Author Topic: What happens to BitAssets if we snapshot BTSX to create a fork?  (Read 3209 times)

0 Members and 1 Guest are viewing this topic.

Offline hasher

  • Full Member
  • ***
  • Posts: 87
    • View Profile
split DAC data for this:
the ability to sell one and buy more of the other depending on their preferences.
?? am i missing something

Offline tonyk

  • Hero Member
  • *****
  • Posts: 3308
    • View Profile
It is truly amazing how this place attracts so many bright minds. I rarely feel stupid in my everyday life... and when I do it is usually by someone that has more knowledge in one particular discipline (or a group closely related disciplines). And yet I  came here and 'met' BM, who is not only too smart to believe he is real, but well diverse in  his knowledge. If this was not enough, I also met people like you *arhag* who are too  expert-y in more than one walk of life. (and the examples neither start nor end with you, which is the truly amazing thing).


So after this long intro... I read your post and I expected something like:
-Math inconsistency;
-Economics inconsistency;
-IT flows;

Most of the above probably would have been above my pay grade to tackle but tried to find what you mean never the less...

But you surprised me.. you address more or less 'the social issues' of a potential fork... amazing really...
can we please indeed close the curtains and keep all this talent to ourselves ...
... if for no apparent reason I get invite to stay, that is.


I yes, I know for a fact that the voting functions have been existing in the BTSX for probably more than 10 versions, not fully implemented but in the works more or less. They are not #1 priority as of now, and I kind of agree that they should not be....

Code: [Select]
blockchain_list_proposals [first] [count]                                                           
Lack of arbitrage is the problem, isn't it. And this 'should' solves it.

Offline arhag

  • Hero Member
  • *****
  • Posts: 1214
    • View Profile
    • My posts on Steem
  • BitShares: arhag
  • GitHub: arhag
True, but I do not see where do you see the problem looming from... Are you saying we have suddenly managed to double everybody's wealth/value... kind of 'Ripple style'? or something else?

It is not necessarily a problem. People just need to keep in mind that if we allow for the possibility of splits they need to be aware that they can be forced out of their BitAsset position (at a fair price at the time of the split) even if they do nothing. It also requires them to be aware enough of the split (and when it is going to happen) to quickly get back into their position after the split (because I assume they are holding BitAssets rather than BTSX for a reason).

Also, because of the complications introduced to the customers of the DAC (the BitAsset holders), it is not an operation that should be taken trivially. Nor is it something we can just do and then let the market decide on if it was a good idea; the act of doing it without considering the shareholders is the problem we are trying to avoid (customer irritation if the DAC split unnecessarily, and risk of black swan if the DAC didn't do the split when shareholders wanted it and ended up drastically losing shareholders, and market cap, as they fled to a fork that was created without the cooperation of the original DAC). This is why I think it is important for the DAC to allow shareholders to vote on whether they want to do such a split; and before actually doing the split in some proposed specified time in the future, to require a large (maybe >75%) share approval of the proposed split operation, regardless of whether the voters prefer the original or the forked-off one, and it would also be smart to gauge shareholder opinion on which fork they prefer and only allow the split operation if the shares aren't too skewed towards one or the other (if they are too skewed towards the original, they shouldn't split; if they are too skewed towards the new fork, they should just implement that fork's rules on the current blockchain and not bother with a split and annoyances to BitAsset holders that comes with that). This is just another handy tool made possible if we design the DAC to allow delegate proposals ratified by shareholder approval like I have been trying to push to the community for a while now.



Offline tonyk

  • Hero Member
  • *****
  • Posts: 3308
    • View Profile
Actually, I invasion only the new (the forked-off) DAC to start with no asset positions... and the original DAC to go through a kind of 'calculated' liquidation of all bit Assets...

Track where the wealth of a BitAsset holder is going. Let's say I have 1 BitUSD on BitShares X. If this fork process happens at a price of 30 BTSX/BitUSD, at the end there will be two DACs: the "old" (although still modified) BitShares X1 and the new forked-off BitShares X2 (there would also be the old BitShares X but that would be effectively dead because the delegates would stop updating that blockchain and the shares would be worthless anyway). I would no longer have 1 BitUSD on any chain, but instead I would have about 30 BTSX1 and 30 BTSX2. Perhaps the new price of BTSX1 would be approximately 0.0166 USD, and same for BTSX2, so I still have the same value as before the fork. Although, the prices would start diverging and changing over time. So, I would probably want to get back into my old position of 1 BitUSD as soon as I can (after all I had 1 BitUSD before for a reason: I wanted price stability). I can choose which share to do this on, or I can do it on both (splitting the value according to the ratio I wish).
True, but I do not see where do you see the problem looming from... Are you saying we have suddenly managed to double everybody's wealth/value... kind of 'Ripple style*'? or something else?

*Reference to how they increased the XRP in circulation a month or so ago without actually diluting them.
« Last Edit: October 04, 2014, 04:58:55 am by tonyk »
Lack of arbitrage is the problem, isn't it. And this 'should' solves it.

Offline arhag

  • Hero Member
  • *****
  • Posts: 1214
    • View Profile
    • My posts on Steem
  • BitShares: arhag
  • GitHub: arhag
Actually, I invasion only the new (the forked-off) DAC to start with no asset positions... and the original DAC to go through a kind of 'calculated' liquidation of all bit Assets...

Track where the wealth of a BitAsset holder is going. Let's say I have 1 BitUSD on BitShares X. If this fork process happens at a price of 30 BTSX/BitUSD, at the end there will be two DACs: the "old" (although still modified) BitShares X1 and the new forked-off BitShares X2 (there would also be the old BitShares X but that would be effectively dead because the delegates would stop updating that blockchain and the shares would be worthless anyway). I would no longer have 1 BitUSD on any chain, but instead I would have about 30 BTSX1 and 30 BTSX2. Perhaps the new price of BTSX1 would be approximately 0.0166 USD, and same for BTSX2, so I still have the same value as before the fork. Although, the prices would start diverging and changing over time. So, I would probably want to get back into my old position of 1 BitUSD as soon as I can (after all I had 1 BitUSD before for a reason: I wanted price stability). I can choose which share to do this on, or I can do it on both (splitting the value according to the ratio I wish).

Offline tonyk

  • Hero Member
  • *****
  • Posts: 3308
    • View Profile

I have spent my fair share of time thinking about this ... but at the end of the day the  most simple solution seems to be the best (as usual), in my view:

-BTSX holders get their share -> 1 per BTSX;
-Short position holders get: Collateral - number of bitAssets* Feed price at the time of the snapshot;
-bitAsset holders get -> number of bitAssets * Feed price number of bitAssets* Feed price ;

The sum of those above should be equal to the total the DACs total current share supply...

Yeah, maybe so. Basically in a single moment you force BitAsset holders to buy BTSX and shorts to cover at the exact same price (the price at the time of the fork), thus removing all BitAssets in existence on the DAC, and then fork as you would with a typical cryptocurrency. Then you just need people to re-establish the market depth and issue BitAssets into existence again on both new DACs.

Notice that it still requires participation from both those who want to fork off and those who are apathetic about the spun-off fork, which is really my larger point. You can't just sit on your BitUSD for a year and expect it to remain there in the same form if the shareholders of the DAC holding your BitUSD decided to do a split. I think the blockchain needs to provide a mechanism to figure out if enough shareholders are interested in a split so that they can justify pulling off the above mechanism (potentially to the annoyance of some BitAsset holders).

Actually, I invasion only the new (the forked-off) DAC to start with no asset positions... and the original DAC to go through a kind of 'calculated' liquidation of all bit Assets...
Lack of arbitrage is the problem, isn't it. And this 'should' solves it.

Offline arhag

  • Hero Member
  • *****
  • Posts: 1214
    • View Profile
    • My posts on Steem
  • BitShares: arhag
  • GitHub: arhag

I have spent my fair share of time thinking about this ... but at the end of the day the  most simple solution seems to be the best (as usual), in my view:

-BTSX holders get their share -> 1 per BTSX;
-Short position holders get: Collateral - number of bitAssets* Feed price at the time of the snapshot;
-bitAsset holders get -> number of bitAssets * Feed price number of bitAssets* Feed price ;

The sum of those above should be equal to the total the DACs total current share supply...

Yeah, maybe so. Basically in a single moment you force BitAsset holders to buy BTSX and shorts to cover at the exact same price (the price at the time of the fork), thus removing all BitAssets in existence on the DAC, and then fork as you would with a typical cryptocurrency. Then you just need people to re-establish the market depth and issue BitAssets into existence again on both new DACs.

Notice that it still requires participation from both those who want to fork off and those who are apathetic about the spun-off fork, which is really my larger point. You can't just sit on your BitUSD for a year and expect it to remain there in the same form if the shareholders of the DAC holding your BitUSD decided to do a split. I think the blockchain needs to provide a mechanism to figure out if enough shareholders are interested in a split so that they can justify pulling off the above mechanism (potentially to the annoyance of some BitAsset holders).

Offline tonyk

  • Hero Member
  • *****
  • Posts: 3308
    • View Profile

I have spent my fair share of time thinking about this ... but at the end of the day the  most simple solution seems to be the best (as usual), in my view:

-BTSX holders get their share -> 1 per BTSX;
-Short position holders get: Collateral - number of bitAssets* Feed price at the time of the snapshot;
-bitAsset holders get -> number of bitAssets * Feed price number of bitAssets* Feed price ;

The sum of those above should be equal to the total the DACs total current share supply...
Lack of arbitrage is the problem, isn't it. And this 'should' solves it.

Offline arhag

  • Hero Member
  • *****
  • Posts: 1214
    • View Profile
    • My posts on Steem
  • BitShares: arhag
  • GitHub: arhag
I had a very complicated analysis I wrote up trying to figure out who benefits and who loses in the case of moving BitAssets over to the chain with the majority shareholder consensus. I thought it worked out to an acceptable solution where BTSX holders remain unaffected, BitAsset holders keep their value (but with some small risk of black swan events in certain situations), and where shorts mostly get their pre-fork value with maybe a minor penalty due to margin call risk in certain situations. But the analysis was very complicated and probably wrong, so I am not going to post it here.

Basically, the conclusion is that this process isn't trivial. It is risky if we do it wrong. And we probably should figure out how to do it right because forks are inevitable (and if it is not for BitShares X, it might be for one of the other DACs that also want to carry BitUSD).

Offline Empirical1.1

  • Hero Member
  • *****
  • Posts: 886
    • View Profile
I've been thinking about this subject too.

I was thinking along the lines that there would probably be a majority supporting an idea & that it would be best for all stakeholders if that majority would probably take the current BitAssets. Regarding the case in point, (BTSX + dilution) I imagine the developer voting block would swing it in favour of dilution in which case that should stay as BTSX.

However if you could make some reasonable estimations that the majority of shareholders excl. that voting block would have voted for the original BTSX. Then you have a case that there would be enough demand for a BTSX fork. However the fork with the lower majority should start from zero BitAssets imo, and it should probably change its composition slightly till the point where the minority would have won the original vote.

Offline arhag

  • Hero Member
  • *****
  • Posts: 1214
    • View Profile
    • My posts on Steem
  • BitShares: arhag
  • GitHub: arhag
I just thought of a particularly tricky situation regarding snapshots of DACs with BitAssets and was hoping to get some reassurance or explanation of how we could handle this hypothetical situation.

Let's say we wanted to at some point create a snapshot of BTSX stake to fork off a BitShares X clone that allowed for dilution. The idea is that at the point of the fork people have equal stake in BTSX and BTSXD (BitShares X with dilution), but then they have the ability to sell one and buy more of the other depending on their preferences. After the fork, the price of BTSX would quickly drop while the price of BTSXD (after being available for trading on exchanges) would quickly rise from zero, until they reach a new initial equilibrium. Let's say the market is initially equally split on which DAC is better, so the price of BTSX suddenly drops to half after the fork. Isn't this a huge black swan risk to the BitAssets on the original chain? It would be even worse if the market consensus favored the new chain more.

Are we supposed to split the BitAssets (both holdings and amounts owed by shorts) between the two DACs? Meaning we actually create two new forks (BTSXD and BTSX'). BTSX' would be the continuation of BTSX but with the adjusted BitAssets. The idea is that after the fork, BTSX becomes completely worthless (and all BitAssets on their chain are also worthless), but people effectively "continue" that chain with the BTSX' chain and its adjusted BitAssets. Also, they have the remainder of their BitAssets on the BTSXD chain. BitAsset holders still keep the same value of BitAssets as before, and they can use cross-chain trading to get the BitAssets over to their preferred chain.

The problem is that we need to decide how to split the BitAssets between the DACs depending on how we expect the value of BTSX to be split into BTSX' and BTSXD. This decision needs to be made before the actual fork, but we won't know the true market valuations until after the fork. Thus, we need some mechanism of estimating BTSX holder consensus of how much they think each future DAC is worth. If the estimation is too inaccurate, we risk a black swan event in one of the forked chains (accuracy of estimation needed is of course dependent on the minimum collateral ratio).