agriimony
commented
3 years ago agrimony Apr 24
Yes I understand the calculations but I would like to know why specifically you chose 0.17 and not 0.25 or some other number? Seems rather arbitrary to me and would like to know more.
So for example, why do you think 0703% total fees puts us at an optimum spot? If volume and the network effect (because of reduced fees) is what you are after, instead of subsidising the maker, we could also temporarily remove or reduce the AirSwap fees which would bring us down to 0.743%. So you would have to justify why subsidising using AST would be more beneficial in this case.
VladislavW Apr 26
So for example, why do you think 0703% total fees puts us at an optimum spot? If volume and the network effect (because of reduced fees) is what you are after, instead of subsidising the maker, we could also temporarily remove or reduce the AirSwap fees which would bring us down to 0.743%. So you would have to justify why subsidising using AST would be more beneficial in this case.I guess we can’t 100% predict the volumes but which we will get relative to the percentage provided as a subsidy and we have to experiment. I would like to update my calculations a bit according the math i did recently:
MM charges 0.875% anyway from each trade (for AirSwap it is 0.743%)
So each DEX should adopt and consider that each trade done on their platform will additionally charge 0.875% so the platform should adopt its fee to stay competitive
AirSwap has the option to set 0.13% fee so that would be 0.875% total MM fee (other platforms should set 0% to reach the same fee level)
AirSwap prepaid 0.743% fee to the MM (on some volume) and was charging 0.3% from each trade, which was 0.575% more competitive in relation to other DEXes My first question here: Why didn’t AirSwap set the fee to 0.4% or even 0.5% if the trades would still stay competitive? I suppose that is because makers took that difference of <0.575%
Now MM charges 0.743% from all AirSwap trades and AirSwap fee was reduced to 0.13% so there is 0.875% (same as 0% fee for other DEXes) and there is 0% window for the makers now. Thats why volumes have dropped, because makers don’t have any “fee window” nowIf so makes sense to:
Ask MM to add 0% slippage option
Reduce AirSwap fee to 0.1% or less
Subsidize makers some % with AST to help them make more competitive prices and win more tradesGenerally i would like propose the following:
Reduce AirSwap fee to 0.1% (it will help us our makers to have additional fee window with 0.843% fee total compared to 0.875%+ our closest competitors)
Subsidize our makers with AST from the AST treasury with the easy configurable value. I propose to that that amount to 0.143 (it is 0.7% of total fee compared to 0.875%+ our closest competitors which is 0.175% window for our makers to make more competitive and profitable trades) for start and watch how the things are going for 1 week.
As the next step the AirSwap protocol fee and rebates amount should be easy configurable to help us strike the best possible balance between rebates, volume, and feesgreypixel Apr 26
Ask MM to add 0% slippage option
I’ve mentioned this in the design suggestions channel on discord, but I will re-iterate here because I believe it to be very important: adding a 0% slippage option on its own does not necessarily mean RFQ will be selected more often.
Further explanation
For an AMM, slippage tolerance does not affect the "quote" you receive, only the acceptable difference between the quote and the number of tokens you actually receive if the market moves between the time that you're served the quote, and when your transaction is mined. Therefore if MetaMask did not also update its selection algorithm, it would most likely result in more transactions failing, and an exodus of users from MetaMask.
We should therefore also suggest that selecting 0% slippage tolerance either completely removes AMMs, or both deprioritises AMMs during quote selection, and displays a very clear warning in the UI if the user goes to manually choose a quote.
My suspicion is that many MetaMask users are unaware of slippage tolerance, and of those who are aware, few actually understand exactly what it means. I have doubts over how many people would click it.
Justification
I have this opinion because I believe those who think and care about slippage tolerance are generally those looking to ensure they are achieving an efficient/good value trade. I think if they were looking with this level of detail they would most likely not use MetaMask due to the 0.875% transaction fees that could be easily avoided by going to the DEXes directly.
A good point of reference would be to ask MetaMask what % of their users have adjusted default slippage tolerance already.
An alternative option would be to suggest always displaying an RFQ quote, which in the case of a more “competitive” AMM quote would mean showing two. We could suggest that the RFQ quote was shown as a ‘guaranteed price’ and perhaps a ‘low chance of failure’, compared with the AMM which shows the price range of tokens that may be received with the text ‘higher change of failure’ (or similar).
EDIT: the other alternative would be to flat out ask MetaMask to default to preferring RFQ. I assume this is not likely but I have no idea.
Reduce AirSwap fee to 0.1% or less
I prefer this option to directly paying for volume from the treasury, however I still feel it may be too early to jump to this conclusion. Is there a reason our makers can’t quote competitively despite the level playing field of fees across MetaMask providers? In fact we have a very slight advantage currently.
Subsidize makers some % with AST to help them make more competitive prices and win more trades
Same point as above (is this really necessary for competitive quotes?)
100% agree with those saying that subsidies should be performance based, and should be stepped based on meeting criteria for trading volume achieved and uptime targets.
I believe using KPIs like this, particularly for uptime makes this a more palatable treasury spend, because it can be at least partially seen as covering the operational cost of running a high availability service.
Also not opposed to AST treasury incentive for uptime, and points incentive for volumes.greypixel Apr 29
Comments following the new wording:
The idea is to empower coordinators to selectively run rebate programs and onboarding packages for makers and integrations case by case in the interest of meeting volume targets. That proposition have to be named PQI (Price Quote Incentives).TBH what is being suggested is a bit vague, I’d prefer to see something more specific. I’ve put an example down below, but am completely open to other suggestions, I’d just like a bit of clarity before I vote.
The intention is not to be prescriptive, but to provide some guidelines for co-ordinators who already have a lot on their plate. All numbers are plucked out of the air, so don’t get tied up on actual $ values.
Anything in [] would be adjusted by co-ordinators based on discussions with makers.
Onboarding stimuli
This would be an acknowledgement of the effort taken to get a server up and running, and serving orders. This could be split into two components:
[Fixed amount] provided at least [X] swaps (number, not value) are served in [onboarding period].
[Variable amount] based on $ volume over the most recent [measurement period] (could be removed and replaced with performance incentive below).
Example with numbers
Performance incentives
Intended to reward reliability and/or performance. Any of the following:
Fixed payment for uptime, e.g. 99.99% → [10k AST] per month
Payment for average daily volume volume given a [threshold] is met, e.g:
0 - [$500k] → 0
[$500k+] → 0.035%Notes:
The onus would be on the Makers to keep a log of all successful transactions in an agreed format that would allow us to verify $ values.
Similarly uptime would need to be provided by makers based from an agreed third party service
All payments would be in sAST (stakeFor), with conversion from $ value done at the time of payment.
The success of this system depends on attracting and keeping makers, so there is a balance to be struck between making the incentives attractive, but not so attractive that makers would come for the onboarding incentives then leave once they finish.
I am assuming that paying retrospectively avoids the MetaMask fee increase effect when the incentives end.astholder May 1
Just a few thoughts and comments from me :slight_smile:
Rationale
The idea is to empower coordinators to selectively run rebate programs and onboarding packages for makers and integrations case by case in the interest of meeting volume targets. That proposition have to be named Rebate.
Each new Rebate should be written in the same way as AIP, it should have at list Summary, Specification and Rationale part for the community to be acquainted with each proposed Rebate.
If community has at list one member who disagrees with proposed Rebate and is able to argue his disagreement, this Rebate must be nominated for a weighted voting without points accrual (like is was done to configurations before).IMO - I don’t think we need to write/vote an AIP everytime we onboard makers/integrations or apply performance based incentives. It’s not very efficient. Furthermore, there could be instances when the maker prefers more anonymity. I think since the AirSwap coordinators are doing the stakeholder engagement it makes sense we discuss these incentives internally and then informed the community. Any objections could be raised in the discord and we can discuss. Perhaps as a compromise we can define a maximum value of the incentives packages the Coordinators are allowed to utilise. Anything proposed that exceeds the maximum could be voted on like an AIP or configuration vote.
In short, I think the AIP should call to empower coordinators to negotiate rebates on a case by case basis while sharing as much info as possible with the community to stay accountable.
As for the incentive - I think sAST is a pretty good mechanism - but each case is unique and dependent on the interests of the individual maker/integration etc. I’m not sure a strict rule on what the incentives are (currency or amount) is the best way to go - it really limits our negotiating options.
greypixel May 6
In terms of AIP wording, I think the vast majority can be removed - my points & comments were never really intended to go into the AIP wording - they are not concise enough, just there to encourage discussion.
This could probably be distilled down to a few short bullet points.
My main concern is makers creating good volume during incentive periods, then not being able to keep it up, or worse, disappearing, after the end of the incentive period. This was the reason for retrospective performance-based incentives, and rewarding in sAST.
I do understand that some makers may not want sAST, but I think it should be the default starting point.
candiceb May 6
It does make sense to me to have high volume because it will effectively make people aware of Airswap and perhaps attract investment. However, we still need to be able to drive more organic volume without subsidy. For me, i would think it is appropriate now to sub because we are using MM and the fee is high for makers.
I wouldnt subsidize like this after we launch our own web-app. Can we have a time limit on this Aip?
temporary incentive package for makers on MM*I against using treasury for further stimulus or so after we have our web-app.
It should be perhaps percentage rebate or so later. Like if you reach x volume, you get % of fee back kinda thing from the revenue, not treasury. Some sort of tier-system.
We should not sub new makers with Asts without seeing their performance when we have our own web-app. We should have tier rebate system. They get % fee back .the more volume, the more they get back. For example, You earn me $100000, i give you back $10000. You earn me $10000, you get $500.
candiceb May 25
So this aip doesn’t show me a specific plan but example of x and x.
Are we voting again later for a specific rebate racket/tiers?
I think this aip should be bit more specific with details and plans and how implement, specific tiers or formula for rebate.
Summary
Goal is to increase the trading volumes and to attract new makers and integrations to the AirSwap protocol. The proposal is to empower coordinators to selectively run rebate programs and onboarding packages for makers and integrations case by case in the interest of meeting volume targets.
Specification
Example 1
The example of how AirSwap is going with MM and why do we need rebates:
Example 2
Rebate proposition could consist:
Notes:
Credits
Thanks to greypixel, astholder, agrimony, Justin, Don, zzew12 and Schalla for their contributions to this AIP
Copyright — All proposals are public domain via CC0.