Committee member
  • Hero Member
  • *
  • Posts: 3482
    • View Profile
    • Steemit Blog
  • BitShares: abit
  • GitHub: abitmore
Re: Announcement on BSIP42 relevant actions
« Reply #33 on: October 03, 2018, 06:28:51 pm »
We can say the promises were made by BM. Then BM left. We have to make our own decisions.
It wasn't ONLY BM, we ALL were there, it was a community consensus we all (including you) have been living with for at least 3 years, so it is totally inappropriate to blame the redeem-ability or collateral requirements on BM alone.

I was not blaming BM, but describing the fact. We all know BM made his own decisions, we only got the options to accept or not. I'd say most of us didn't understand the design at the beginning and blindly accepted it or hoped it will work. But he has left, and now we need to make our own decisions, it's absurd to think we have to stick with an impossible promise forever.

Quote

According to the analysis I made earlier in this thread (please read), some of the promises were too heavy, we're unable to keep all of them. You can't have your cake and eat it. You want the 10M bitUSD in your wallet be fully backed, but in case when BTS market cap dropped to below 10M, they're simply unable to be fully backed if you, as a holder, do nothing other than hold.

Changing strategy is hard, but when it's needed, we have to do so, otherwise there is no future. I'm well aware of the risks.

I am not sure you do understand the risks. If you do, you're willing to risk others funds in this haphazardly conducted experiment.

> "haphazardly conducted"

Sorry, my English is not so good, are these words objective or subjective? Could you please use simpler words?

Quote
I do agree if both bears and bulls are promised they can't loose it's a problem. However to some degree that is indeed the situation. One or the other is going to complain there is no balance or it is skewed and needs to be "adjusted". I agree with you in your last sentence, except "they're simply unable to be fully backed if you, as a holder, do nothing other than hold." People that use leverage to trade with are used to the broken, under collateralized mainstream system and thus object to BitShares policies. Sorry, when you come to BitShares it's different for reasons, mostly the safety of funds, and you must maintain your collateral or you will be margin called. We don't need any changes that weaken investors collateral or allows fractional reserves, incentivizes accumulation of debt or skews the market via collusion or manipulation.

If course if a leveraged trader refuses to manage his collateral, well, should know the risks and not do that. It's the trader's responsibility to know the rules of the platform / game. If s/he fails to do so and the market accelerates its downward trend, the loss is on his/her back. It's not proper to blame the ecosystem for not being a mainstream institution. It wasn't intended or designed to be the same. Why can't people understand that? Do they want to repeat the mistakes all over again anew?


Due to different visions of BitShares the platform and the bitAssets. I guess bitcrab explained better than I had, have you read that? Just a few pages above.

Quote
Witnesses are paid positions, meant to contribute. Witnesses (read: including you) are the nearest to price data because they produce price feeds, and data is all over there, just need some time/efforts to collect them and summarize and show them to people who asks. Keep asking others to report the result/findings IMHO is not very good behavior.
Why is asking not good behavior? Why is providing a coherent report / summary being avoided so much?

Attitude.

Why don't you gather the data and make the report? Why I have to do that? Only I have the skills to gather data and make a report? You're my boss? I owe you something?


Quote
Why (as Thul3 comprehensively asked) are the BSIP42 advocates using such coercion and force to change the ecosystem rather than playing by the consensus rules OR make a solid case why they need to be changed that is persuasive enough to gain the consensus sought? Could it possibly be b/c our principles are not aligned? I have to ask, since you said nothing to me about them. Principles are important, do you disagree with that?

There has been many discussions in months, perhaps you didn't participate in? Is it my fault that didn't notify you? I wrote the BSIP document, I created workers for vote, it's all played by consensus rule. The workers are "voted in" / accepted. TBH I didn't get what you're asking.

Quote
If want the process to be more methodical and systematic, please propose the methods and see whether others will follow or convince them. There is a 30+page thread (and other threads) in Chinese forum shows how the algorithm is evolving: https://bitsharestalk.org/index.php?topic=26315.0 , a translator would help.

Actually some "findings" were posted, perhaps not in your favorite format:
* https://bitsharestalk.org/index.php?topic=26966.msg322398#msg322398
* https://bitsharestalk.org/index.php?topic=27170.msg322352#msg322352
* https://bitsharestalk.org/index.php?topic=26315.msg322288#msg322288
* https://bitsharestalk.org/index.php?topic=27203.msg322834#msg322834
Yes, a translator would help. The language barrier is tough for most of us. However, it isn't the place for BSIP42 opponents to fix the BSIP, it's for the authors of it to make their case with facts and data, and make sure that info is made available widely. The links you provided don't do that IMO. It's far from the data and analysis a financial system needs to make a quality engineering decision. The billions of funds invested deserve much better.

> "it's for the authors of it to.."

It's hilarious. It's actually community efforts. Just because I started the writing of a document, then it's my job to maintain it all the lifetime? As a community member, a stake holder, why don't you contribute to the document? If you don't understand Chinese, if you don't understand the charts, if you don't know how to find the data, you can hire someone to help you if you REALLY WANT to get the things done.

Quote

About redeem-ability:
I already acknowledged if inappropriate promises are made to 2 contrary positions it's a problem. You have a different take on how to resolve it than I do. You want promise A to be forgotten and I want promise B to be forgotten or reduced. Both of us want more liquidity, we're just not sure how to get it without upsetting someone.

I also think it's quite interesting that there is almost no talk about why things are the way they are now, no consideration of history. Your reference to BM wasn't analysis to facts to say why his reasoning was [all] wrong, you just used him as a scape goat. I'm not saying he was right, but you didn't show where his reasoning is wrong. WHY is an extremely important question. You may be annoyed by it, but it is essential to have the freedom to ask it and explore alternatives.

If there is indeed an "impossible trinity" of promises to community, it needs to be fixed. The question is which leg of the trinity needs to go? Why not add a 4th leg (as discussed at BitFest to publish an additional adjustment value to apply to market feeds)? We all know how unstable 3 legged chairs are.

Off topic, but we (who learnt some physics) all know how stable a triangle structure is. https://zh.scribd.com/document/199099598/Why-is-the-Triangle-the-Strongest-Shape

Quote

I advocate for a position that provides the best balance between bears and bulls. I favor Austrian economic principles over the debt based Keynsian model that uses oligarchs to centrally "manage" an economy which has over a century of demonstrating how poorly that model works to help people live freely.
BTS account: abit
BTS committee member: abit
BTS witness: in.abit

Offline sahkan

  • Full Member
  • ***
  • Posts: 188
    • View Profile
    • BitShares DEX
Re: Announcement on BSIP42 relevant actions
« Reply #34 on: October 04, 2018, 05:52:45 pm »
xeroc asked witnesses to chime in on this conversation, but majority did not. Friday is coming up and I would like to know what is the final consensus? Are we sticking with the current feed prices on BitUSD, BSIP42 or a modified BSIP42?  Starting tomorrow I will start feeding my BitUSD prices in accordance with an outcome from this conversation.

Offline xeroc

  • Board Moderator
  • Hero Member
  • *****
  • Posts: 12722
  • ChainSquad GmbH
    • View Profile
    • ChainSquad GmbH
  • BitShares: xeroc
  • GitHub: xeroc
Re: Announcement on BSIP42 relevant actions
« Reply #35 on: October 05, 2018, 02:06:11 pm »
bitCNY has a 5% force settlement offset for long time and this does not impact the value of bitCNY, the value of smart coin do not rely that much on force settlement. if we can not guarantee tight peg+liquidity+fair price force settlement, I believe we should give up the last one.
Just because people do not use force settlement does not mean that the
ability to do so provides value.
IMHO, force settlement (I settle bitCNY for its collateral) is a core
feature that provides a floor to the valuation of bitCNY.

Unless of course, you are talking about the 5% offset. Of course, if it
doesn't effect the value, nor it's liquidity, we could as well get rid
of the offset.

I agree it's OK to not allow MCR<1 in the MCR based solution.
I agree that the overall collateral ratio need to be >1, the problem is, how to handle some single margin call orders with CR<1? I think it's bad to let these single orders trigger global settlement process. can we just leave these orders there and wait they be eaten or disappear?
I believe we all agree here now.

There have been some options to deal with individual call positions that
go <100%:

* Take call positions that reach <(100+x)% away from users and give it to
  the issuer - effetively a x% penalty for the shorter.
* Allow traders to bid for the call position to provide additional
  collateral in exchange for pro-rata shares of the debt.

And possibly others.

I believe there are incentive - actually the “negative feedback” make this logic appears: if there are demands on smart coins (premium>0), there will be incentive to borrow smart coins and buy BTS, this is true incentive that we had ignored, it means if we can make more demands on smartcoins, we can help the BTS price to go up.
I think we all agree here as well. The disagreement lies in the means
that we use to achieve that goal. So far, BSIP42 meant that the price
feed formula is modified.

However, there has been plenty of discussion also instead of faking
the price feed, to use the margin call ratio as well as the short
squeeze protection ratio to provide a similar solution.

For instance, according to Peter Conrad, the current situation may well
be achieved by using the actualy/fair price feed and a short squeeze
protection ratio of 0%. That would lead to margin calls to execute at
the price feed instead of additional penalty of additional 10%. The only
drawback (that might not be one) is that the short protection ratio
cannot be negative. That means that in case there is a margin call
pending in the books, people that want bitUSD will provide a "premium"
to the market to snatch bitUSD from the margin calls.
If this becomes a problem, the price feed could still be "modified", but
at all other times, the price feed would reflect the fair price.

Give BitShares a try! Use the http://testnet.bitshares.eu provided by http://bitshares.eu powered by ChainSquad GmbH

Offline pc

  • Hero Member
  • *****
  • Posts: 1404
    • View Profile
    • Bitcoin - Perspektive oder Risiko?
  • BitShares: cyrano
Re: Announcement on BSIP42 relevant actions
« Reply #36 on: October 05, 2018, 03:01:11 pm »
For instance, according to Peter Conrad, the current situation may well
be achieved by using the actualy/fair price feed and a short squeeze
protection ratio of 0%.

Minor correction: I think that with MSSR=0 the fake price feed could be kept *closer* to the real price, not that the feed price would always have to represent the fair price exactly.
Bitcoin - Perspektive oder Risiko? ISBN 978-3-8442-6568-2 http://bitcoin.quisquis.de

Offline armin

  • Full Member
  • ***
  • Posts: 91
    • View Profile
Re: Announcement on BSIP42 relevant actions
« Reply #37 on: October 06, 2018, 08:03:04 pm »

For instance, according to Peter Conrad, the current situation may well
be achieved by using the actualy/fair price feed and a short squeeze
protection ratio of 0%. That would lead to margin calls to execute at
the price feed instead of additional penalty of additional 10%. The only
drawback (that might not be one) is that the short protection ratio
cannot be negative. That means that in case there is a margin call
pending in the books, people that want bitUSD will provide a "premium"
to the market to snatch bitUSD from the margin calls.
If this becomes a problem, the price feed could still be "modified", but
at all other times, the price feed would reflect the fair price.

From what I understand, the MSSR being >0 actually serves the purpose of offsetting the risk of holding smartcoins from going into global settlement mode. In global settlement mode, the smartcoin won't be properly pegged anymore.

The current MSSR seems to be a balancing act to make global settlement mode much less likely, as the margins calls get back to the designated MCR quickly. If the MSSR = 0, global settlement mode becomes more likely and holding smartcoins becomes much less attractive actually (which is the opposite of what bitcrab is aiming for, that is if the holders of the smartcoin realize this)

Thoughts? Maybe I'm wrong to understand that it makes global settlement more likely.

Anyhow, MSSR=0 seems like a decent experiment
« Last Edit: October 06, 2018, 08:05:22 pm by armin »

Offline bitcrab

  • Committee member
  • Hero Member
  • *
  • Posts: 1067
    • View Profile
  • BitShares: bitcrab
  • GitHub: bitcrab
Re: Announcement on BSIP42 relevant actions
« Reply #38 on: October 07, 2018, 03:58:40 am »
Just because people do not use force settlement does not mean that the
ability to do so provides value.
IMHO, force settlement (I settle bitCNY for its collateral) is a core
feature that provides a floor to the valuation of bitCNY.

Unless of course, you are talking about the 5% offset. Of course, if it
doesn't effect the value, nor it's liquidity, we could as well get rid
of the offset.

if we now reset force settlement offset of bitCNY to 0, it will greatly increase the chance to hurt the bitCNY creators and the whole system.
I haven't found a solution that can enable smartcoin holders to settle in "fair" price but at the same time do not hurt the debt position owners.


I believe we all agree here now.

There have been some options to deal with individual call positions that
go <100%:

* Take call positions that reach <(100+x)% away from users and give it to
  the issuer - effetively a x% penalty for the shorter.
* Allow traders to bid for the call position to provide additional
  collateral in exchange for pro-rata shares of the debt.

And possibly others.
make sense, I am also considering another possibility: possible to integrate the feature to force settlement: enable smartcoin holders to settle the margin call orders with CR<(100%+x) with added penalty to the debt position owners?

I think we all agree here as well. The disagreement lies in the means
that we use to achieve that goal. So far, BSIP42 meant that the price
feed formula is modified.

However, there has been plenty of discussion also instead of faking
the price feed, to use the margin call ratio as well as the short
squeeze protection ratio to provide a similar solution.

For instance, according to Peter Conrad, the current situation may well
be achieved by using the actualy/fair price feed and a short squeeze
protection ratio of 0%. That would lead to margin calls to execute at
the price feed instead of additional penalty of additional 10%. The only
drawback (that might not be one) is that the short protection ratio
cannot be negative. That means that in case there is a margin call
pending in the books, people that want bitUSD will provide a "premium"
to the market to snatch bitUSD from the margin calls.
If this becomes a problem, the price feed could still be "modified", but
at all other times, the price feed would reflect the fair price.
please do not call the current feed price as "faking price feed", BSIP42 just redefined this parameter, if it is confusing to still call it "feed price", I think we can rename it to "guide price" and show in the UI.

just setting MSSR=100% will not get the same result as current setting, as there is no "negative feedback" logic inside.

Offline bitcrab

  • Committee member
  • Hero Member
  • *
  • Posts: 1067
    • View Profile
  • BitShares: bitcrab
  • GitHub: bitcrab
Re: Announcement on BSIP42 relevant actions
« Reply #39 on: October 07, 2018, 04:18:03 am »
For instance, according to Peter Conrad, the current situation may well
be achieved by using the actualy/fair price feed and a short squeeze
protection ratio of 0%.

Minor correction: I think that with MSSR=0 the fake price feed could be kept *closer* to the real price, not that the feed price would always have to represent the fair price exactly.

under BSIP42, in bear market, maybe setting MSSR=100% will help to make the feed price more closer to market price, but in bull market, it will make the feed price farer away from the market price, as it encourage borrowing even when the smartcoin is over supplied, so need bigger negative feedback.

I think it's the best choice to move forward the MCR based solution to get the purpose of  "feed price reflect the real market price".

Offline abit

  • Committee member
  • Hero Member
  • *
  • Posts: 3482
    • View Profile
    • Steemit Blog
  • BitShares: abit
  • GitHub: abitmore
Re: Announcement on BSIP42 relevant actions
« Reply #40 on: October 07, 2018, 10:53:41 am »
Just because people do not use force settlement does not mean that the
ability to do so provides value.
IMHO, force settlement (I settle bitCNY for its collateral) is a core
feature that provides a floor to the valuation of bitCNY.

Unless of course, you are talking about the 5% offset. Of course, if it
doesn't effect the value, nor it's liquidity, we could as well get rid
of the offset.

if we now reset force settlement offset of bitCNY to 0, it will greatly increase the chance to hurt the bitCNY creators and the whole system.
I haven't found a solution that can enable smartcoin holders to settle in "fair" price but at the same time do not hurt the debt position owners.

Define "hurt"? Debt positions are attached with responsibilities, we can't deny it.

When smartcoin is oversupplied (which is reflected as trading below par value), actually borrowing more will hurt the peg thus we can say it hurts the ecosystem. In this case, in order to encourage people to borrower less, there need to be methods to reduce debt or keep debt at a certain level.
* the original approach is to encourage people to force-settle, in this case, a positive offset does the opposite;
* the BSIP42 approach (as well as a MCR adjustment approach) is to force the least collateralized positions to be margin called

Either way the cost is paid by the debt position owners.

When smartcoin is under-supplied, to maintain the peg, we encourage people to borrow more. In this case, force-settlement doesn't help and thus should be discouraged. People are still possible to buy from market at fair price even if force-settlement is disabled. The offset doesn't really matter here.

Quote

I believe we all agree here now.

There have been some options to deal with individual call positions that
go <100%:

* Take call positions that reach <(100+x)% away from users and give it to
  the issuer - effetively a x% penalty for the shorter.
* Allow traders to bid for the call position to provide additional
  collateral in exchange for pro-rata shares of the debt.

And possibly others.
make sense, I am also considering another possibility: possible to integrate the feature to force settlement: enable smartcoin holders to settle the margin call orders with CR<(100%+x) with added penalty to the debt position owners?

Please be aware that any added penalty to debt position holders when smartcoin is under-supplied (in a shortage) will push up the premium (please reread JohnR's analysis), thus will cause bigger adjustments to other parameters in order to maintain the peg.

That said, if reducing MCR to 150% and keeping MSSR at 110% resulted in a tight peg, if we change MSSR to 130% (to punish the debt position owners), it's likely we need to reduce MCR to 140% or so to keep the peg.

On the other hand, if we change MSSR to 100%, likely we can increase MCR a bit.

BTW I don't think any bidding or settling mechanism is better than margin call mechanism. BSIP42 and other MCR-based approach are essentially making more use of the margin call mechanism.

Quote

I think we all agree here as well. The disagreement lies in the means
that we use to achieve that goal. So far, BSIP42 meant that the price
feed formula is modified.

However, there has been plenty of discussion also instead of faking
the price feed, to use the margin call ratio as well as the short
squeeze protection ratio to provide a similar solution.

For instance, according to Peter Conrad, the current situation may well
be achieved by using the actualy/fair price feed and a short squeeze
protection ratio of 0%. That would lead to margin calls to execute at
the price feed instead of additional penalty of additional 10%. The only
drawback (that might not be one) is that the short protection ratio
cannot be negative. That means that in case there is a margin call
pending in the books, people that want bitUSD will provide a "premium"
to the market to snatch bitUSD from the margin calls.
If this becomes a problem, the price feed could still be "modified", but
at all other times, the price feed would reflect the fair price.
please do not call the current feed price as "faking price feed", BSIP42 just redefined this parameter, if it is confusing to still call it "feed price", I think we can rename it to "guide price" and show in the UI.

just setting MSSR=100% will not get the same result as current setting, as there is no "negative feedback" logic inside.

Will discuss below.

For instance, according to Peter Conrad, the current situation may well
be achieved by using the actualy/fair price feed and a short squeeze
protection ratio of 0%.

Minor correction: I think that with MSSR=0 the fake price feed could be kept *closer* to the real price, not that the feed price would always have to represent the fair price exactly.

under BSIP42, in bear market, maybe setting MSSR=100% will help to make the feed price more closer to market price, but in bull market, it will make the feed price farer away from the market price, as it encourage borrowing even when the smartcoin is over supplied, so need bigger negative feedback.

I think it's the best choice to move forward the MCR based solution to get the purpose of  "feed price reflect the real market price".

Peter's idea is actually manually setting a MSSR, then dynamically adjusting price feed, so there is a feedback.
Xeroc's MCR-based idea is also dynamically adjusting MCR, so it's also based on feedback.

Practically they and BSIP42 can all achieve tight pegging if the market condition is within a boundary.

The differences are:
* when smartcoin is under-supplied, setting MSSR to zero and adjusting price only (but not MCR) will result in margin calls never got eaten, thus may accumulate risks as time goes by. in UI, it will show as more and more debt positions hanging.
* adjusting MCR only (but not price) can NOT avoid globally settling when there is a black swan event. Why I think globally settling is bad is explained earlier in this post and/or other posts.
BTS account: abit
BTS committee member: abit
BTS witness: in.abit

Offline bitcrab

  • Committee member
  • Hero Member
  • *
  • Posts: 1067
    • View Profile
  • BitShares: bitcrab
  • GitHub: bitcrab
Re: Announcement on BSIP42 relevant actions
« Reply #41 on: October 07, 2018, 01:45:22 pm »
Define "hurt"? Debt positions are attached with responsibilities, we can't deny it.

When smartcoin is oversupplied (which is reflected as trading below par value), actually borrowing more will hurt the peg thus we can say it hurts the ecosystem. In this case, in order to encourage people to borrower less, there need to be methods to reduce debt or keep debt at a certain level.
* the original approach is to encourage people to force-settle, in this case, a positive offset does the opposite;
* the BSIP42 approach (as well as a MCR adjustment approach) is to force the least collateralized positions to be margin called

Either way the cost is paid by the debt position owners.

When smartcoin is under-supplied, to maintain the peg, we encourage people to borrow more. In this case, force-settlement doesn't help and thus should be discouraged. People are still possible to buy from market at fair price even if force-settlement is disabled. The offset doesn't really matter here.

when the debt position is in sufficient collateral, a smartcoin holder can forcibly buy the collateral, this is definitely hurting.

maybe this is what is called "responsibility" of debt position owners according to the initial design of BTS, but what had happened is that we have to adjust parameters now and then to avoid hurting debt position owners, and finally we found force settlement do not provide better buying condition to smartcoin holders than just buying in DEX.

in traditional financial systems, I haven't seen that the borrower's collateral can be forcibly "bought" when the collateral is sufficient.


Peter's idea is actually manually setting a MSSR, then dynamically adjusting price feed, so there is a feedback.
Xeroc's MCR-based idea is also dynamically adjusting MCR, so it's also based on feedback.

I am aware of this while replying, I just want to tell this will make things complex, we may need to set MSSR=100% at bear market while increasing it at bull market, the effect is just "make feed price closer to market price".

Offline abit

  • Committee member
  • Hero Member
  • *
  • Posts: 3482
    • View Profile
    • Steemit Blog
  • BitShares: abit
  • GitHub: abitmore
Re: Announcement on BSIP42 relevant actions
« Reply #42 on: October 07, 2018, 04:32:39 pm »
Define "hurt"? Debt positions are attached with responsibilities, we can't deny it.

When smartcoin is oversupplied (which is reflected as trading below par value), actually borrowing more will hurt the peg thus we can say it hurts the ecosystem. In this case, in order to encourage people to borrower less, there need to be methods to reduce debt or keep debt at a certain level.
* the original approach is to encourage people to force-settle, in this case, a positive offset does the opposite;
* the BSIP42 approach (as well as a MCR adjustment approach) is to force the least collateralized positions to be margin called

Either way the cost is paid by the debt position owners.

When smartcoin is under-supplied, to maintain the peg, we encourage people to borrow more. In this case, force-settlement doesn't help and thus should be discouraged. People are still possible to buy from market at fair price even if force-settlement is disabled. The offset doesn't really matter here.

when the debt position is in sufficient collateral, a smartcoin holder can forcibly buy the collateral, this is definitely hurting.

maybe this is what is called "responsibility" of debt position owners according to the initial design of BTS, but what had happened is that we have to adjust parameters now and then to avoid hurting debt position owners, and finally we found force settlement do not provide better buying condition to smartcoin holders than just buying in DEX.

in traditional financial systems, I haven't seen that the borrower's collateral can be forcibly "bought" when the collateral is sufficient.


Define "sufficient collateral"? Almost all the time all debt positions have more collateral than debt, from this perspective, they should never be "hurt" in terms of margin calls or forced settlements? Apparently this is not our rule.

If you mean 175% is sufficient, I disagree. 175% perhaps is sufficient sometimes, however, when a smartcoin is oversupplied, IMHO that means 175% is NOT sufficient; when under-supplied, I'd say 175% is too high.

So, if we adjust MCR dynamically and keep feed price unchanged, for debt position holders, it's not always "safe" to keep their positions if they have only 175% collateral ratio or even 200%.
BTS account: abit
BTS committee member: abit
BTS witness: in.abit

Offline bitcrab

  • Committee member
  • Hero Member
  • *
  • Posts: 1067
    • View Profile
  • BitShares: bitcrab
  • GitHub: bitcrab
Re: Announcement on BSIP42 relevant actions
« Reply #43 on: October 08, 2018, 09:03:06 am »
Define "sufficient collateral"? Almost all the time all debt positions have more collateral than debt, from this perspective, they should never be "hurt" in terms of margin calls or forced settlements? Apparently this is not our rule.

If you mean 175% is sufficient, I disagree. 175% perhaps is sufficient sometimes, however, when a smartcoin is oversupplied, IMHO that means 175% is NOT sufficient; when under-supplied, I'd say 175% is too high.

So, if we adjust MCR dynamically and keep feed price unchanged, for debt position holders, it's not always "safe" to keep their positions if they have only 175% collateral ratio or even 200%.

if sufficient collateral become insufficient because of feed price or MCR adjusting, and then the debt position become margin called, it's a well defined and acceptable process. no problem.

but according to the current force settlement rule, smartcoin holders can always settle the debt position with the lowest collateral ratio, regardless whether the ratio is sufficient or insufficient,  regardless whether the smartcoin is in shortage or oversupplied, I don't think that's a well designed rule, and in practice, the force settlement just provide an arbitration tool, it is not a must for smartcoin pegging.

Offline pc

  • Hero Member
  • *****
  • Posts: 1404
    • View Profile
    • Bitcoin - Perspektive oder Risiko?
  • BitShares: cyrano
Re: Announcement on BSIP42 relevant actions
« Reply #44 on: October 08, 2018, 12:20:02 pm »
1. I don't think debt positions are "hurt" by forced settlement, because settlement happens at the fair price. (But that's an old argument.)

2. It should be possible to keep a balance using smartcoin parameters. E. g. with the settlement offset the debt position receives a reward for being settled.
Bitcoin - Perspektive oder Risiko? ISBN 978-3-8442-6568-2 http://bitcoin.quisquis.de