I think that in order to get more efficiency from the budget system we need to do one or more of the following things:
a) Lessen the time factor--There is a real sense of urgency, in that we have only one month (at most) to plan and make proposals to the network before the funds or irretrievably "lost" for future development.
b) Given that any unallocated funds are lost forever (by simply not being created), we have a situation in which there is essentially zero cost for failure. If a proposal doesn't work out, then so what? The funds would have been lost anyway. The flip side is that almost any proposal with even the slightest chance of bringing value to the ecosystem is worthwhile, from a strict cost-benefit analysis. Take the latest proposal to integrate with a major exchange: that's just over 2,000 DASH or $30,000 USD. Not spending that money would effectively increase the value of all DASH in existence by half a cent, which is infinitesimal.
c) Communication must be improved. The creation of the new subforum is an excellent start.
My suggestion is to allow the network to "bank" any unspent funds so that they can be used at a later time by worthy projects. This has several advantages:
1) There's an incentive for saving. Today's unused funds will likely be worth quite a bit more in the future, due to price appreciation. This month's proposals, therefore, essentially have to compete against the possibility of even better proposals in the future. We don't want to fund marginal projects today, since that takes money away from potentially better projects in the future.
2) It eliminates the time factor, since there's no longer a hurry to spend funds before we lose them (i.e. before the month's superblocks are created). That gives more time to craft better proposals, more time for communication, and greater ability to hold contractors accountable (since there's no rush to approve expenditures).
3) It alters the cost-benefit analysis, because now the cost of disapproving a proposal is lower and the cost of approving a proposal is higher, since the funds if unused would be available for future projects.
P.S. It is vitally important that masternode owners realize there is risk in everything, and that it is up to them to assess the risk. There have been, and will continue to be, budget proposals that simply don't work out very well. That's simply how the world works--there are no guarantees in anything. All we can do is assess the risk and make the best possible decision with the information we have. If something doesn't work out, we learn from it and use that knowledge later. Failure is to be expected; if we never fail, it's because we never aspired.
Couple of VERY important points you bring up on this that I've never understood and never agreed on...why can't the funds be "banked"
This urgency to spend the funds before they are "burned" is ridiculous. This should be addressed...unless there is some compelling technical reason as to why we can't do this, this requirement should be removed.
And second back to my original point...and one you touch on here...we need a very clear way to assess the performance of the budget system. Certainly we are all grown ups here(well most are here...
lol ) and can assess risk using our own criteria. The problem is that the information to make this kind judgement of system is incomplete and not readily available. This is quite problematic.
So to recap...2 things need to be addressed, we should be advocating the removal of the time limit on spending the funds in a limit time frame and 2 creating a portal to which not only masternode operators but the community at large can gauge the performance of our system.
The issue was a technical one if I remember correctly. People balked at members of Core being in control of the "bank" (for good reason) -- and it wasn't feasible for funds to be in a n of m multisig account as the the m (number of masternodes) is always changing. If fact, I don't think a proposal can even be paid out to any type of multisig account currently...
12.1 and Sentinel may allow such a functionality and it does make sense to explore this "bank" if it is technically feasible. A dev can probably offer more insight.
Just to clarify, by "banked" I mean "uncreated, but able to be created in a future superblock." Essentially, I'm suggesting that the network "remember" how many Dash have gone uncreated, and be able to create that many Dash in a future month.
For example, under the present system (hypothetical) with a maximum budget of 7450 DASH per month:
Month 1 -- 5218 DASH paid to proposals, 2232 unpaid and never created
Month 2 -- 6938 DASH paid to proposals, 512 unpaid and never created
Month 3 -- 6277 DASH paid to proposals, 1173 unpaid and never created
Month 4 -- 9000 DASH in proposals have enough votes to be paid, but the limit is still 7450 DASH, meaning *at least* 1550 DASH worth of proposals end up being unpaid
Under a new system where unused funds are "banked" or "remembered" by the network:
Month 1 -- 5218 DASH paid to proposals, 2232 unpaid and never created, network remembers that it can create up to 2232 more DASH
Month 2 -- 6938 DASH paid to proposals, 512 unpaid and never created, network knows that it can create up to 2744 more DASH
Month 3 -- 6277 DASH paid to proposals, 1173 unpaid and never created, network knows that it can create up to 3917 more DASH
Month 4 -- 9000 DASH in proposals have enough votes to be paid, network pays all proposals, since there are plenty of funds to do so (7450 DASH for the month, plus 3917 previously "banked" or "remembered" DASH that could have been created, but weren't)
Now, you don't want this to go on forever; uncreated DASH eventually have to "expire." The original discussion of DGBB involved the possibility of "banking" uncreated DASH, but the worry was that superblocks might end up creating an insane amount of DASH one month due to all the uncreated DASH being expended all at once. There's also the risk of perverse incentive: if a massive amount of uncreated DASH was remembered by the network, a few masternode owners could collude and create a worthless proposal that would vote all the funds to themselves. It would hurt Dash, but they huge payout would make it worthwhile.
I think it would be beneficial for the network to have a three month "memory." Nothing crazy, but enough to be a little flexible.
