Quote from: Agent86 on October 01, 2014, 02:00:37 am

Quote from: tonyk on October 01, 2014, 01:39:05 am

And here is the other side of the equation:
how about we restrict the longs to also buy just at the beg?

It will definitely fortify the peg. Everybody is selling and buying just and only at the peg, and the shorts compete just by the amount of collateral provided.

Perfect peg. No? What more one can ask for...

You can't restrict free trade/exchange between BTSX and BitUSD... the trading could just move off chain.

I agree that you should not, ... but your system just did that for one side of the trade... so what prevents us to do the same for the other counterparty? And it will inforce the peg like... ice

Quote from: Agent86 on October 01, 2014, 01:25:03 am

How "short lived" the highly collateralized positions are is irrelevant if you always have compete to put up high collateral whenever you re-short.  It may not have made a huge difference overnight because there are still a lot of low collateral positions out there, but if those are forced closed or we started the system with collateral prioritization from the beginning then you would see the difference.

Ok, let assume for now they are highly collateralized (not the main objective of your proposal, (but I agree that finding a solution to increase collateral  it is good) ,btw). How this makes the peg tighter?

Quote from: Agent86 on October 01, 2014, 12:01:17 am

Quote from: mf-tzo on September 30, 2014, 11:46:20 pm

Quote

I guess I just don't see people "dropping their pants" to unload bitUSD if they know that holding for less than a year will give them face value.  I'm not saying your view doesn't have merit but I think if it's that critical the over-creation of bitUSD probably already happened and now most will be grandfathered anyway - it would be kind of nice to reset the market in the short run.

I didn't make any sense from that comment but then again I am tired so I better go to sleep now..What do you mean "reset the market in the short run"?

I'm just describing forced covering as "resetting the market" (all positions are opened new from scratch and excess bitassets are removed from circulation).

My point was that there is a limit to how much people will undervalue their bitassets when they know they just have to wait a bit to get the full value.  Making shorts re-short every month seems like a lot of work to me.

So if (short_order_opened(time) >= 30 days, close_order());

is a lot of work, but new rules for shorts is not?

Quote from: Agent86 on October 01, 2014, 12:01:17 am

I'm just describing forced covering as "resetting the market" (all positions are opened new from scratch and excess bitassets are removed from circulation).

My point was that there is a limit to how much people will undervalue their bitassets when they know they just have to wait a bit to get the full value.  Making shorts re-short every month seems like a lot of work to me.

So if (short_order_opened(time) >= 30 days, close_order());

is a lot of work, but new rules for shorts is not?

Quote from: Agent86 on September 30, 2014, 11:05:16 pm

Forced 30 day covering feels pretty aggressive to me.  I feel like having it just be one year (consistent with the inactivity fee) would be fine.  I'm not saying there's a right answer but it seems unnecessarily short to me.  I think the main benefit of a new short window would only be achieved if current shorts were not grandfathered in - we could sort of hit the reset button on the market that may have created a bit too much bitUSD supply in the early days without short restriction.  I don't think there's a moral imperative to give current shorts an exception but it is what it is…  I'm just not sure having a short 30 day with grandfathering cover has a real advantage over a one year force cover.

It is very necessary because without it the varying demand for BitUSD will cause the peg to suffer.... suppose there is a massive increase in BitUSD demand (BTSX Bubble) so the supply doubles pairing to shorts who don't want to cover for a year.  Then all of a sudden people want out of BitUSD (Bubble is Over)... there are not enough buyers on the market and the long-term shorts don't want to cover (they have potentially lost money as BTSX was falling so don't want to realize their loss.   The only solution the market has at this point is for BitUSD holders to sell at a loss large enough to motivate long-term holders.   The fact that they are selling at a loss will discourage future BitUSD holders from buying.   

In the long-run, you can increase the forced covering period or remove it all together... but in a young illiquid market it is certainly necessary to ensure that there is plenty of liquidity and that BitUSD will always revert to the peg within a month of any increase in supply.

This is partly in response to the following conversation (which starts here https://bitsharestalk.org/index.php?topic=9464.msg123288#msg123288) I had with bitmeat.

So generally, I suck because I answered 4 of his question (albeit jokingly) and Agent86 answered only 1 of them but politely.

It does not matter that A86 is the mastermind behind this new short market rules. It does not matter that I have managed to collect like 85% of the answer HOW those market rules work. I have done that through scraping through post partly relate/explaining said rules and through tests in the BTSX client. Test that as we speak have left  non-cancelable orders, orders stretching not only in one market, but appearing in several (bitUSD/BTSX and bitBTC/bitUSD in that case).

But the person who designed those market changes, the person who knows 100% of not only HOW but also WHY those market rules work, as well as what are the benefits from them cares to politely answer just one question (from time to time) when several questions are asked. Requests for explanation of those rules have been plenty – in posts and even entire threads have been dedicated to  that! But the only person that knows the real answers (other than the developers themselves, but we all know they have more work than humanly possible to handle) does not care to post them.
But for that I suck and he does not!

I also suck because before giving the right answers to bitmeat, I have to dig out what the problem of mf-tzo was regarding his short orders. https://bitsharestalk.org/index.php?topic=9361.msg123205#msg123205 The usually helpful xeroc was on the wrong path regarding the issue… the reason for that – nobody has written straight all the new rules regarding shorting. So, your rule digging, just 85% sure about those rules, in house sucker, had to help the guy.
But for that I suck and the above it all politely answering from time to time, masterminds of those rules does not!


Cheers!

Speaking of which. You should create an ELI5 for the new market rules and have it stickied. Correct me if I'm wrong but it seems to me the new market engine is meant only for bot trading? The lack of understanding would certainly explain why BitUSD volume has tanked.

Quote from: Agent86 on September 29, 2014, 06:28:54 pm

Quote from: emski on September 29, 2014, 05:53:05 pm

Network Maintainers' (aka Delegate's) job is measurable, strictly defined, and can be audited. It can even be decentralised (Imagine each delegate split into X different subdelegates that need all approve the same block). Their job is network support and security.

You propose to merge these accountable/auditable strictly defined jobs with investment politics, dilution and subjective  "increase the value of DAC".
Here I disagree with you.

I am also of the opinion that delegates should be separate from workers.  I also feel strongly that dilution should require majority (50%+) approval of "active" shares. 

Gamey, I think your idea that merging this function into the role of delegate simplifies things is wrong and it will have the opposite effect.  It creates more complication and need for "brainpower" than keeping the roles separate.

discussion from a while back:
https://bitsharestalk.org/index.php?topic=5467.msg73251#msg73251

You operate under the assumption that people have spare time to learn the 101 delegates to begin with.  This has shown to be wrong.

We haven't really cracked keeping track of 101 delegates. 

A person spending their inflation in a wasted manner will be likely brought to light quite quickly. 

Having 2 lists of people keep track of etc just adds obvious complexity, but it might happen not be as novel and so it makes more sense to people.  So I agree if the board of directors is sufficiently small it could make it simpler for a user to keep track of, yet it is still another page on the GUI, another separate race, etc etc.  I understand it fits into pre-existing defined roles, but that doesn't mean it actually makes it all simpler.  Perhaps it does, but it definitely does not from the coding end and makes the codebase even more complex.

Also by making a smaller board of directors, you give any 1 person more power than spreading the power around 101 delegates. I'm not sure thats actually a good thing.

You operate under the assumption that people have spare time to learn the 101 delegates to begin with.  This has shown to be wrong.

We haven't really cracked keeping track of 101 delegates.