Why not?

Quote from: abit on February 01, 2016, 03:04:19 pm

IMHO there are flaws. Market making is NOT a 100% safe business, it will definitely loss in a trend, so the system itself should never dilute to operate such service.

Meh. You are arguing that once there is a sufficient amount of liquidity on the books, that there will be sufficient incentive for people to manipulate the market to their advantage. You are effectively saying that the DEX is broken, as once a sufficient to amount of liquidity exists that it will be manipulated and profited off of from bad actors. What your are saying is a possibility whether my proposal is enacted or not. It is possible as long as there is enough incentive on the orderbooks, which can occur naturally. How do you know this isn't already taking place as is?

I get what you are saying, but while bringing up a good reason the proposal should not be implemented, you are also stating the DEX is broken and unfixable. As soon as there is enough liquidity (incentive) then bad actors will manipulate the market for easy profits. This seems more like a flaw with Smart coins than with the proposal as it could easily happen without it.

I've posted the reason above but you didn't quote it.

One of the challenges is: if we have huge liquidity, we shouldn't simply move the feed price when price on external exchange moves, especially when liquidity of external exchanges is less, otherwise it's a good target to manipulate/exploit. But, since the IOU USD backed by exchanges are real USD, on the opposite bitUSD is NOT fully backed especially when huge amount of them are created via dilution, the feed price should move/follow the price on external exchanges, otherwise bitUSD would be considered "not pegged well". Imo BitUSD can be considered as fully backed if there hasn't been unlimited BTS diluted to create them. I'm not saying that smart coins are already broken, but I suspect that it will break if doing this way.

Exactly this:

The challenge is preventing the feed from being abused among a few other things.

Quote from: abit on January 31, 2016, 10:52:14 pm

If the system dilutes unlimited BTS for market making, what will happen when a whale or a group of whales manipulates the price of BTS?
* buy the BTAs which are borrowed via dilution, dump BTS to external exchange, settle
* borrow BTA, and sell to the system, pump BTS on external exchanges, buy BTA back from system

The autonomous liquidity operations would not ever buy SmartCoins. It only shorts them, and then places them on the sell side of the market above the price feed. I don't understand what you mean by "sell to the system" since the "system" is never buying any coins.

If this works how I think it works, couldn't someone do this type of an exploit as Bitshares exists today?

I wrote above post from a manipulator's perspective.
Since your "autonomous liquidity operations" (in my word: your bot) place orders into the market, I named it "the system", whatever price you place it, a manipulator can take advantage of your open orders. Once your smart coins are bought by the manipulator, you have open short positions, which can be settled by the manipulator. Your bot would passively accept all the commands sent by the manipulator, thus cause losses.

//Update

If this works how I think it works, couldn't someone do this type of an exploit as Bitshares exists today?

Since no one provides unlimited liquidity without moving the price. The manipulator has no opponent to worth exploiting.

Quote from: cube on February 01, 2016, 04:22:42 am

Quote from: jakub on January 31, 2016, 05:57:38 pm

Moreover, I think you can return 16k BTS to the user immediately after s/he upgrades, as you'll be sure to be compensated in 90 days by the vesting process.
This way you loan 16k to the customer but this loan is absolutely NOT safe.
It all makes sense if you have other ways to make profit on this customer.

Corrected.

The vesting period of 90days means a fluctuation of bts price.  From a business perspective, you need to keep accounting of the price fluctuation at the point of sales and the potential of losses if bts price dropped by end of the vesting period.  Some may not be interested to offer a non-profit faucet because of that.

IMO it IS perfectly safe, assuming you are BTS long anyway.
You're right, this will not suit everyone but this assumption holds for most BitShares-related businesses, including bitcrab's transwiser.
You give away 16k BTS today and you're sure to get the full amount back in 90 days.
You need to freeze some of your BTS capital this way, but in return you get a happy customer having access to low transfer fees.

I'm still curious why @bitcrab is not willing to consider this as an alternative option to "1 BTS for transfer".

Probably it need to freeze too many BTS's. Probably you don't know how many potential (ab)users are there in China. 16K BTS per user means 1.6M per 100 users, 16M per 1000 users, imo it's too easy to be exploited.

//Update:
Even the 16K BTS is paid by the referrer, the 4K upgrading fee seems to be still high for some users who want 1 BTS transfer fee.

yes, and then everyone refers their own accounts and marketers get nothing. but go for it, if you guys add 0 fees you might as well give this project the final blow

If one register a new account which referred by herself, she won't actually save money until she upgrade the first account to LTM (to be able to withdraw from cash-back object), which will give the original referrer/marketer 80% of upgrading fee, so it's not "nothing" but "only thing". However this upgrading would come after a long period, or even never, so yes in this case the original referrer will get nothing.

cd libraries/fc
git checkout master
git pull
cd ../..
make witness_node


If the system dilutes unlimited BTS for market making, what will happen when a whale or a group of whales manipulates the price of BTS?
* buy the BTAs which are borrowed via dilution, dump BTS to external exchange, settle
* borrow BTA, and sell to the system, pump BTS on external exchanges, buy BTA back from system

Risks are too high IMO.

If price is decided by external exchanges, to be safe, fund for market making inside Dex have to be much less than volume/depth of external exchanges. However, if the fund are too few, it makes little sense, since no enough liquidity will be provided to the Dex..

So the Dex should find price by itself. For example run a LMSR bot with huge fund. But how to peg?

So a possible solution would be, do market making on external exchanges and Dex. When volumes on external exchanges are high, market making on the dex will be safe. If no depth on external exchanges, dex will die.