Ethereum's solution to scaling is worth mentioning since I do cover EOS a great deal. Ethereum is developing the Raiden Network. To understand the Raiden Network, it is easiest to just understand the Lightning Network.
The Raiden Network is a system of payment channels between established, regular payment arrangements. Using encryption and the methodology of the lightning network idea, two parties can create a payment channel to handle their peer to peer transaction off-chain. This off-chain arrangement reduces fees, and reconciliation to the blockchain can occur afterward or periodically through their agreed arrangement.
In my view payment channels only works for businesses, and in most cases, B2B (business-to-business). The only business-to-consumer model I have seen for the lighting network design is micro payments for video streaming, but once such a video stream were reconciled to the blockchain, it would still cost the standard network fee, making the whole thing impractical today.
A Future of Not Owning Your Private Keys
If in the future, the only way to receive reasonable fees for Bitcoin and Ethereum will be to have your funds stored on a business website, willing to setup payment channels with others in the industry. We then have to conclude this will economically force people to keep more funds on 3rd party websites.
In a space where everyone has been advised to store funds on personal wallets, and have control of your private keys, every month that passes, this becomes less economical.
Payment channel fee reductions will just further incentive this behavior of keeping funds on 3rd party websites, rather than in private wallets.
Bittrex could create a payment channel with Coinbase, and both would experience drastically reduced fees, and once daily or even multiple times a day, they could settle the payment channel and reconcile it to the blockchain. When payments are being sent back and forth between two parties, the lightning network protocol makes a lot of sense.
For an individual user, sending a one time transaction to move their funds from a personal wallet to an exchange, they will not be able to use a payment channel. It simply will not fit that format.
I imagine, cloud wallets will be the solution, where the cloud wallet holds your funds, and they are able to conglomerate enough movement of funds to create a payment channel with exchanges such as Bittrex and Coinbase. This still means a loss of controlling your private keys, unless a solution is found.
My Preference is DPOS - (Bitshares / Steem / EOS)
The alternate solution is scaling on-chain, where users are able to keep full control of their private keys, without suffering drastically higher fees. This also has the advantage of making it very transparent to investors and other users, as to the actual network traffic occurring on a given blockchain, rather than moving all the transactions to black boxes, because blockchain is inconvenient for scaling.
I assert, blockchain is not inconvenient for scaling, but rather, developers have lacked imagination and innovation to resolve the core problem of scaling.
EOS aims to achieve 1 million transactions on the blockchain. With other technology and development choices, they believe businesses will be able to go much higher than that threshold.
I will say, though, that the market does not always favor the better method. In this case, however, EOS has many other advantages that the market will certainly favor. Even if Ethereum transactions per second get raised by moving it off-chain, on-chain will still remain limited, and also the on-chain block times are slow. EOS will give a sub 2 second response time. On-chain Ethereum will never do this.
Dash's Instant-X Architecture
It is also worth noting that Dash is scaling without moving transactions off the blockchain. InstantX is their technology to have masternodes confirm transactions in the space of time similar to credit card processing.
Their latest goal in this area, is to beef up their masternode system so the masternodes have minimum requirements in terms of response time, and hardware requirements. The goal here is very similar to the DPOS method, where you have a select group of decentralized servers that meet a specification threshold, so they are able to serve the entire network with fast response time and transaction speed.
Scaling On-Chain is Possible - Lazy, Unimaginative Developers are to blame here
Bitcoin and Ethereum have chosen a path, which in the future, may haunt them. LN, while a neat feature, it is not a true scaling solution. It also incentives their networks to behave poorly.
Increasingly, we have seen from both Bitcoin and Ethereum, that their protocols force more and more centralization, and now that their projects have both chosen this direction, we will also see more centralization of funds in the future, as well.
It's a recipe for a catastrophic event.
Crushing the internet with knowledge, Ted! Let's go EOS!
EOS should have "FTW" as its monicker.
I like that you are pushing the discussion beyond just Lightning Network, which I am super excited for, but I know that it is not the holy grail of scaling Bitcoin that we will ultimately need. I am very curious about the DPOS model, and would like to hear some counter arguments to it before making up my mind and just jumping on the EOS bandwagon. I will definitely be doing my research. Thank You.
It is claimed by some, DPOS is not centralized. Steemit operates on a system of 20 witnesses plus another 100 or more witnesses.
The claim then is, "See, there are only a 100 points of failure! Take down the 100, and the system collapses." It's a true statement, but the argument neglects how centralized Bitcoin and Ethereum has become with the centralization of mining pools and professional mining operations.
This was a criticism Vitalik brought against DPOS and EOS recently, and I elaborated this point of Ethereum centralization fully. My conclusion was that perhaps 30 points of failure exist for the Ethereum protocol, to determine what it is, and it's future. Something like EOS when it launches would have perhaps 60 to 120 points of failure.
The argument of Bitcoin and Ethereum having widespread decentralization is really not a true argument, but it's the main argument brought against DPOS.
https://steemit.com/eos/@crypto-investor/centralization-of-cryptocurrency-bitcoin-ethereum-and-eos
Agreed. the centralization direction will also see more and more hacking incidents.
What are the implications for the EOS project right now then? It seems as if their project next summer might be a day late and a dollar short now!
With all of the pressure from Chinese markets and the ICO crackdown the situation is looking a bit grim for EOS. I still have a majority of my trading portfolio invested in them however since I got a smoking deal on them at 79 cents each token.
No because EOS scales on-chain, Ethereum does not do this. Also, EOS provides the sub 2 second response times among other features such as parrallel processing, Ethereum will never have those features.
EOS also with architecture choices will still do more tps than Ethereum. Worse case, they could implement lightning network on EOS and easily do hundreds of millions of transactions per second if needed. From what I have read, they are aiming for 1+ million transactions per second on chain, and with tech choices, it could be multiple millions per second on chain without resorting to the LN solution.
Why cant ethereum developers do it in blockchain
They would have to gut Ethereum and rebuild it from what I know. Back in the day when Ethereum was built, there was a false notion that having longer block times in the order of minutes meant more security. This concept has been proven false in the last couple years, but Bitcoin and Ethereum are still ideologically married to slow block times.
Really, Dash's team understands the situation just like Dan's team does... for networks to work as well as they should, you need a select group of servers providing the backbone of the network as efficiently as possible. They need to be servers tied into internet backbones, with low latency, and system requirements. This backbone needs to confirm transactions in under a few seconds.
Ethereum just wont take that path for ideological reasons. Also, if they tried, they'd probably have another alt-coin on their hands and a lot of complaints from people who fail to understand what is going on.
Isn't "a select group of servers" a centralizing aspect? Which would also increase single point(s) of failure?
Raiden Network is blazing fast ;)
Thank you great post!
Nice to read such a great post. Love it.
Great post,
Follow me for a follow back
its a very helpful post, thank you :)
We think alike crypto-investor!
Our own recent update of the RDNs...
https://steemit.com/cryptocurrency/@harpooninvestor/hot-crypto-of-2018-the-raiden-network