Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - kani

Pages: [1] 2 3 4
1
Benefits over transaction fee distribution:
1. Market Fees produce bitAsset of choice -- no need to convert BTS to bitAsset for dividend payment.
2. Does not violate LTM agreement
3. Does not mess with referral expectations

"Your share in the Decentralized Exchange"

Producing revenue from market activity is a natural income source for an exchange.  Sharing it with Bitshares users makes sense to me.
We already produce revenue from market activity via transaction fees without the need for additional market fees.

I've numbered your points for easier referencing:
1. You can pay transaction fees in assets other than BTS, so we can already immediately distribute bitUSD to bitUSD holders instead of converting it from bitUSD -> BTS prior to redistribution. Unless this is an automated process right now? The committee in charge of the fees assigned to the reserve pool could provide additional information here.
2. The agreement was that the membership you purchased was to exist for a lifetime, not that the benefits were permanent for the lifetime of your account. These benefits should be subject to change, given sufficient network consensus.
3. The referral system being allocated 80% of transaction fees was unfair to begin with, there was an expected replacement income stream for asset holders in the form of a bond market which never materialized. It's time for asset holders to have a slice of the pie allocated to them. Referrals don't keep users around, dividends do.

Thank you for the thoughtful reply.

In regards to #1 response:

Yes, that is theoretically true.  But the number of transactions which are paid in bitAssets is very small.  Which means it is not terribly useful source of revenue for a bitAsset dividend.

2
If you put $10 from one of your pockets to another one, how much do you owe at the end?

You can technically borrow whatever on DEX, but until you give away or sell whatever you borrowed, you are not in short position and you debt is zero. By allowing yield harvesting you are not going to increase liquidity.

Bob borrows $10 from Alice.  Then Bob moves it from right pocket to left pocket.  It changes nothing.  Bob still owes Alice $10.

You are confused here.  The problem is that people can short to themselves by just setting up another account.  What does that accomplish that is worth paying for?  This is a widely understood issue that people have dubbed "yield harvesting".

No, I'm not confused here.  The blockchain tracks collateral-to-debt ratio.  This is not a problem if dividend is sent only to those with current debt and not holders.  If dividend was to shorters and holders then yes, I see the problem.

Trust me, you're confused.  The point of paying shorters would be to incentivize them to provide liquidity.  In other words, we want them to borrow bitAssets and short sell them to someone else who wants/needs to buy them.  That is useful activity and is worth rewarding.  But if that person just shorts the bitAssets to himself in another account, then that is NOT useful activity and should not be rewarded.  That is the definition of "yield harvesting".  There is no way we can know if someone is legitimately shorting to other market participants, or just yield harvesting.  That is why we can't reward bitAsset short positions.  Please stop insisting otherwise.

Shelved.

Anyway, I'm shelving the idea in favor of the simpler "market fee revenue to holders" I outlined above.

What you're describing is not simpler.  You keep talking about market fees vs. transaction fees, as if we should be directing one type of fee for one purpose, and another type of fee for another purpose.  Right now we have general income from fees paid, period.  That income is split 80/20 between the referral program and the network.  We're talking about shifting a portion of that income currently going to the referral program in order to 1) replenish the reserve pool 2.5x faster, 2) drive demand for bitAssets by paying interest, 3) reward LTM accounts with a dividend. 

Another thing.  You keep saying you prefer "market fee distribution" because there is "no need to convert BTS to bitAsset for dividend payment."  Again, your idea of separate "market fee distribution" and "transaction fee distribution" is overly complicated.  Also, no one is talking about converting BTS to bitAssets.  You are either confused with some other proposal on another thread. Or you are intentionally confusing matters.

It is my understanding that this proposal would pay bitUSD dividends for bitUSD held, which would require BTS to bitUSD conversion at some point.  If I'm mistaken, then I'm sorry. 

And I really do feel that revenue sharing from market fees is a better option.

3
Benefits over transaction fee distribution:
* Market Fees produce bitAsset of choice -- no need to convert BTS to bitAsset for dividend payment.
* Does not violate LTM agreement
* Does not mess with referral expectations

"Your share in the Decentralized Exchange"

In my opinion suggested model doesn't affect LTM members at all. Lets see:
LTM member pays fee in advance (LTM mebership fee) and gets 80% cash back, on every fee they pay to the network.. It is simple bulk discount.
Proposed change ONLY affects non LTM users which pay same fee as anyone else. They doesn't get any cash back only.
Their fee goes 20% to the network and 80% to referrer and registrant coalition.

Proposed change increases network cut from 20% to 50% decreases referrer coalition cut from 80% to 20% and introduces new beneficiaries like bitAsset interests (20%) and LTM members interests(10%).

Non LTM users can decide whenever they like to pay LTM membership fee and get 80% cash back on their fees. As I understand current model, such action breaks referrers dill and network keeps all of membership fee. Am I right?

Ah, but the suggested model does affect LTM members in a non-obvious way.  If the LTM has self-registered sub-accounts with the expectation that the registering account gets 80% back in vesting from those accounts, then the proposal increases fees for that user.

4
If you put $10 from one of your pockets to another one, how much do you owe at the end?

You can technically borrow whatever on DEX, but until you give away or sell whatever you borrowed, you are not in short position and you debt is zero. By allowing yield harvesting you are not going to increase liquidity.

Bob borrows $10 from Alice.  Then Bob gives the $10 to Frank.  It changes nothing.  Bob still owes Alice $10.

Yeah, right. Now look at this: Bob creates 10 bitUSD and sends it to his own account. Not to Alice, not to Frank, but to himself. How much does Bob owes?

Tip: Bob owes nothing to nobody. Think about it.

Bob now has $10 in another account.  But it was borrowed from the blockchain with BTS as collateral.  He still owes the blockchain $10.  If the collateralized debt could be erased simply by moving bitUSD around, don't you think MPA would have failed a loooooong time ago?

5
If you put $10 from one of your pockets to another one, how much do you owe at the end?

You can technically borrow whatever on DEX, but until you give away or sell whatever you borrowed, you are not in short position and you debt is zero. By allowing yield harvesting you are not going to increase liquidity.

Bob borrows $10 from Alice.  Then Bob gives the $10 to Frank.  It changes nothing.  Frank doesn't owe Alice $10, Bob does.

You are confused here.  The problem is that people can short to themselves by just setting up another account.  What does that accomplish that is worth paying for?  This is a widely understood issue that people have dubbed "yield harvesting".

No, I'm not confused here.  The blockchain tracks collateral-to-debt ratio.  This is not a problem if dividend is sent only to those with current debt and not holders.  If dividend was to shorters and holders then yes, I see the problem.

Anyway, I'm shelving the idea in favor of the simpler "market fee revenue to holders" I outlined above.

6
I'm just going to post my final thoughts on all this.

I think changing the transaction fee distribution is muddled with problems.  I do not favor this approach.

And while I tend to think paying a market-fee based dividend to shorters would have numerous benefits, it is rather complicated for an ordinary user, so I will shelve it here.

In finality, I think the following idea is my favorite:
* Generate revenue from smallish (0.01% to 0.1%) BTS and bitAsset "market fees"
* Distribute BTS market revenue to LTM members
* Distribute bitAsset market revenue to bitAsset holders

Benefits over transaction fee distribution:
* Market Fees produce bitAsset of choice -- no need to convert BTS to bitAsset for dividend payment.
* Does not violate LTM agreement
* Does not mess with referral expectations

"Your share in the Decentralized Exchange"

Producing revenue from market activity is a natural income source for an exchange.  Sharing it with Bitshares users makes sense to me.

7
Quote
And I honestly don't see why it can't be paid only to shorters (collateralized positions, only those who have current debt).  The 'yield-harvesting' "problem" mentioned above would not seem to apply if dividends are paid only to those with current debt.

The 'yield-harvesting' "problem" mentioned above does seem to apply, because there is no way to know who actually have current debt and who shorted to themselves.

Can you explain this?  It is my understanding that the blockchain retains a collateral-to-debt ratio per account and is enforced at a blockchain protocol level -- so this (debt portion) is the metric I'd choose.  How could someone short to themselves?


Someone can borrow bitUSD and send it to his other account. He is not in debt in this case and he shorts nothing on the market, but he is eligible for receiving yield. And you can't tell if someone is a legit shorter or a yield harvester just by looking at account debt, because you don't know who owns which account.

Um, no.  You can't erase your debt just by sending bitUSD to another account.  The debt remains on the original account.  The second account now holds bitUSD with no debt.  This argument makes no sense.

8
Quote
And I honestly don't see why it can't be paid only to shorters (collateralized positions, only those who have current debt).  The 'yield-harvesting' "problem" mentioned above would not seem to apply if dividends are paid only to those with current debt.

The 'yield-harvesting' "problem" mentioned above does seem to apply, because there is no way to know who actually have current debt and who shorted to themselves.

Can you explain this?  It is my understanding that the blockchain retains a collateral-to-debt ratio per account and is enforced at a blockchain protocol level -- so this (debt portion) is the metric I'd choose.  How could someone short to themselves?

Quote
And another reason market fees are better than transaction fees is that is does not change the rules 'retroactively'.  Anybody who doesn't like new market fees can choose not to trade in that market.  Whereas changes which lessen LTM benefits are forced on those who thought they got something different.  Also consider openledger who graciously faucets new user accounts.  They do so with the understanding that they get a cut of the user's transaction fees via registration (and referral, if not otherwise filled) percentages.  I think it's a bad idea to change this now.

This issue is easy to resolve. Let CCEDK and other current LTM owners have their 80%, but stop selling new LTM, because this does not make any sense.

This does not change the increased-fee effect on self-registered sub-accounts.  If the changes could be applied to new users only (at a future date... once implemented), then I'm probably okay with it...

9
This is all dandy, but misleading.  The claim that this proposal does not raise user fees is absolutely FALSE!

A LTM account pays 20% to network, and gets 80% back in vesting.  And anyone who has self-registered sub-accounts, is the same -- 20% to network and 80% back to the registering account.  Don't mess with this -- any changes represent a major breach of trust to anybody who paid the $80-$150 to become a Life Time Member.

Do it with market fees.  Then there is no worry about how to convert to a chosen bitAsset, and is market driven -- more volume creates more revenue.

And I honestly don't see why it can't be paid only to shorters (collateralized positions, only those who have current debt).  The 'yield-harvesting' "problem" mentioned above would not seem to apply if dividends are paid only to those with current debt.  (And consider the VERY POSITIVE price effect this would have on BTS.  Many more will short bitAssets to chase the dividend payment and locking up that BTS for bitAsset of choice.  More BTS locked up means upward pressure on BTS price)

EDIT: And another reason market fees are better than transaction fees is that is does not change the rules 'retroactively'.  Anybody who doesn't like new market fees can choose not to trade in that market.  Whereas changes which lessen LTM benefits are forced on those who thought they got something different.  Also consider openledger who graciously faucets new user accounts.  They do so with the understanding that they get a cut of the user's transaction fees via registration (and referral, if not otherwise filled) percentages.  I think it's a bad idea to change this now.

10
Thanks to those who commented on what I wrote.  I have little time for detailed response but it's clear there is a wider perspective here which I did not consider.

But I will reinforce one idea in-particular: Substantial revenue for paying BTS and bitAsset interest can be made via "Market Fees".  This alleviates the risk of reserve pool managing short positions (or even touching the reserve pool at all).  One advantage is that "Market Fees" are produced in *that* asset (ie: bitUSD market fees come out in bitUSD, and etc).  The transaction fee schedule then remains, in effect, an anti-spam measure and does not take away from the existing referral system.  Plus there is no in-between step of converting BTS from transaction fees into the currency of choice.

Now, I do believe the market fees must be kept small to keep Bitshares a competitive trading platform.  Something like 0.05 to 0.1% is probably okay???  Would need broader community feedback and analysis.


EDIT: I really do think revenue sharing from exchange activity is important...  Does this phrase ring a bell?:
  • "BitShares - Your share in the Decentralized Exchange"
Let's make that statement a reality.

11
I know there are plenty voicing opinions in this thread, but I'd like to take a moment to share mine.

First and foremost, I do not agree with any actions that add strain to the reserve pool (without a very strong argument for long term benefit).  Witness pay has just increased and more active workers are draining the pool faster than increased activity can keep up.  Long term goal should be a sustainable pool which balances pay with fees.

Second.  There should be no change to LTM fee percentages (small changes to non-LTM is fine with me).  Many have chosen LTM knowing the cost-benefit of 80% fees returned.  Don't mess with this.

Third.  We must explore motivations and desired outcomes for any suggested changes.  What problem are we trying to solve?  What impact does this have on the long-term health of the Bitshares ecosystem?

In my opinion, the two things which would help Bitshares the most is:
  • increased supply of bitAssets
  • usable liquidity

The reason being that more active and usable markets will attract new traders to the DEX.  Rewarding those who simply hold assets does not improve the DEX.

I come back, then, to the idea suggested a little while ago: Reward those who have bitAsset debt (collateralized position) with (new) market fees from same bitAsset markets.  Only those with debt collect dividends (from the market-produced trading fees -- aka "market fee") at a pro-rata basis.  Simply holding a bitAsset does not yield dividends.

What does this do?
  • encourages more to short, increasing supply
  • encourages riskier shorts closer to MCR to get most bitAsset for the BTS (which then results in more forced settlement activity)
  • encourages those who do short to actively make markets to increase usable liquidity, which in turn provides volume and increases dividend from market fees

If the idea is adopted, I'd like to see it applied (as a test case) to a less-active bitAsset first before moving on to the big ones.

Thanks for reading!

12
Carbon Wallet "C"

Why?  :P

A link to its graphene base!

 +5%

13
General Discussion / Re: Incentivize SmartCoin collateralization
« on: January 24, 2017, 02:51:59 pm »
Hey guys,

Popping back in to bump the original idea for this thread: Pay dividends to those who collateralize a SmartCoin by raking a small market fee from DEX activity in that SmartCoin.

I like that other ideas have been presented here, but I do not think they are the right ones.
  • Zero fees for orders which sit inside spread is not enticing enough.  The fees are too low for that to matter — if you can make an extra 0.2 BTS by placing orders further in orderbook, is of no benefit.
  • Blockchain selling Smartcoins.  Too big of a change with significant risk to Market-Pegged mechanics.  Borrowing SmartCoins with collateral is not without risk, and is best handled by market actors.  The blockchain cannot assume this risk.

The original idea presented here is a comparatively small change which incentives the right behavior:
  • Creation of additional SmartCoins
  • Higher returns for higher market activity
For a trustless solution this would require a change to blockchain protocol, as dividends should be automatic and made by the blockchain.

The biggest concern I have about this approach is that it will encourage shorters to chase minimum MCR to get the most SmartCoin for their BTS.  Can others comment on the impact of this?  Is this good or bad?

Remember the tagline:
“Your share in the Decentralized Exchange”

This idea brings that statement closer to reality.  For it rewards those who collateralize BTS for SmartCoins with proceeds generated by DEX activity.

14
General Discussion / Re: Incentivize SmartCoin collateralization
« on: January 18, 2017, 05:21:30 pm »
Not to beat a dead horse, but there's one more point I meant to make and forgot.

This idea has a NET ZERO effect on Bitshares inflation -- it does not touch the reserve pool.

15
General Discussion / Re: Incentivize SmartCoin collateralization
« on: January 18, 2017, 05:05:31 pm »
But yeah, you may be right...  I might not have a full grasp on the complete picture.  Which is why I'm not going push the idea too much longer.

While I do think there is some merit here, perhaps someone else has something better.  Bottom line is that with all the talk of taking fees for other purposes, I really think it's best to use any new fees to improve markets.

(EDIT: And what better way to improve a market than by taking fees FROM that market FOR that market)

Pages: [1] 2 3 4