Quote from: TPTB_need_war on October 25, 2015, 12:51:34 AM

I have of course been thinking a lot about how the competing technologies for block chain scaling compare to mine (linked bulleted items below are to the prior research I did):

Lightning Networks (LN)
Iota (DAG) tangles (IoT)
eMunie (vaporware not even a white paper)
Bitshares 2.0 (BTS)
Dash Evolution

Lightning Networks

I have a unique insight on the apparent fact that LN is unique in that it is the only one wherein not all participating "full"[1] nodes have to see all the microtransactions processed by all the nodes, thus it scales exponentially and qualitatively better in the sense that not every "full" node on the network has to have the same (least common denominator) baseline of network connectivity and CPU processing power (although BTS claims 100,000 TX/s single-threaded throughput for block chain updates on commodity CPUs so perhaps CPU isn't a practically limiting factor in any case).

LN is not compatible with existing block chains (including my opinion it will wreck havoc on Bitcoin) because it will requires extreme levels of block chain scaling because it will drive a proliferation of payment channel opening and closing transactions and even competing payment channel networks, each requiring a distinct TX on the block chain. Thus while LN scales the microtransaction volume well, it paradoxically requires a block chain that can scale well also. Thus I have come to the conclusion that LN is only really compatible with my PoW block chain redesign. Although my block chain design can scale transactions very high, it will also have this aforementioned (least common denominator) baseline which will probably limit the network to about 100,100 TX/s on commodity servers for "full" nodes.

LN also has the downsides that it isn't end-to-end principled and can't be always available to between every payer and payee, which is where when combined with my block chain design, then mine fills the gaps to provide the end-to-end capability and the complete availability. It can't provide anonymity (well at least not the ring sigs and value hiding that we can get on the block chain). And realistically LN will require trusted commercial servers to make it function well (not trusted with any funds, but trusted to not DoS and to optimize payment flows with complex algorithms and global knowledge, reduce latency, etc). Thus LN will be a spy network for government and corporations. But sometimes users don't care and they may need this extra scaling (and probably even lower TX fees) for the tiniest and highest volume of microtransactions.

Each of my design and LN separately (and especially together as I am proposing above) can continue to function reasonably well in the event of a partitioned internet and network downtime without allowing double-spends to wreck chaos on the partitioned forks, whereas the existing (PoW and PoS) block chains end up in chaos with double-spends on each partitioned fork.

Iota (DAG) tangles

My analysis thus far (subject to change if I learn something new about this) is that the only significant advantage offered by this interesting new technology is that the issues with large block sizes for block chain scaling are avoided, which my block chain scaling design fixes without incurring some of Iota (DAG tangles) faults. The "full" nodes still need to see all the transactions on the network. I don't really see how it can handle network partitioning since there is nothing to prevent double-spends on multiple partitions. And the confirmation times are not yet characterized in a way we can quantify in comparison, so I can't say how fast it can confirm transactions. I expect my design to be less than 1 second.

Bitshares 2.0

Digging through the hype, the actual performance is only less than 100 TX/s on today's commodity hardware and networks.

Dash Evolution

Main issue is it doesn't address the large block chain blocks aspect of the block chain scaling issue facing for example Bitcoin. And the claimed immunity to 51% attacks is really just the factual obfuscation that masternodes are making too much money (officially endorsed form of cheating) to cheat the design flaws. The lead developer of Dash replied.

[1] "Full" node means different things in these different designs, but I have equated them in the sense of the node that must have global knowledge about either all the transactions or in LN's case all the payment channels.