I got 2^256 problems, but a fee ain't one


#1

We’re back for some more basic cryptoeconomics! Recently, our fearless DApp/Browser/Whatever else needs to be done in between Team has proposed that we charge a fee on all stickers in our upcoming Sticker Market.

In fact, the idea that we charge as much as 50% of the total sales has been suggested (in the doc above). Consider that the App Store only charges a 30% fee on transactions that happen in apps distributed through it. I think this is not good. Very not good, in fact.

There must be more discussion around how the fee is justified and calculated (esp when app stores only charge 30%). We need to ask if it will not scare away any potential artists, and why we need to charge a fee at all?

Isn’t the thinking around our whole business model that, if we can create a bunch of compelling SNT use cases such that there are a variety of different “token sinks”, ie ways to spend your SNT, it will mean that each SNT’s value increases, which is how we fund the project long term? Surely this means, as a community-driven, OSS project, that we don’t need to charge fees at all?

If we increase the utility of SNT such that it’s value increases, we can potentially generate enough “revenue” from our token holdings to keep the project sustainable. It shows:

  1. Real commitment to being truly open source and community driven
  2. NO FEES is a HUGE attraction for poor artists who get ripped off on other platforms, which will lead to network effects/virtuous feedback loops.
  3. Real skin in the game from us: not charging a fee commits us to continuously striving to do more and better things with SNT such that its value increases enough to continually cover costs.

The whole point of any smart contract needs to be to demonstrate a different way of doing “business”, that is, of handling the artefacts of how we relate to each other (transactions) in ways which are as anti-rival as the software they’re written in. Otherwise, what was really the point of any of it?


#2

+1 to no fees. The user and artist already pay in other ways, especially during this alpha period, and the barriers to adoption need to be shaved off as much as possible.


#3

In fact, the idea that we charge as much as 50% of the total sales has been suggested (in the doc above). Consider that the App Store only charges a 30% fee on transactions that happen in apps distributed through it. I think this is not good. Very not good, in fact.

FWIW Line & other markets take 50%, so that’s where the figure came from.


#4

Thanks for triggering a discussion @cryptowanderer. We’re progressing fast
and we’re spread more thin than before, which means we all need to speak up fast to make sure enough eyes safeguard our principles.

Agree to ‘justified and calculated’. Which is why I’d be curious to see some data to support our decision making.

Can we model how many ways of spending SNT (utilities), value, and spendings we would need to be self-supporting without fees? Even estimates would be better than sailing without a map;)

Not sure who to look at for this, so any thoughts on how we might accomplish creating such model are welcome.

If the service we offer for the moment costs more than we get out of it in terms of value to sustain the platform (not enough utility and spending projected), IMO we can discuss a fee (of any structure or %). We can only reach a long term business model if we’re sustainable enough to get there.


#5

Good point about adoption barriers for creators @Bruno. Even if the overall commission is less than on other platforms, we do have a sizably smaller user base, currency and convenience barriers. All things to keep in mind.


#6

This is hard for any number of reasons. The value of SNT depends on so many factors external to the sticker market. It is correlated to the price of ETH in non-obvious ways (though this can be modelled to some extent on past data); to one-off and hard-to-predict events like exchange listings and if, for e.g. SNT became a kind of collateral in Multi-Collateral DAI; to other SNT use cases (present and future, no good data as yet from any though); and many more I have not thought of here. That said, @barry did a pretty cool observablehq notebook for ENS usernames, so he could maybe help out?

You do hit upon perhaps a better solution, though. Perhaps we can simply use some % of the swarm costs currently (and this could be as high as 50% of the current costs for maintenance maybe), make an estimation of the sales per month and then calculate a fee that would “cover” those maintenance costs.


#7

Something to consider about modeling utility value is that unlike more traditional assets where the valuation models are generally agreed upon and the differences are over what the inputs to those models are, here there is no established method or general consensus on what the models are lets alone the inputs to those models.

In light of that, I recall something Ray Dalio once said when asked how he would value Bitcoin. He said he would start by figuring out who all the participants in that market are what their interests are. I think such an Ontology for Ethereum would be useful, although I don’t know of anyone compiling a complete market profile. One part of the picture is this: https://diar.co/ethereum-ico-treasury-balances/, but there are also miners, employees, users, investors, etc all who are participants and have their own interests.


#8

Using the ENS model as an example, the biggest lever on future value was user growth. Changing the price had less impact.


#9

Clarifying my earlier reply:

App stores take 30% of any in-app purchase, including stickers, and LINE takes a further 50%. The result is that creators in these markets get a total of 35%.

The reason they did this was to offset cost of maintaining the market, and that’s something we need to think about as well. The white paper pretty straightforwardly reasons that stickers are a great means of generating (typical) revenue.

That said, I’m also curious about the longevity of this project and whether it can be secured exclusively through token value.

  1. Real commitment to being truly open source and community driven
  2. NO FEES is a HUGE attraction for poor artists who get ripped off on other platforms, which will lead to network effects/virtuous feedback loops.
  3. Real skin in the game from us: not charging a fee commits us to continuously striving to do more and better things with SNT such that its value increases enough to continually cover costs.

Unlike other markets, the Status sticker market will be completely open and permissionless to submit (thanks, extensions!). Anyone can register a pack of stickers for sale. That’s a huge benefit to creators who otherwise have to wait for approval and in many cases, can’t even get their work listed.

And even a steeply cut 50% of sales back to artists is better than standard.

But more importantly, giving 100% of sales back to the artists is not a token sink, is it? Those tokens remain in circulation for the artists to spend or hold.

To be less profiteering about it, we could take that percentage of sales not directly to our wallet, but instead burn or recycle the tokens, as Jarrad suggested in #status-core-dapps.

Pros and cons to that?

Thanks for the thought provoking pushback, @cryptowanderer—great to have people engaging with the contract specs at this critical phase.


#10

In the strictest sense (i.e. what V is talking about), no it isn’t. However, it still increases the velocity of SNT overall (a good proxy measure for economic health a la Douglas Rushkoff). Essentially, more things for people to spend SNT on should translate directly into higher demand for SNT, which should increase the price at which it is traded.

I quite like the idea of burning some small % as a donation to the maintenance of Status as an OSS, community-driven project. Think I could likely live with that. Plus, in tandem with @richatmakerdao, we can hold the world’s first real Digital Burns, where we sacrifice MKR and SNT fees collected in a cathartic moment of release in order to protect the system as a whole and keep things in balance :heart: (If we do come up with some sane fee, we could use the same for the Editorial section in the DApp Store, so we get to burn more SNT :fire:)


#11

I would add a requirement of X SNT to actually publish a sticker pack, so that we don’t end up with a Spore-like situation or flat out spam stickers. That X SNT should be burned.


#12

Evenso SNT value can increase it is probably bounded. Assuming we plan to spend those - increasingly more valued - tokens surely the pool status hold will deplete. How do we account for that?


#13

I would add a requirement of X SNT to actually publish a sticker pack, so that we don’t end up with a Spore-like situation or flat out spam stickers. That X SNT should be burned.

Charging a small fee on receipts (that is burned) has the same net effect, and I like that we lean to more permissionless and then institute curation mechanisms on top of them (yes, bonded curves for everyone!) to prevent spam.

Even so SNT value can increase it is probably bounded. Assuming we plan to spend those - increasingly more valued - tokens surely the pool status hold will deplete. How do we account for that?

Surprise, surprise: this can also be modelled with a cool curve! Ideally, we want the rate at which our holdings decrease to match roughly the “revenue” we earn from the SNT price increasing. Not sure of the exact right equation off the top of my head, but it can certainly be done and then we can figure out the rate at which SNT value needs to increase over time given more and more SNT use cases :stuck_out_tongue_winking_eye:


#14

Or it means that we should charge the minimum possible fee to cover the operating costs of such an activity. I think it’s possible that whilst the overall utility of SNT rises, the wider crypto market falls in the short term, bringing the price of SNT down with it. Charging a small fee (to cover the operating costs of the activity) ensures continuance without relying on the SNT price fluctuations to cover costs.


#15

We should clarify what we are trying to accomplish with the fee because that will drive the tradeoff decisions made. Charging a fee to provide short term operating cash flow (which likely means sacrificing user growth) will look different from creating a structure to maximize the long term value.

IE: Charging 30% with a transaction growth rate of 100%/yr will produce a smaller net present value than charging 0% in year 1-10, and 30% from year 10 onward with a transaction growth rate of 300%/yr.


#16

Great topic and discussion. I’m generally in favor of no fees, largely for the reasons outlined in OP. I see the revenue stream argument as well.

Let’s talk about this at Core Dev call on Monday: https://github.com/status-im/pm/issues/11

PS Speaking of sticker revenue and Line, I passed by a Line Friend event in Taipei, here’s what this looks like. What would the equivalent look like in a few years for Status? To what extent could it be community driven and profitable to token holders? DS


#17

Adding utility to SNT is a crucial target. The sticker market is perfect for that. I’d focus on making transacting in SNT as fun as possible and derive the design choices based on this ‘fun principle’. Ideally, the users of Status will love the stickers as much as transacting with SNT. How other platforms structure sticker markets, should not be the benchmark. The benchmark should be the joy which the users will experience while transacting on the Status platform. The fee should be structured to incentivise both the creators of the stickers and their users. While of course, it is essential to make the right design choices upfront, the fee structure may change over time, especially when a change brings satisfaction to all parties involved. With the vast Status Treasury, burning or recycling and the percentages taken or not taken do not matter much at this stage. From an economic standpoint, the number one concern should be to maximise the number of SNT transactions for the utility of SNT to unfold. In the long run, the volume of the SNT transactions is the key lever of the sustainability of Status.


#18

+1 to @j12b and @rachel’s comments here, where a small fee is charged for the maintanance of the project. We would likely want to do something similar in the Teller Network as well.

As discussed above, the fee’s can then be pooled, where the projects that get funded from that pool can be decided upon through SNT voting through the DAO. This would be similar to how 10% of block rewards in DASH are used for its future development.


#19

Just playing devil’s advocate here, but probably there are others constraints too.

e.g. we could transfer 10 SNT to each user sending a sticker. I certainly would find that fun! And transaction volume would probably significantly increase too.


#20

So, I thought I’d chime in here and elaborate on my position made in #312-janitors.

I am not a proponent of the idea for fee the way it’s being presented, why? Well it boils down to incentives, we need to pay people to use Status, in the sense that people need to be able to profit from participating in the network, that’s why network incentivization and user aquisition engine are so important.

The reason we’re not working on the User Aquisition Engine is because it’s network effects on steroids and Ethereum doesn’t scale yet, although we’ll implement it after the lesser SNT use-cases.

The mental model of rent extraction, is largely rooted in the gatekeeper business model which is predominant in Web 2.0. “I provide service, you have to go through me to get said service, I extract rent from you by taking a fee”. That’s what Uber does, they middle man a network, and they also prevent the network from voluntarily improving the network.

LINE is doing exactly the same thing. Users can’t improve the sticker market by contributing code, they have to have LINE’s approval (thereby censoring and distorting the marketplace) and LINE takes a massive cut. What a shit deal for users. Doing this we’re making more work for ourselves, by enshrining ourselves (Status GmbH) as the provider’s (and governers) of the solution, it’s incredibly inefficient.

Infact it’s common for web2.0 business models to waive the fee until they get large enough network effects and then shanghai their userbase by dumping a fee on them later, it’s very feudal.

I heavily considered typical business models when Ethereum first started and came to conclusion that they don’t work. why?

Aside from them being inefficient, Ethereum is weird, in decentralised application it’s very difficult to concentrate power as the middle-man, because we disintermediate them. That’s why Status is open source, to minimize the ability for us to centralize and work against the users in the event we attain real network effects.

Better business models revolve around making a market and then acting as a node within that market, or some kind of transactional layer that requires the token to function (which is why messaging incentivization is important for SNT and our real asset)

This allows people to buy into the network, and improve it with their self-interest to maximise their return. Token’s represent networks of belief, they believe in the idea the token represents, even if that’s as simple as expecting the value to appreciate.

You can view Ethereum as a medium of pure free trade, more concretely anyone can take your smart contract code, your application code and ‘neuter’ the fee and re-deploy it, they can do this for less than $0.50 in 30 seconds or less. It’s basically piracy, they don’t have the overhead of creation of it.

This move undercuts your service and is more competitve than your service, and will win in the long run as your fee cannot compete with zero-margins.

So how do you prevent them from undercutting your service? Well, you look at the incentives.

The question isn’t “how do we recoup our development costs?” (which is insignificant in the scheme of things, and I understand that most people in Status haven’t experienced the appreciation of a token’s value)

Also don’t forget that the network participants paid for this, it’s not our money, they entrusted us with their money. We are working to reinforce their belief in us.

The question is "how do we make the bad behaviours unprofitable? or how do we make participation profitable?"

Remember, no one is going to use Status unless we (the network) pays them, unless they can earn from onboarding people into the network.

Also by virtue of you holding SNT, it’s very important to understand that you are part of the network, taking a fee isn’t going to make you richer, it’s probably going to make Carl and I richer and by our good graces we decide to put it back into the network (btw it’s not a charity, we’re SNT holders too and we want the value to go up for ourselves out of our own self-interest).

If you view every SNT holder as part of the organisation (think of them like shareholder, remember how I keep saying there is no us vs them, no internal vs external) then you want everyone in the organisation to profit, to prosper from what the network creates.

If we hold that idea in mind, and we want people to, then the ‘fee’ should be burning of the SNT. This is deflationary and therefore rebalances the value into the remaining SNT. Therefore everyone profits and the need to work against the network, to pirate the code is mitigated.

Because piracy of the code recognises value of the code, of the product - then by simple self-interest it’s better to participate in the network, buy more SNT and get more people buying stickers, which means getting more people to use Status to buy more stickers.

You want to capture that individuals interest, their enthusiasm and allow them to profit and contribute, thereby the entire network benefits and you’ll create stronger network effects.

All of our SNT use-cases should exist to create stronger network effects, by allowing everyone to earn more value by participating in the network. You want people working for the interest of the network and what it stands for, you don’t want to make them work against the network.

If you want to recoup development costs, then maybe it’s acceptable as an voluntary donation to the DAO in the UI.

Rent extraction that the network doesn’t profit from is short-sighted, alienates and works against our network. The very thing we’re trying to grow. This is bad.

Now if we do have a fee, there’s a big question of pricing it. I’m open to suggestions. Some ideas to explore are quorum based setting or some kind of difficulty-like adjustment.