The asset issuance fee of 0.5 XCP works well at the moment. It’s high enough to discourage squatting and low enough for anyone to issue an asset.
If the XCP price increases a lot, the fee will be too high for many desirable uses (e.g. issuing voting tokes). We should also be prepared for big drops in the XCP price. Say the fee was reduced to 0.05 XCP due to a massive upswing, but then the price crashes. If the fee remains at 0.05 XCP, we might then see a Dogeparty-like scenario where squatters register thousands of assets.
Correct me if I’m wrong, but the fee can only be adjusted with a protocol change. This will always need to be informed some time in advance, so the fee can remain at a bad level for too long.
I know there’s a plan to adjust the fee based on the money supply [ xcp fee = 0.5 * moneysupply/2,600,000 ]. This solves another problem - running out of XCP.
What I suggest is another input as well - the length of the token name. The formula could be
xcp fee = 0.5 * moneysupply/2,600,000 * 7/length
It would reduce the problems I initially described. If XCP increases in price, it will still be affordable to generate longer names. Important projects will always be willing to pay more for a short name. If XCP gets cheaper, it will still be costly to register massive amounts of short, catchy names.
I think that token names will be handled differently (on a separate level) and this will not provide the right incentive unfortunately. The current protocol names are not necessarily the ones that user will see anyway.
I see your point on that protocol names are not necessarily what the users will see. An exchange could for example display descriptions / full names of pre-approved tokens only. Users wouldn’t need to know the asset protocol name.
However, I believe most projects will want a short, easy-to-remember asset name, e.g. Swarm (SWARM), LTBcoin (LTBC), Storjcoin X (SJCX), and FoldingCoin (FLDC). This compares with how companies on the stock exchange are listed with tickers.
A higher price on short asset names would ensure that any future BIG project would get a good ticker. For other uses a longer (and cheaper) asset name will do.
The formula I suggested is of course just one possible implementation. Ideally the fee should be a fixed dollar amount, but I don’t think that’s possible. A vote among XCP holders would perhaps be the best, although it’s easier said than done.
What is “moneysupply” - is that the exchange rate of XCP/USD?
It’s the total amount of XCP in existence. Every time a 0.5 XCP fee is burned, the moneysupply declines by 0.5.
Just discovered that a free human-unreadable class of assets is planned.
It pretty much addresses the same concerns - thumbs up!
… although a simpler solution could be to let any name >12 chars (or so) be free to register?
Yes I saw your comment on Github and I agree with it.
Moneysupply - > Token supply would be better, but it doesn’t matter since now I know what you mean
By the way I was wondering is the way to calculate it just to deduct from the original issuance the number of assets divided by 2?
I think the amount of burned XCP is roughly the number of assets divided by 2, yes. But it’s not exactly. Earlier the fee was higher, I believe. Wasn’t it 5 XCP? There is (or will soon be) as fee for dividends as well.
The moneysupply at the time of writing is 2,647,142.08 XCP.
The dividend fee kicks in a week, 1000 blocks from now at 330,000.
@JPJA, yes, it was when it was “young” and when 5 XCP couldn’t buy you a loaf of bread
They’re trying to strike a balance but it’s difficult - on the popularity changes, XCP/BTC rate changes and people’s view of what may be lucrative changes, so at times 5 XCP seemed cheap, then expensive, etc.
I collected and posted basic stats and links to sources (below), I think it’s good to have it in one place.