I’m not sure if that’ll fly. At some point development should be funded by the companies that run on NEM. If you tell people right now that fees for namespaces are held in a special account and that it will prob goto the devs (regardless how deserving they are) that will look fishy as fuck. Especially with the amounts still sitting in the funds.
Now the devs need to be payed but i don’t think this is a way with which people will be happy.
IMHO it would be cool if that account would be used for node rewards ideally automatically (i.e. implemented at protcol level).
2 Likes
Agree with pat. I think there are already enough funds available - especially if we get a better $<->XEM value in the coming years 
And:
In general I dislike the idea that the fees are send to a human controlled account, even if it’s multisig. This may work for now but it’s always a bad idea to involve humans and their specific interests in a technical process, especially if we talk about money. If you ask me, this should be handled by the protocol somehow. Having people involved in the decision of what should happen with the funds is a bad idea for long term. Who should decide? Who has the keys? … I don’t like that.
Have not thougt about the detatils but could it be an idea to distribute the Mosaics fees proportional to all existing accounts as some kind of interest?
As the existing funds will short term get the most of the fees (because they currently have the most XEM) this has the same effect and can support new projects and the devs. Later on more and more will be distributed to the XEM owners - it’s automatic, determinated and no human interaction needed.
Cheers
Owon
1 Like
Those funds are not harvesting so they don’t get any fees.
Please read my state again, I’m not talking about harvesting, I’m talking about proportional distribution to all accounts as interest, not sure if that will work but I think it’s anyway better than just sending it to another multisig account. 
There should be some split eventually between paying for development, paying out to supernodes, and paying into the community fund.
Companies cannot be relied on to pay for everything. Ideally that should be the case, but XBT has shown that things don’t work out in the real world.
Giving money to everyone will make no one happy as it will be such a small amount given out to so many accounts. It would also prevent future development from being paid for, so the system will slowly die, unfortunately 
Well, it’s not the point to make stakeholders happy 
. I just think it’s a bad idea to trust people to mange that fund in a wise way. Who decides? Who controls the keys? Which developer will be paid? Which not? People that today agree in a common decision may have conflicts later on. You always have to trust a party you don’t know! Humans with different personal interest are involved. In future a lot of money may be involved. As Pat said: “that will look fishy as fuck” - And thats why it should be handled without human interaction - at least thats my point of view.
I kind of agree that it doesn’t need to go to fund development at this point as there is already XEM set aside for that, but we do need to think about long term sustainability and this is one way. If there are others, that would be good too, but at some point money will be going into an account that helps fund the main devs. Maybe someday it could be set up completely with account rules with a 64/64 multisig, meaning that even if two people didn’t want to make a change, then it couldn’t be changed. The account rules would then control the funds and disperse them to set parameters every so often. That would be nice I guess.
One problem is we can’t give the funds directly to harvesters. That is because there is a known way to beat the system and it is suspected that NXT is often beat and that is why they get so many scam assets. Just like there was a scam asset for NEMstake. One might ask, “why would somebody waste quite a bit of money, like 10,000 nxt, to make a scam asset?” And the answer is they didn’t.
A person with two nodes and enough stake can wait until one of his nodes is about to harvest and in that block include a sock puppet transaction for creating the scam asset. In that way a person has successfully created an asset but really just passed the tokens from the left hand to the right so to speak.
We need a way to beat that in NEM. We don’t want to have scam namespaces or people squatting on them using this cheat. Therefore the fees must go to an account not controlled by the person making the block. At the same time, its not really clear what to do with these funds. “Use it for funding development” seems like a good enough answer, but maybe somebody can come up with something better.
I wouldn’t mind it going into the node rewards fund as a bonus to people running nodes for the time being until dev funds get low, and then at that point it switch to fund the main devs. But then that gets a bit tricky of when/how to make that pass off.
Hi jabo,
I think you may have understood me wrong, I am aware of the cheat you mentioned, that’s why is made the proposal:
I think it could be realised as a special type of tx, maybe one of the devs can leave a comment on this. At least it would elimiate the need of trusting people and will solve your harvesting cheat issue.
the thing that worries me about that is blockchain bloat. Lets say we have 2000 accounts right now and I need to spread around 50,000, wouldn’t that make me have 2000 transactions? Or is there another way? Like maybe just give it to one or a few accounts based on some factor?
As far as I know someone can’t harvest his own tx fees. Why are asset creations except from that in the first place ?
Just trying to understand this.
Afik the db and chain operates on balances, it may be possible to handle this on one tx. But would need confirmation from devs. If not possible in one (the same) tx than forget my proposal.
the block harvester itself isn’t the one that made the transaction. it is that person’s second account/node with asking for the transaction, except that transaction isn’t being broadcast to the whole network, it is only being broadcast to the colluding node/account.
So sooner or later the cheater gets a block and in that block includes that transaction for their ghost sock puppet account.
right…obviously
Okay how about this.
The fee goes into a special kind of bond account. After 10 new blocks are harvested one of those harvesters gets the fee. I’m sure there is a sane way to have all nodes agree on who it is. I’m not sure if that’s feasable but maybe look at the hashes of those blocks and see which one is the most similar to the hash of the block where the asset was created and then give it to that harvester. If all some are equaly similar then wait until a more similar one comes along (or split it). Noone can controll the hashes so that could be a way. I’m sure the devs could come up with a better way.
I do want to point out that I think the devs should absolutely be rewarded and I’m aware that this would prob be a massive change to the protocoll. I just think there have to be better ways than collecting network fees. I mean people always talk about that there is too little incentive for harvesters and then we take a way some of the fees on top of that ?
it could maybe be split 5 ways and sent to the last 5 accounts that harvested a block? that doesn’t seem like a simple and elegant solution though.
I think the best solution is still to use those funds for dev, pay supernodes, and pay community fund. There are flaws as Owon pointed out, but I think guaranteeing that NEM development is unsustainable and has no long-term future is a more serious flaw.
Mosaics come soon though. So then NEM will have a more or less constant stream of fees into that “fund” while still having a shit ton in all the other funds (that can’t be used anyway without completely crashing the markets and devalueing those funds).
I think currently there is too much focus on funding the future and what not and completely forgetting that we still need to have some XEM to get to the general public.
Yep, that is why I think doing some sharing to the supernodes makes sense. That way anyone can join in AND the network will be supported 
Well, super node bonuses would be a way to spread the fees out more evenly across many harvesters and would get us away from the known cheat we are trying to avoid with giving to a single harvester directly.