With the current difficulty, a single 4.6 CLAM output/claim would be likely to take a great deal of time for a user to "claim". So much time, in fact, that it may not even be economical to do so.
An alternative might be to give undug CLAM additional weight and chance to stake. This would make the process of claiming more likely, but still limit digs overall.
This is, however, probably not fair to existing stakers.
To make it more fair, the block reward could be removed from digging stakes. This would mean that undug CLAMs would not give a 1 CLAM reward when staked.
This, however, would be a change to the money supply/inflation.
An alternative might be to attribute "missed" stakes, due to claiming stakes, to normal stakers either at the next block, or into a pool/window spread out over subsequent stakes.
That would leave us with a situation where unclaimed CLAM have additional weight to stake more quickly, but are still limited based on current difficulty and the block time. When unclaimed CLAM stake, they would not give a reward, but instead add their 1 CLAM reward into a pool. When normal blocks are staked, that pool would be apportioned out.
Not sure how I feel about this idea - interested to hear what dooglus, xploited, and the rest of the gang think.
It is definitely more simple than the idea I outlined yesterday; though without some of the additional advantages.
Will give it some thought.
All this just to make it fairer to require that distribution outputs stake before they can be dug? I don't see how that solves our "problem". Maybe it will slow down the digging, but it won't change the end result - the active supply is getting inflated 50% by someone who plans to dump all the new supply. Dragging that out so it takes 2 years instead of 1 year doesn't help us, I don't think.
Requiring that old outputs have to stake before they can move also destroys fungibility. Some coins in my wallet would be of a different class than others. That was one of things you seem keen to avoid.
That was an idea suggested by a user earlier - I was playing devil's advocate and attempting to build on it to make it more reasonable.
I think your fungibility argument is accurate.
The digs would be placed into a separate state, such as immature or a new state labeled "unclaimed" or something.
Interestingly, it would incentivize new users to have familiarity with the client and staking - a net benefit for the network.
I haven't worked out every alternative, but I think it could be set in a variety of ways - the key idea being to find trustless variables which give us data on either demand or supply.
blocksizeThis creates a sort of conglomerate "supplier" and artificially sets supply. You would have a moving window for target blocksize. Target blocksize would be based on a window of previous blocksizes, increasing when average blocksize is greater than the previous target and decreasing when it is below. This would usually create a bloat risk whereby an attacker could create spam transactions perpetually forcing the blocksize upward. Enforced minimum-per-byte-per-block fees would positively correlate and rise with block size. This should make demand move opposite of supply until they find equilibrium. It would also make bloat attacks such as the one above more costly the farther from the natural equilibrium it is pushed. Space above the target blocksize would be available for transactions, but come with either a progressively higher enforced minimum per byte or a reduction in subsidy which would make in uneconomical for a staker to include transactions below a certain per-byte-per-block fee. This would allow additional space in each given block above the target, at the cost of increased fee - which gives us DDoS resistance without artificial ceilings.
coin-days-destroyedThis gives us a measure of money supply velocity. It could be viewed as a proportion between days-destroyed / total-days-created or in other forms. This is a measure of demand. I haven't dug into this much but it seems like there should be a reasonable way to utilize coin-days to measure transaction activity/velocity and adjust the fee upward during times of increased activity and downward during times of decreased activity. In this case, supply would be a vertical line and the system would find equilibrium by adjusting the demand curve upward or downward via adjusting fees which would in turn place inverse pressure on the measure of demand. This would result in a supply-demand gap at equilibrium - but, in this case as smooth pointed out earlier, "suppliers" don't exist in the conventional microeconomic sense, so this likely doesn't represent the inefficiency it would in a normal market.
subsidy reductionIn this case, we apply a cost to supply. It is similar to the target blocksize window idea hinted at in that idea, except we do not have a target. Instead, we create a progressive curve that penalizes subsidy on blocks as total byte-size increases. This decrease must be offset by the fees paid in the transactions included (which raise total byte-size) and suggests that a staker would be unwilling to include transactions in which the cost of inclusion is below the reward(fee) of inclusion. This turns the vertical supply line into a sloped line and places downward pressure on demand via fees.
Pretty much anything which applies pressure on either the supply and demand curves could be used to create an inverse relationship and adjust them to find equilibrium.
In the end, even utilizing trustless chain data you end up picking a sane floor/ceiling and drawing a curve or line between the two. The additional data simply shifts that line upward or downward.