The crypto community obviously values decentralization. If you asked 100 people why decentralization matters and what implementing a decentralized protocol looks like, I think you would get 100 different answers. In this thread I'll share my take on the question.
-
-
Separate from implementation, people have specific operational expectations: - How much control the community should have at a given point in time - How much deciding power should the
@BloomToken company have - Is ok if an organization has a larger say than an individualShow this thread -
For example: - Why does the Bloom company have final say over X decision? Isn't that centralized? - What if some organizations on the
@BloomToken protocol have a large amount of BLT (and therefore influence)? Isn't that basically centralized?Show this thread -
From what I've seen, everyone has their own subset of ideological motivations, operational expectations, and implementation requirements when it comes to decentralization.
Show this thread -
The ideological reasons are strong perks, but I don't think they alone will drive mainstream adoption. We already have a lot of ideologically motivated technologies (Tor, BitTorrent, PGP, etc) and their core usage is in critical adversarial situations.
Show this thread -
Likewise, I don't think a generalized technical motivation alone will drive adoption. Many software engineers know the phrase "worse is better." Better tech doesn't win on its own. Timing, market forces, and ease of use play a large roll.
Show this thread -
I've been saying *generalized* technical perks of decentralization to refer to perks of decentralized tech in isolation. I agree that an internet built with IPFS where links don't break would be better, but that alone won't convince average users to adopt the technology.
Show this thread -
I do think that the economics behind specific decentralized protocols will drive adoption though if it saves people money. Likewise, specific technological perks that reduce costs in ways that centralized technologies can't will drive adoption.
Show this thread -
In particular, I think many Ethereum-based protocols in the long run can beat centralized businesses by winning the race to the bottom in terms of profit margins for individual transactions.
@gems competes with mechanical turk by not skimming money off of each unit of work.Show this thread -
Applying the technical perks can make compelling business use cases. For
@BloomToken, crypto + IPFS make identity+financial history persistent and verifiable no matter what business you work with. Lower costs per business + seeing past checks increases confidence for everyone.Show this thread -
Operationally, I think it is important to have an incentivized business building out the protocol and making core decisions especially for the first few years. It is hard to gain enough context about an early stage company to decide what is best in the long term for adoption.
Show this thread -
For
@BloomToken in the near term, we ask the community for input on decisions and spend hours a day (between different team members) discussing decisions. A strong core team needs to be making key operational and protocol decisions for the first few years though.Show this thread -
Ideally, there is a strong core team that is working towards getting the protocol to a state where it is valuable and usable, even if the central company dies. This requires long term vision and a full time effort. Well intentioned outsiders will miss important context.
Show this thread -
With that said, the protocol being built should be improvable without the core company. With
@BloomToken, we want to create an economic incentive to improve the Bloom score (the credit score within our lending infrastructure). We'll drive improvements until it hits critical mass.Show this thread -
In terms of what matters software wise when implementing a decentralized protocol, I try to answer this question: could someone else recreate our app (in terms of what the user sees on the website) even if we shut down our servers?
Show this thread -
This doesn't meant that it should be easy to recreate. For example, only the minimum amount of logic and data should be tracked in your smart contracts. IPFS should hold the data that you want to guarantee wont change.
Show this thread -
For example, our user invite system for
@BloomToken tries to be easy to use while also fundamentally using the smart contract. For the user, they just put in their friend's email and submit an Ethereum tx. Behind the scenes, a one time shared secret is used in the contractShow this thread -
For our voting system, anyone can vote but we filter for Bloom users off chain. Votes are weighted by amount of BLT, but that weighting happens off chain. Determining the current weighted vote totals per block requires a lot of syncing work. You can recreate it though.
Show this thread
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.