Comments I made today on John DeGoes blog.
http://degoes.net/articles/blockchain-myths#comment-3763542281
Regarding Myth#4, I agree that third-party rents will be disintermediated.
However, I disagree that highly successful distributed applications will each separately run their own proprietary distributed ledgers. Afaics, you’re missing the elephant-in-the-living-room, which is decentralization and decentralized ledgers, which is currently a fairy tale and consortium blocks don’t qualify. Users want open data as companies want open source. Cross-chain synchronization (side-chains et al) of consensus is afaics an insolubly flawed concept. As for decentralized scaling, I’m working on that specifically.
My point is orthogonal to the likely proliferation of token types. My opinion is these tokens will be hosted on a ledger (e.g. ERC-20 tokens on Ethereum) that no company controls, because users want open data, not continuing with companies having proprietary control over data as is the case with the Internet now. We’re going to disintermediate all the Facebooks, Twitter, Reddit, Github, etc.. and their retarded centralized control over data. The notion of multiple proprietary side-chains interopting is afaics not secure (c.f. my link above). Byzantine fault tolerance and security of the consensus dictates one-to-rule-them-all. Asserting that we could copy data from proprietary ledgers would not be consensus over those copies. Discussing this point deeper really requires a forum format, not comments.
For example of the egregious negative network effects scaling costs of closed data, Disgus isn’t displaying these comments when I try to archive these pages on archive.is/.org.
P.S. there’s 3 links above but they don’t seem to be underlined or highlighted in my browser. Just hover the mouse over the text to find them. Also you turned me on to Scala in 2009 when I was lamenting HaXe. And I was mentioning Scala/FP to the founder of IOHK (Cardano) in 2013 in Bitcointalk before he created Ethereum.
http://degoes.net/articles/blockchain-stack#comment-3763563535
Please refer to my comment in your prior blog wherein I provided my rationale for disagreeing with some of your analysis of Myth#4. However, I do agree with you that a killer app could disintermediate the “consortium blockchains” which is basically how I classify all of them at this time. Also afaics the term blockchain is not general enough. I prefer decentralized ledger. A technological solution doesn’t necessarily have to be based on non-interleaved blocks.
Also I disagree that users will want to manage many different tokens for payments, yet they may collect many different token types as assets for trading (analogous to baseball cards). I think we’ll end up with only a few or even only one for the underlying token that operates the ledger and for making payments. But there’s a technological bifurcation of the future, depending on whether off-chain and side-chain technologies will be popular given that many of us argue they’re Mt. Box failure directed. I linked to some technical information in my comment in your prior blog.
http://degoes.net/articles/coc-review#comment-3763594792
I must say I detest all these CoC. I would much prefer decentralized curation, which is what I’m attempting to bring to the party. I hope we can respectfully disagree.
http://degoes.net/articles/blockchain-stack#comment-3768272216
John Degoes wrote:
Timm wrote:
There might be things like identity services running on popular blockchains. If you need identity in your DApp, you can't just run your own blockchain.
Yes. I'd say the analogue is how today's SaaS applications often call out to 3rd party APIs—even though they are not "built" on those APIs in any sense of the word.
I expect we'll see use of 3rd-party blockchains for verification or exchange or even deployment of smart contracts.
Again John, I disagree about a proliferation of ledgers. Consistency of data and consensus is a key factor.
Follow-up discussion:
http://degoes.net/articles/blockchain-stack#comment-3769837548
"MMoRPG over the blockchain" do not make sense.
I was writing on Steemit some technological and economic analysis for why I think they make sense (apology not so easy for me to locate the link quickly). And I’m working on (the real-time confirmation) technology to hopefully make it a reality. I agree with John, be very careful the devil is in the details. And you’ll probably be surprised if decentralized ledgers disintermediate nearly everything on the Internet.
Economically afaics, having all the massive server infrastructure requirements available turnkey could enable more Indie-level creativity in the MMoRPG arena thus disintermediating the moat of the well capitalized vendors.
Why would you need a blockchain for it? Don't get me wrong, I am really open for discussion. Which property of the blockchain technology is so needed that the price of bringing in the blockchain complexity is justiced?
In addition to the second paragraph I was adding (as an edit) to my post while you were writing your reply, the decentralized ledger offers transparency and prevents proprietary ownership of the data. Per my comment in the prior blog John linked to, merely replicating data is not sufficient to extinguish proprietary control, because consensus of the data has inertia.
I agree with this sentence 100%: “decentralized ledger offers transparency and prevents proprietary ownership of the data”. But is this a feature that MMoRPG need? As John noted (and using your terminology) running decentralized ledger is very expensive.
Did you read my comment at John’s other blog?
http://degoes.net/articles/blockchain-myths#comment-3763542281
Open data is important because it means I as a user can’t be cheated, can’t be banned, can’t be censored, can’t be locked into one vendor’s proprietary crappy software and APIs which doesn’t give me the range of features I want. MMoRPG’s aren’t just solo games, they’re social networks. Users don’t want some vendor stomping on their degrees-of-freedom w.r.t. their relationships and contacts on the network. So afaics, I think MMoRPG on decentralized ledgers could mean more choice for users and eventually much better games and social networking within those games. Also possibly users become (via ledger enforced transparent contracts) investors in the game by holding the game’s tokens (with their ownership not at the whim, opaqueness, and potential bankruptcy of a vendor controlled MMoRPG). But MMoRPG is an extreme use case and we don’t have to succeed there first. There are much lower hanging fruit with less technical/marketing/hen-egg-critical-mass challenges to address first. Btw, bankruptcy is going become a much more common phenomenon and the need for the youth to find alternative sources of livelihood also will become very important in the coming years:
@anonymint/global-contagion-of-bank-failures-coming
Unlike the recession and stagflation of the 1970s with the high interest rates, inflation, and stagnating economy, we now have 10X higher sovereign debt loads and bankrupt pensions, and instead of being in their prime working years now with an aging retiring demographic bubble of boomers in the West who were born right after the soldiers returned home from WW2. The fundamentals for the West are much worse than they seem because of hanging by a thread with the ZIRP/NIRP that was buying the sovereign debt to keep kicking the can down the road, but the day of reckoning is approaching when some governments can’t sell their bonds and default rendering their countries/states/cities into third world zones analogous to the fall of Spain from world empire to destitute in the 1600s because their Conquistadors-dependent society was addicted non-productively to stolen gold flowing in from the colonies.
As for decentralized ledgers being more expensive, from the standpoint of an Indie-creator, it would I think be less expensive on a total risk-cost analysis if all the server infrastructure just exists and the users pay for it as they use it, so the Indie-creator has much lower upfront capex and can experiment with trial-and-fail more often without going bankrupt. IOW we amortize the capex/risk costs over all the applications that run on the ledger. A counter point is that Indie-creators may not have sufficient resources to produce compelling games given the massive investment also in other aspects of the production such as software and media, but ecosystems arise to demand and amortize these costs over groups of creators becoming more diversified and efficient than monolithic shops.
As for the usage costs, the massive amount of unnecessary redundancy in validation is one of the aspects of costs and scaling that I’m trying to fix in my technological work. And I doubt (haven’t checked though) that server costs are even a significant fraction of the revenue that MMoRPG vendors earn so I assume even if the costs to users was 10X higher it would be an irrelevant cost.
http://degoes.net/articles/blockchain-stack#comment-3768272216
There might be things like identity services running on popular blockchains. If you need identity in your DApp, you can't just run your own blockchain.
Yes. I'd say the analogue is how today's SaaS applications often call out to 3rd party APIs—even though they are not "built" on those APIs in any sense of the word.
I expect we'll see use of 3rd-party blockchains for verification or exchange or even deployment of smart contracts.
Again John, I disagree about a proliferation of ledgers. Consistency of data and consensus is a key factor.
RE: “Consortium blockchains” (e.g. DPoS & Tendermint) can’t Internet scale