hrmm, kinda got rambly and my thought train seems to have jumped the track. damn.
Following your (criss-crossy) line of thought, here's another ramble on pools and miners:
I think that pools are a
necessary evil. The smallest of miners, would be practically impeded of participating on the mining network, if there were no pools, and given sufficient nethash that their average time to find a block would be leaning towards infinity (exaggerated, yes).
On the other hand, pool concentration, and the lack of a technical necessity for running a node (for the majority of participants), present us with a very common dilemma:
nodes are too few, and they lack any substantial form of incentive.I think this problem was indeed neatly approached by Spreadcoin, in that it attempted to prevent pool formation entirely. I think it failed at the original goal, but the fact that it resulted in a modified trust model (note the existing private pool for Spread) is quite interesting.
I'm not sure the same existed for Spread as I am not sufficiently familiar with it, but with Ziftrcoin, I know that there is an incentive to solo-mine, built on some of the same principles that Spread initiated, and by providing knowledge over the current transaction pool, solo-miners can claim a higher reward in their block finds (5% IIRC).
I think it is this latter approach that deserves some further thought. If we could work out a way to incentivize solo mining, and at the same time encourage more full nodes to exist, we'd see a much healthier network (as well as more secure, relatively speaking ofc).
I do believe that a 5% reward boost is too small of an incentive to pursue this approach sensibly, and would suggest something as high as 20%. My rationale is that: technically, solo-miners are usually at a disadvantage, because part of their mining resources are implicitly tied up with the business of running a full node (hence, some potential hashrate performance is lost). Additionally, external miners will usually perform best, seeing as they run code that is systematically being optimized for that function alone, and are less concerned with wide cross-platform support and code portability (compared to a full node).
Personally, I run a full node and I'm happy solo mining, even though I get a performance drop when compared to pool mining. Thanks to Smooth's recent updates, the performance drop is rather tiny (in the region of 5% or so), though my motivation is not so much an economical one, as it is of participating with a full node and contributing towards a healthy network.
Most miners are economically motivated, period. I think it would be nice to leverage on that motivation, and drive some of these miners towards a more meaningful contribution to the network (solo-mining, full node incentives). This ramble is just my collection of past ideas on this matter, and food for thought as much brighter minds are around
