There is actually no point to require minimum 192% (MCR*MSSR) starting collateral anymore. Before change in shorting rules (with old rules which are given in documentation) it did make sense, because margin call could be triggered at this threshold. Now minimum collateral can be set to MCR.

It has a positive point, it prevents someone to set 175% collateral and get margin call 2 seconds after finishing its operation if the price move a little.

When you say that it makes no sense, is there a way to avoid it ?

Because when I'll increase the MSSR to 140%, the collateral asked will be around 250%, which is crazy high. I want to incentive shorters, not the oposite.

Yep, I agree !

I also think we shouldn't stop there with 10% of the fees. We should keep braistorming and find other things to add so the referral program become so interresting that it will start attarcting to whole crypto community. We have an amazing tool and we do nothing with it.

Sent from my SM-G935F using Tapatalk

Quote from: yvv on January 07, 2017, 11:08:20 pm

Quote from: svk on January 07, 2017, 09:44:19 pm

No he's correct, margin calls only happen once the call price goes below the feed price.

Then documentation should be corrected, because it says that margin call is triggered when the call price is between feed price and SQPP.

The documentation is a copy paste of a write up that I did before BM changed the mechanics to make it so that margin calls are only triggered when the call price goes below the feed price, there's a github issue for this in the graphene repo.

Call price is DEBT * MCR / COLLATERAL.

This is extremely confusing indeed ! If the documentation is not right on certain points, I really don't know where to go to find information to make myself able to take the right decisions.

I need to change this parameter (MSSR) as soon as possible. I don't want to have this market in a dangerous position in the next high volatility event.

Can you confirm a couple more think ? (and correct me if I'm wrong)

If I set the MSSR at 140% and keep the MCR at 175% :

1) The minimum collateral to borrow a BTWTY will be at 245% ( 140 * 175 ) ?
2) Is there a way to avoid asking for so much collateral ?
3) The margin call will be triggered at 175% of the price  ?
4) I won't trigger any margin call by increasing the MSSR because they will be triggered at 175% of the price feed ?
5) Once someone has less than 175% collateral, he will be margin call and its order will sit at 140% of the price feed ?
6) The price at which the margin call order sit in the market move with the price feed ?

I like the idea, I like it a lot ! It seems easy to set up.

It would also incentive ppl to get the LTM because you would get your money back through time.
Ppl live the win-for-life kind of stuff.

If we make it that you can have your money back in 2 years in current MC and daily Tx rate, we could see a massive wave of speculators coming in BTS only to create a LTM. That would push up the market, increase our reserve pool in BTS and also in $, increase our available money to spend on dev, have lots of ppl playing with the DEX.

The number of new LTM will at some point reach an equilibrium because the more LTM there are, the less you win and the more time you'll have to wait to have your money back and start earning. Also, the first ppl coming to get their LTM will increase the number of Tx in the DEX and may push forward this equilibrium with a never ending reenforcing circle.

The committee could be responsable of setting the % going into the pool for LTMembers. Adjusting this % could increase or decrease the new creation of LTM.

Damn ... I really find this idea excellent !!

Is there any part of this idea that is not doable or may create problems ?

@Chris4210, another thing to add to your list

Sent from my SM-G935F using Tapatalk

A tricky proposition indeed.

I like the idea for no fees on very small transactions.  But a percentage-based fee on higher value transactions will simply drive users to other blockchains where it is cheaper to conduct business.  I oppose this idea.

Bitshares is special because of its low fees -- you don't see the same bot activity on other platforms due to the relatively high transaction costs on those platforms.  Increased fees just might kill the growing bot ecosystem, which Bitshares gravely needs.

If the referral system can be improved without disrupting the other factors that make Bitshares great, I'm all for it.  But don't kill it with substantially higher fees.

It wouldn't disturb bots, they could operate without fee. That would be even better for them !!
With a system like this one, they would only need to pay a fee when they actually fill an operation. With a low 0.1% that should be a problem at all. Poloniex have lots of bots running with a much higher %.

I've just listened to the BitShares special Hangout.

The fees actual structure have been mentioned but without further thoughts. Someone said something like "didn't we have kind of a solution for that?" and the conversation on the topic died.

Just as a reminder the solution we've encountered a while ago was a very good one !!!

The conclusion was that the best way to handle fee without killing the referral program was to have a hybrid system with rate limited Tx and percentage based fees.

I'll use an example with BTS to explain it :

A transaction from 0.00001 BTS to 200 BTS has not fee at all. (Micropayment possible !)
A transaction bigger than 200 BTS has a 0.1% fee (no one will ever complain to pay 1$ on a 1000$ Tx nor to pay 100$ to move 100,000$)
We could also have a top cap. Tx higher than 100,000$ keep that 100$ fee for example.

To avoid the possible attacks allowed by the Free Tx, we limit the number of transactions someone can do in relation to your funds in BTS (rate limited Tx).
So the more you can abuse the system, the more BTS you need to have in it ... and the less you want to screw it up ;p
It also implies that any company on BTS should have some fund in BTS if they plan on having lots of Tx.
I don't know how they manage that part on the Steem blockchain but their solution may also be interesting.

Having this new system give another chance to the referral program. If you refer a person who will, in a no so distant future, pay its gasoline, cigarettes or groceries with Blockpay, they will produce much more fees that what they would have with the 2015 stable fees structure.
Let's imagine a person who is using crypto for gasoline and from time to time for other stuff in a typical month. Gasoline twice, 10 cigarettes boxes and a couple of groceries at the local small shop accepting crypto. That would be around 250$ paid, so around 0.25$.
60% of that goes to the referrer, so around 1.8$ a year.
I think that it would be enough to motivate people to start again thinking to promote BitShares!

The trading on the platform could also take 0.1% and free order creation and cancellation. Other feature on the platform could integrate this system and produce more and more fees for the referrers.

A Little story to show how reviving the referral program is important. I spent months on building Bitsharesfcx.com website. I started with learning HTML, CSS and some PHP from knowing absolutely nothing .... really nothing, not even what <p> or </br> meant !!! My two friends translated the whole BitShares.org website, the BM blog, flyers, infographics, articles.. we've also translated the book from Max Write.
We have done all of this because we thought BitShares is amazing, we want to help to spread the word and BM just had given to us the perfect tool, the referral program. We made the math and we concluded we could invest a shit load of time in the website. We were going to be the gateway for the millions of people that speak french but not English well enough to learn about BitShares in the next years.
Well, since the referral system has been put in sleep mode, there is a very perceptible lack of motivation in our team and it's very understandable. We are also in sleep mode, nothing has really changed for the last year or so. We may have spent all this time for nothing, that isn't the kind of thought that gives you the motivation to keep working.
If the referral program would be back, providing a way to produce incomes by promoting BitShares, we would probably end up beeing back on the project.
Conclusion, the referral program works, the problem is the current fee structure.

I wanted to share my experience and summarise what I still have in mind about this topic so we can include it in the @Chris4210 BitShares improvement list.

All right, I get it this way :

First you have the MCR (minimum collateral ratio) : 175%
Then you have the MSSR (or SQP) : 110%

The shorter can't short with 175.1% collateral !!! The minimum is 175% * 110% = 192%

The collateral can go below of 192% as the price moves. When it reach 175% it gets margin call and tries to buy at 110% of the price feed.

If I would want to set the MSSR (or SQP) at 140%, the shorter will still be margin call at 175% BUT he would need to short (borrow) BTWTY with a minimum of 245% of collateral (140 * 175).

If my explanation is right (I think so), the remaining question would be, wich percentage for the MSSR would be enough to avoid situation creating black swan risks and not too much so shorter doesn't feel like locking too much collateral ... ?

There is also the option to lower the MCR while increasing the MSSR so the collateral needed wouldn't be that different but that could be even more dangerous.

Just for have some numbers to reflect on :
         
110   175   19250   >>> MSSR (SQP) / MCR / Minimum collateral to borrow    (this row is the current one)
120   175   21000   >>> MSSR +10%
130   175   22750   >>> MSSR +20%
140   175   24500   >>> MSSR +30%
150   175   26250   >>> MSSR +40%
160   175   28000   >>> MSSR +50%
         
110   160   17600   >>> current MSSR / MCR at 160%
120   160   19200   >>> MSSR +10% / MCR at 160%
130   160   20800   >>> MSSR +20% / MCR at 160%
140   160   22400   >>> MSSR +30% / MCR at 160%
150   160   24000   >>> MSSR +40% / MCR at 160%
160   160   25600   >>> MSSR +50% / MCR at 160%
         
110   150   16500   >>> current MSSR / MCR at 150%
120   150   18000   >>> MSSR +10% / MCR at 150%
130   150   19500   >>> MSSR +20% / MCR at 150%
140   150   21000   >>> MSSR +30% / MCR at 150%
150   150   22500   >>> MSSR +40% / MCR at 150%
160   150   24000   >>> MSSR +50% / MCR at 150%

Quote from: yvv on January 05, 2017, 05:35:38 pm

And if condition for margin call is met, it does not take the best sell offer event if there is liquidity. I have a hard time to understand why is it done this way.

but in docs:
 The squeeze protection price acts as a price ceiling, meaning the forced margin order will not execute at a very high price in an illiquid market.

That part makes sense. Without this rule, in an illiquid market, there would be no limit to the price the margin call order would buy back the asset.