... and the post has now being removed. Frankly fixing Darksend could be worth a lot of money so no hard feelings towards TPTB_need_war for trying. If TPTB_need_war actually has a solution, I am sure the Dash community will pay him for it and it will be money well earned. Does the Monero community have to be concerned about all of this. I doubt it.
All mixing that is
active requires, obviously, activity. If you don't get activity up you won't ever significantly increase transaction speed of CoinJoin / DarkSend transactions. Fortunately, Monero mixes
passively and therefore doesn't require activity of other participants on the network.
Actually there is only an activity threshold above which offchain mixing can be just as fast as onchain. Actually to do ring mixing correctly so that rings never can overlap in ways that allow combinatorial unmasking, Monero should require activity, but I was apparently never able to get Shen-Noether to understand this during our Reddit discussions last year (because he is a condescending prick in the same mold as Gregory Maxwell who thinks he is too smart, actually they appear to hobknob together sometimes), so afaik Monero remains "broken" (suboptimal). The advantage of adopting my idea for preventing combinatorial unmasking, is it would also make the block chain entirely prunable, not just compressable (which afaik is what Monero and BBR erroneously label "pruning").
The key breakthrough is to remove the simultaneity requirement (lol, I am the one who fixes Gregory Maxwell's broken CoinJoin in 10 minutes of my spare time while my head is deep in designing a programming language), and on further thought I've decided I want to embarrass noobtrader (to show my appreciation for his disrepect) so I went ahead and I think figured out how to eliminate the simultaneity requirement in CoinJoin! I figured out how to eliminate the short-term trust aspect! Another advantage is it can radically improve the robustness of decentralized exchange as well.
It will reduce the block chain size considerably. I also see how to put a viewkey in it. And the mix anonymity sets can be huge, say 50 or 100 transactions per mix (or more!). The disadvantage is the masternode can see the correlation of inputs to outputs. But just like any mixing method, if mix over and over, the probability of your anonymity set being known to any one party diminishes in probability.
You'd still need stealth addresses to achieve the delinking from the recipient's public key.
Another potential advantage may be that this technique I've just invented gives you IP address obfuscation inherently, which is one of the big weakness of Monero. Monero adds I2P integration to attempt to overcome this weakness.
My apologies to ArticMine. I am rushing so much, that I mistakenly (cross-eyed) attributed the above quoted disrespect to him.Edit: my discovery for offchain mixing (fixing CoinJoin) to remove the simultaneity requirement, is essentially something like a ring signature, but the signers don't need to include the other signers in their signature. The ring is formed by the masternode. It is quite clever. I don't know why I didn't think of it before! The reason Gmaxwell didn't think of this, is it because it requires the knowledge I did for the fixing the atomic DE protocol of TierNolan. The insight comes from the work I did there.