Quote from: TPTB_need_war on October 21, 2015, 11:26:21 AM

Quote from: jwinterm on August 15, 2015, 08:57:20 PM

With all the hype surrounding Ethereum launch, I'd like to start putting together a comparison table of all the 2.0 platforms/coins/tokens/thingamajigs. The ones I have in mind are Bitshares, Burst, Nem, Nxt, Qora, Counterparty (Dogeparty/Clearinghouse), and Ethereum. I'm interested in mapping out their similarities and differences, particularly what each is capable and incapable of doing, but also other differences such as validation scheme (PoW or PoS), initial distribution, etc.

With that being said, here is the table:

Link to html version of table: http://myrcraft.com/2p0platformtable.html

I am going to be adding my comments about each of these coins in this post. Please check back on this post from time-to-time as I will be updating it as I complete my research gathering process. I will correct any incorrect information as I learn or it is pointed out to me. The statements below are what I think to be true based on the level of reading I've done to this point.

BitShares 2.0

Any engineer should know the antithesis of reliability is lack of fault-tolerance, i.e. depending on only 1 node for each block validation. Variability in the network hiccups, DoS-resistance, hardware, and other aspects of witnesses will mean that some can't keep up with the 3s block time every time or they drop transactions and transactions need to propagated to further witnesses and blocks. So either will have unreliability on the real-time promise, and/or this will push approval voting for witnesses that are centralized by those who have the resources to defend their nodes and maintain uptime and performance loads. These issues don't matter as much with longer block times and/or lower expectation of tps, but if you are seriously expecting 1000s of tps in 3s block times sequenced (funneled) into a queue of witnesses then issues will amplify exponentially not just linearly.
As far as I know, the 3s block time with 1000 tps will be vulnerable to DoS attacks in terms of incoming transactions, because it is known a priori which witness will be producing each block. All PoS shares system suffer this issue, but less of an issue until you start to go for very fast block times, very high tps rates, and very small transacted values for microtransactions with real-time dependency on performance. In PoW the winning witness is unknown a priori and is randomized. Thus if ever Bitshares becomes a serious competitor to a properly designed high tps system, you can probably expect DDoS attacks on Bitshares causing it to become unreliable and suffer outages that belie any claim of real-time performance. I am sure many users have noticed significant delays and downtime of the Bitcointalk.org site due to DoS attacks. The system could be (re-)designed so each witness delegates transaction verification to a large distributed group of transaction verification servers (e.g. perhaps the other witnesses) and only accepts connections from those servers blacklisting any that DoS, but this means the same transaction can be sent to all the servers thus exponentially increasing network bandwidth requirements and the cost that must be charged to fees. Complex coordination algorithms might be attempted but complexity is increasing, so normally resilience suffers and corner cases proliferate. As the transacted value of the valid microtransaction decreases, the DoS cost per transaction increases proportionally. Viewing this as an analogy, it is well known that in multithreaded systems that contention (coordination) load can starve actual throughput and even to the point of gridlock. This appears to me to be fundamental issue that can't be fixed within the design of round-robin rotation through preassigned order of witnesses. Typically all solutions to coordination and even in Bitcoin's case lead to increased centralization and I am sure Bitshares will find this to be true. They've already centralized by adopting reputation and voting schemes in order to get a deterministic node for each block in order to eliminate network coordination load that plagues PoW system but they just pushed the problem onto DoS and they will need to centralize more to address this fundamental issue.
PoS systems are inherently more centralized because reputation and voting are power vacuum paradigms, meaning that power amasses more power and those who don't play the game theory for power lose power. Democracy is always corrupt for this reason. Thus these systems are the antithesis of the reason we are hear for crypto which is to have trustless protocols without power vacuums that can't be gamed by those who aggregate power. Now I am aware that PoW has analogous failure modes in that coordination load is driving centralization and the 25 - 51% attack exists, but just because that is so it doesn't mean PoS is a solution. And I maybe someone actually has a solution. We will have to wait for that solution to appear and make a choice between lesser evils in the meantime.

DPOS claims[1] that by having the stakeholders in the system vote, that the controlling group which is the corporation comprising the developers is not in control. Well publicly listed entities allow shareholders to vote, and that doesn't absolve the classification of investment securities. Ostensibly Bitshares is trying to not run afoul of the criminal and civil liability that results from unregistered investment securities, but my interpretation of the law[2] is they may be still acting as a controlling group since investors depend on them to add value to the investment and the future performance of the investment (again I am not making any declaration that they are or are not, I am raising awareness on this issue for potential investors and participants).

[1]

https://bitshares.org/technology/delegated-proof-of-stake-consensus/

Quote

This design was chosen to ensure that delegates technically have no direct power and that all changes to the network parameters are ultimately approved by the stakeholders. This is done to protect the delegates against regulations that may apply to managers or administrators of cryptocurrencies. Under DPOS, we can truly say that the administrative authority rests in the hands of the users, rather than either the delegates or witnesses.

[2]

https://bitcointalk.org/index.php?topic=1211093.msg12739508#msg12739508
https://bitcointalk.org/index.php?topic=1211093.msg12722193#msg12722193

The new anonymity features in Bitshares are incomplete and not at the level[1] of Cryptonote combined with hiding values which was independently invented by both myself and separately Monero's Shen-Noether.

[1] https://bitcointalk.org/index.php?topic=1211093.msg12711191#msg12711191
https://bitcointalk.org/index.php?topic=1211093.msg12711400#msg12711400
Quote from: TPTB_need_war on August 27, 2015, 03:56:39 AM
Quote from: StanLarimer on August 27, 2015, 03:52:39 AM
This thread has been created to try to provide a no-hype summary of critical information you might just want to know about.

At launch, the big gold-rush excitement will be about signing up new users since those who do will receive up to 80% of the lifetime fees generated by every user they sign up. (So you can get a share of the blockchain's profits without owning any of the BitShares yourself.)

Is this a decentralized coin or a ~~gym~~country club membership?

Sounds like a top-down corporation to me.

Quote from: TPTB_need_war on August 27, 2015, 05:01:18 PM
Oh it is sort of like Amway but anyone can buy direct, no MLM pyramid.

Ah okay I thought I was a crypto-currency. Now I understand you guys are into [small niche] marketing schemes instead. Good luck.

In a controlled environment its already demonstrated 100k tps.. The bottleneck is the network capabilities rather than the core code which is what it demonstrates. Once network capabilities scale up naturally, then 100k comes possible as demand sees fit.

Your assumption which led to a dozen more about which witness is next to sign a block is incorrect, thus your derived assumptions also incorrect. Thus you really have no claim ablut bitshares and the tps without fully understanding the concepts behind the tests and feature itself.