Yes you can "spam" 500,000 times with a minimum fee of 1, but it costs you 500,000+ NEM 
And yes, no one will pay a $1 transaction fee for a 7$ worth transaction. Thats why there will be a hard fork with a NIS patch at some point. Why are you saying that again?
And yes, there will hopefully be a solution for micropayments.
I assume because mob rule might say: "no fork!!".
If that is the case then there is room for a third party NEM payment processor which can do transactions cheaper. Profit incentive means somebody would do this.
I assume because mob rule might say: "no fork!!".
If that is the case then there is room for a third party NEM payment processor which can do transactions cheaper. Profit incentive means somebody would do this.
Yeah, I don't see why a microtransaction payment processor could not be part of the "NEM ecosystem".
Ok so the fee is: x/25000 + ln(x)/5
But if it is smaller than 1, it stays 1.
Thanks
Edit: And you are rounding up. So a calculated fee of 1,001146748 becomes 2.
Yes: max(1, ceil(x/25000 + ln(x)/5))
I would be more inclined to look at a dynamically changing formula base on an established external event if possible (possible as in externally triggered event incorporated into the program so that we don't require hard forking).
Assuming this is possible, then I would propose the following:
Max(y, (xy/2500 + ln(xy)/5), where y is a derived value based on some external event. y is set as 1 today and more than likely y < 1 with the passage of time. But y is not necessarily always <=1. Sometimes it will scale beyond 1 depending on the benchmarked event.
Max(min, (x/25000 + ln(x)/5)) where min is a derived value based on some external event. min is set as 1 today and more than likely min < 1 with the passage of time. Likewise, min is not necessarily always <=1.
I prefer the first one because the first one will shift the entire curve. The second one is not so effective.
Finally, what is this external event? Who will determine it? I don't have an answer yet, but my first thought will be a rule based account event, executed once a year to determine y. I will be more inclined to base it on some fiat value. What immediately comes to my mind is the formula I used for my patent that is based on fiat. ;)
Edit: Can NIS take on a dynamically changing external value? If yes, then we shouldn't need a hard fork.
So then relying on an outside source to control payments is good? How is this different then the systems there are in place now? As for forking it, hopefully the client will either:1) auto update itself or 2) when the new client is released the old stops working and people will have to update to it. I would prefer the first option as this would be easier on average joe.
I assume because mob rule might say: "no fork!!".
If that is the case then there is room for a third party NEM payment processor which can do transactions cheaper. Profit incentive means somebody would do this.
Yeah, I don't see why a microtransaction payment processor could not be part of the "NEM ecosystem".
I assume because mob rule might say: "no fork!!".
If that is the case then there is room for a third party NEM payment processor which can do transactions cheaper. Profit incentive means somebody would do this.
That is decentralization in action ^^.
Unless micro transactions can get covered by another piece of software or something. Maybe you send that NEM to your NEMPay which has smaller transaction fees
AltNemo is considering some idea like this for micro transactions. This should help the unbanked populace of the world who lives on a few dollars a day.
So then relying on an outside source to control payments is good? How is this different then the systems there are in place now? As for forking it, hopefully the client will either:1) auto update itself or 2) when the new client is released the old stops working and people will have to update to it. I would prefer the first option as this would be easier on average joe.
I assume because mob rule might say: "no fork!!".
If that is the case then there is room for a third party NEM payment processor which can do transactions cheaper. Profit incentive means somebody would do this.
Yeah, I don't see why a microtransaction payment processor could not be part of the "NEM ecosystem".
Already happens with Bitcoin. A lot of money goes through BitPay and Coinbase.
Payment processor are inevitable because of flaws in the blockchain like exponential growth and slow transaction speeds.
Bitcoin is already 22,000 MB. If there is 100 million people spamming micro transactions nonstop then we probably have a 1 TB blockchain and 2 hour confirmations, even after they fork Bitcoin to reduce the block size and increase the TPS.
So are you saying that because bitcoin does this then everyone has to do this?So then relying on an outside source to control payments is good? How is this different then the systems there are in place now? As for forking it, hopefully the client will either:1) auto update itself or 2) when the new client is released the old stops working and people will have to update to it. I would prefer the first option as this would be easier on average joe.
I assume because mob rule might say: "no fork!!".
If that is the case then there is room for a third party NEM payment processor which can do transactions cheaper. Profit incentive means somebody would do this.
Yeah, I don't see why a microtransaction payment processor could not be part of the "NEM ecosystem".
Already happens with Bitcoin. A lot of money goes through BitPay and Coinbase.
Payment processor are inevitable because of flaws in the blockchain like exponential growth and slow transaction speeds.
Bitcoin is already 22,000 MB. If there is 100 million people spamming micro transactions nonstop then we probably have a 1 TB blockchain and 2 hour confirmations, even after they fork Bitcoin to reduce the block size and increase the TPS.
There are projects out there that are solving slow transaction times as well as blockchain bloat and some of them are working very well. NEM has to find a way of doing this or it will get left behind.
Ok so the fee is: x/25000 + ln(x)/5
But if it is smaller than 1, it stays 1.
Thanks
Edit: And you are rounding up. So a calculated fee of 1,001146748 becomes 2.
Yes: max(1, ceil(x/25000 + ln(x)/5))
I would be more inclined to look at a dynamically changing formula base on an established external event if possible (possible as in externally triggered event incorporated into the program so that we don't require hard forking).
Assuming this is possible, then I would propose the following:
Max(y, (xy/2500 + ln(xy)/5), where y is a derived value based on some external event. y is set as 1 today and more than likely y < 1 with the passage of time. But y is not necessarily always <=1. Sometimes it will scale beyond 1 depending on the benchmarked event.
Max(min, (x/25000 + ln(x)/5)) where min is a derived value based on some external event. min is set as 1 today and more than likely min < 1 with the passage of time. Likewise, min is not necessarily always <=1.
I prefer the first one because the first one will shift the entire curve. The second one is not so effective.
Finally, what is this external event? Who will determine it? I don't have an answer yet, but my first thought will be a rule based account event, executed once a year to determine y. I will be more inclined to base it on some fiat value. What immediately comes to my mind is the formula I used for my patent that is based on fiat. ;)
Edit: Can NIS take on a dynamically changing external value? If yes, then we shouldn't need a hard fork.
That's the right thinking I guess.
What I had in mind is some parameter in the formula with which it is possible to consider the value of NEM in a stable unit. Maybe an average of the big exchanges and average of different big/stable currencies. But that is a really complex problem because exchanges are coming and going as well.
Base it on XE.com forex rate snapshot at a particular time every year. But if XE.com closes shop we will have problem.
Maybe in the absence of a data feed, we have multisign approval for another source. But we are mere mortals. So, we need to have the ability to add more signatures with the passing of someone or time.
My answer to the problem is to plot the growth in the price of three major cryptosphere, maybe bitcoin, doge, and NXT. Get an average of price increase in ratio to block number and then just make a wild guess.
Assume that by block # xxxxxx NEM will be worth xxxxxx and then will need to have the fees changed. This pattern would then repeat after a certain time.
Gavin is proposing a hardfork in bitcoin right now to happen in a few months that does just this same thing with bitcoin because it too is going through growing problems. Except his proposal isn't with fees, but with block size.
What people are forgetting is that by having 1 million NEM per stake really opens NEM to block chain bloat right from the gate. For example:
The formula is: max(1, x/25000 + ln(x) / 5) where x is the amount in nem.
I think the blockchain bloat is solvable but we have no whitepaper on that.
Someone doesn't like NEM buys a stake from the NXT AE, the day NEM goes live he starts sending out 1 NEM transactions with a 1 NEM fee so that gives him the ability to send out 500,000 transactions (more if they lower the transaction amount). I know it will take awhile but in a month more then 1000 transactions could be sent and this is on top of the other transactions people are making. Now if they share their NEM with two or more people well...
So tie NEM's transaction fee to outside sources plus make a wild guess? If you think about this for a sec that is really scary.
My answer to the problem is to plot the growth in the price of three major cryptosphere, maybe bitcoin, doge, and NXT. Get an average of price increase in ratio to block number and then just make a wild guess.
Assume that by block # xxxxxx NEM will be worth xxxxxx and then will need to have the fees changed. This pattern would then repeat after a certain time.
Gavin is proposing a hardfork in bitcoin right now to happen in a few months that does just this same thing with bitcoin because it too is going through growing problems. Except his proposal isn't with fees, but with block size.
So tie NEM's transaction fee to outside sources plus make a wild guess? If you think about this for a sec that is really scary.
My answer to the problem is to plot the growth in the price of three major cryptosphere, maybe bitcoin, doge, and NXT. Get an average of price increase in ratio to block number and then just make a wild guess.
Assume that by block # xxxxxx NEM will be worth xxxxxx and then will need to have the fees changed. This pattern would then repeat after a certain time.
Gavin is proposing a hardfork in bitcoin right now to happen in a few months that does just this same thing with bitcoin because it too is going through growing problems. Except his proposal isn't with fees, but with block size.
I know it sounds crazy, but I think a shot in the dark is better than no shot.
With the current system, there will definitely need to be a hard fork and serious discussion in the future. With my suggestion, there only might need to be a hardfork.
So tie NEM's transaction fee to outside sources plus make a wild guess? If you think about this for a sec that is really scary.
My answer to the problem is to plot the growth in the price of three major cryptosphere, maybe bitcoin, doge, and NXT. Get an average of price increase in ratio to block number and then just make a wild guess.
Assume that by block # xxxxxx NEM will be worth xxxxxx and then will need to have the fees changed. This pattern would then repeat after a certain time.
Gavin is proposing a hardfork in bitcoin right now to happen in a few months that does just this same thing with bitcoin because it too is going through growing problems. Except his proposal isn't with fees, but with block size.
I know it sounds crazy, but I think a shot in the dark is better than no shot.
With the current system, there will definitely need to be a hard fork and serious discussion in the future. With my suggestion, there only might need to be a hardfork.
How expensive or inexpensive something is unfortunately almost always gets benchmarked against the fiat. Hence, like it or not, it makes sense if we benchmark the transaction against a currency basket.
We also need to understand that if we are decoupled with the inflationary value of fiat, it will definitely be too expensive one day to transact.
I believe the topic of transaction charges is not as simple as we think it to be. It is far more complex when we start to compare with a system in the current financial world that has evolved for over 3000 years. I guess there have also been many hard forks in the financial world that we live in. Just that we never took notice of them. Decoupling gold from the fiat is one of them.
right, the reality is if crypto in general and NEM specifically grows to the point that we think it will (that's why we are here), none of us really know what will happen. it is a game changer.
What about tx's per day? If the number of tx's goes above x they y(fee) drops by z… Is it Metcalfes law that says price is relative to number of transactions, varies above and below but it averages out about right or something?
What about tx's per day per IP?
Well if NEM is a game changer then shouldn't it offer lower transaction fees? If NEM is counting on supporting the network with transaction fees then it made a big mistake. After a couple of weeks it being alive and once people are done trading it then the transaction level will be very low, only a very small amount of people will support the network. If you are worried about chain bloat with lower fees then read my post above about it. Doing hard forks has more cons then pros.