Cashflow (Allow every SAFE token to pay for every SAFE Network service)

The idea of Cashflow is pretty simple (Allow every SAFE token to pay for every SAFE Network service). For example with our ad platform, besides our Plimst token we would also accept other SAFE tokens.

What this does is eliminate the use of an exchange, turn every service/user in an hedge fund and indicate services which tokens are used more. Exchanges take time/money, hedge fund waste time/money (money that could be used if it was in circulation). The more widespread a token is the more usable/valuable it becomes and honestly it doesn’t matter if you paid $1 with SAFEcoin, Clike, SAFE FS or Seeds. Fragmenting value will also take away the users attention from looking at the prices of tokens to what really matter, enjoying services on the SAFE Network and explore. Unfortunately every fun idea also comes with it’s implementation issues, because it would also help if you could send $1 consisting of $0,25 SAFEcoin, $0,25 Clike, $0,25 SAFE FS and $0,25 Seeds, but how do you do it? If you know how to solve this through code on the SAFE Network, I have a bounty of 1000 euro’s :stuck_out_tongue:

This conversation started here, but we’ve hijacked that topic long enough.

I’ve followed the Bancor exchange for these last couple of months and it reflected the current market valuation as close as possible. But to be honest the market will follow whatever value the Bancor smartcontract sets or adjust it for example:
If you can buy BNT for $2.10 on other exchanges and it costs $1.90 on the Bancor smartcontract. You buy it on the Bancor smartcontract, which will make the valuation go up a small % and sell it on the other exchanges and the other way around, so the price always adjust. So price discovery is not a problem, but what it means is that we need an smartcontract exchange.

The best place would be if they can fetch it from a smartcontract similar to Bancor on the SAFE Network, this way we don’t need an oracle to fetch data from the clearnet. The market will follow whatever price is set on the SAFE Network or set it accordingly.

Bancor works like this for the people who don’t know. You send it tokens and it sends you back BNT, you send it BNT and it sends you back the token in it’s reserve. Maybe our smartcontract exchange needs to work like this, you send it tokens and it sends a stablecoin to the service. This would still allow every token to be send in and output the price that Bob asks for an item.

The other option is that you can send every SAFE token to Bob’s wallet and the token value adjust in his wallet, but the price he asked for was paid. This option would diversify his portfolio and spread coins across services on the SAFE Network, but not everybody might see this as a win.

How to implement this gets way over my head, but there are a few places to look for a stablecoin.


Swarm City also developed ‘gas station’, which is intended to allow users to pay for Ethereum gas automatically using ERC20 tokens.


thanks for shearing. i am adding this to my watch list.

1 Like

Ah, you went ahead and created the topic. Good. I was intending to do it, but after mulling some things over… :slight_smile:

@Lee_Travis, my apologies for hijacking the other thread.

Last night I spent coding up some high level suggestion for a smart contract decentralized exchange, based partly on ideas I had a couple of months ago - very similar in the gist to this description by @intrz.
It’s high level in the way that it leaves out the details about the underlying storage/framework (SAFENetwork in this case). Furthermore it simulates agents running the code on separate machines, as actors in the same process. So a lot of simplification. A lot.

The need for atomicity in the transfer step is removed by letting the contracts take ownership of the tokens put up for trade.
I would really like to share the code as to continue refine the idea by reasoning about the code.
I find that I kind of think and communicate more fluently in code… :thinking:
But I want to tie up the loose ends and run simulations on it before I put it up!

Continuing on the subject from that topic:
I’m not convinced by the response from Bancor ppl to the criticism by prof. E. Gün Sirer.
It’s worth trying to use it somehow anyway.

One thing that is not mentioned at all, is that the necessary arbitrageurs will have little effect if there is no other place to exchange the token. So the long tail that would be helped is still needing to be served in some sort of exchange, which they are not, and this is the problem Bancor was supposed to solve.

I think a Bancor-like network absolutely has to be combined with a(t least one) decentralized exchange. NB that there is still no guarantee that the arbitrageurs can do their supposed work, as they still need liquidity elsewhere for this. But the possibility to remove the listing threshold (token volume so low that the fees from it would not compensate the listing overhead) is there when we go decentralized. At least that’s my initial view.

I discern three different subjects here, where the last one, as I see it, is that of the OP Title.

  1. SmartTokens
  2. Decentralized exchanges
  3. In app token agnosticism

The combination of 1. and 2. is what will enable 3. IMO, with some nifty coding.


We’re at the very early stage of tokenization at the moment. Currently most tokens are on ethereum, but that might change in the future.

Whatever platform(s) will be leading platform(s) for tokenization in the future will eventually see tokens for all kinds of stuff. Commodity tokens like digix on ethereum that can be exchanged for gold, but for all kinds of commodities. Tokens representing ownership in all kinds of real and virtual property and much more.

That’s why this is so interesting. I imagine you could have a wallet containing 1 kg gold, a hundred barrels of oil, a 0.1% ownership in some famous artpiece painting and 10 safecoins. Then you go to some safesite and there’s some product or service you want to buy,but you want to keep your safecoins right now and get rid of some oil instead and you just click pay with oil and in the background someone buys your oil tokens and gives you some safecoins which are sent to the service you were paying,but from your perspective you just paid with oil.


Quick question as I haven’t had time to study.

So you are saying that I could have this APPs token and its worth $x and SAFEcoin is worth $y at the time of purchase and the purchase amount is “z” SAFEcoin . Then I can pay with ($y / $x * purchase amount) of the APP’s token?

Is this done by executing a transaction on some sort of exchange? Or is the seller receiving the amount of the APP’s token that is equivalent to SAFEcoin cost amount???

And what happens with volatility where the APPs token was way overinflated for a short period?

I would suggest decentralized exchange. There would need to exist a path with sufficient liquidity for the transaction though. The seller would just receive safecoins or any other tokens that he would choose to accept, but there’s no way to convince every single seller to take payment in absolutely all tokens.

Then there’s the issue of how to atomically transfer multiple tokens. If Alice wants to sell a safecoin to Bob for one bobtoken, then how can Alice transfer her safecoin to Bob while at the same time ensuring that Bob couldn’t run off with the safecoin before transferring his bobtoken. One option would be to transfer it to a smart contract, ethereum style. I think that is what oetyng was suggesting, though I am not sure how or if that is possible without compute being implemented. The other option I suggested would require multisig to support multiple MDs, so that M of N signatures on MD 1 AND at the same time M of N signatures on MD 2 would be required for an ownership transfer which would then be atomic for the two MDs.


And that is coming we’ve been told.

Hey I know NVO has solved this all and coming any day now :wink:


Isn’t that just explaining tokens. I was more after knowing the mechanics of the value transfer.

Also ERC-20, from the link, are done with smart contracts. SAFE doesn’t have those (yet) and may not have them till compute is implemented in a (few) year(s) after launch.

Yes, ERC-20s are just contacts that keep track of everyone’s balances and have some functions for transferring, etc.
For decentralized exchanges (at least the ones being developed on Ethereum), you guys might be interested in this podcast episode. I remember him talking about how various projects are taking different approaches, basically trying to move various parts of the exchange off the blockchain to make them cheaper and more performant.


Ah, I thought what was planned for now was still just multisig for a single MD. I have been thinking a bit more about this concept of multisig for multiple MDs and I think it could open up lots of interesting applications actually. Basically when some group all agrees that a set of conditions has happened they will swap or mutate a set of MDs and all the mutations will happen atomically on each MD, but only once everyone agrees. For example Alice could agree to buy a token from Bob, but also agree that the trade would only occur if Bob bought some token from Charles and Charles bought some token from Dave, so you could have networks of trade that would atomically either happen or not.

I think it’s actually good that compute won’t be finished when SAFE launches. It continues to amaze me that it seems like most of the stuff implement with smart contracts in ethereum could instead be implemented with MDs in SAFE without running arbitrary code on the network itself. If compute was launched simultaneously with SAFE many people might just go the ethereum route with everything, but now everyone will be forced to try out the alternative ways of doing things so once compute is out it’s less likely to be abused for things it shouldn’t be used for.


Looks like lighting network just did an atomic bitcoin to lite coin swap.


Actually you might be right

I suppose the close groups of all the MDs would have to get chained together somehow, to form a larger close group. Perhaps the set of MDs could form something like a virtual MD, with its own Id, that mainly contains a list of all the MDs and other needed data if any. This would add some extra complexity to multisig though.

1 Like

As for Bancor, my understanding is that this is basically a way to create ethereum tokens pegged to bancor tokens. The point of doing this is to allow user generated tokens with a few number of users to have liquidity, by always having their value backed by some reserve.

Basically kinda like the gold standard. On the gold standard supposedly you could take some note to the bank and get gold for it. With bancor your token is backed by a certain number of bancor tokens and as long as the bancor token have liqudity, so would your token.

As far as I can tell, the bancor protocol is only needed for tokens that may not get sufficient liquidity on their own and is a nice way to be able to have user generated tokens that can be easily traded.

On SAFE I suppose a bancor-like mechanism could be implemented and the tokens could just be backed by safecoins.

Trading of tokens that is not backed by another token, could just be done with the decentralized exchange.

Trading of tokens backed by another token could be done by first trading the illiquid token for the liquid token.

Until compute then maybe a third party would have to be introduced for a bancor like protocol. The issue is who’s gonna hold your reserve currency? It probably wouldn’t be that big of a deal though. Basically someone could run a service that would hold your reserve currency in multisig. That owners of that service would not be able to run away with your money, but if for some reason they stopped their service you could risk losing your reserve currency as you’d need their signature to release it. The trade itself would be similar to the decentralized exchange. You send your token to the currency backing service and they would send their reserve (safecoin) back to you in return and the transfer would have to be atomic to ensure that none of you could run away with tokens.

Interesting, but what is the point of this?

If there’s a fixed rate between my token and Bancor I may as well trade Bancor. What’s the point of creating a new tradeable token if it’s value is pegged to something else?

I’m missing something! :slight_smile:

The Bancortoken is now a free floating token and it’s holding it own. The idea is more to create liquidity among the tokens on the SAFE Network.

Maybe this will give a better idea why we want every SAFE Network token to pay for every SAFE service. In this scenario it’s pay everybody with every currency (or supported currency at the moment)

Our users shouldn’t think about exchanging tokens.

My question was about why create a tradeable token that is pegged to something else (Bancor in your example).

Yes it makes it liquid, but I might as well have just used Bancor if the rate is fixed.

I can see some might do this as a form of marketing, but I don’t see any other benefits.

In Tether’s case I think it’s pretty clear (stability of the dollar, if you can call it stable).
Bancor no longer has the pricefloor of 0.01 ETH, they did this to give investors the option to exchange their BNT back to Ether. Now the Bancor token’s price is determined by supply/demand.
Why peg something? For USDT you need the privatekey (freedom), on an bank account your money can just be frozen (Cyprus).

Just creating apps on the SAFE Network is not enough, because people don’t care about privacy or decentralization (Facebook had 2.07 billion monthly active users), our luck is that app developers care about cutting costs. With the Abra app you can pay in euros and somebody else receive pesos and can pay another person in dollars. Value is translated without any waste of time on the users side or psychologically forcing them to look at the prices of tokens.