Centralization vs. Decentralization
avc.com
avc.com
The first step in writing about decentralization is to understand what decentralizaiton means. Premined coins are centralized by desgin--the founders may hold an undisclosed amount of coins that were acquired at no cost. Thus, any fork that threatens the founders can be attacked crushing the incentives of the miners, and thus the security of the network.
Linking to websites such as https://arewedecentralizedyet.com/ is both lazy and meaningless because those metrics are easily gamed, insecure, vulnerable to sybil attacks, and unreliable. Most of the decentralization discussion is a red herring that many use either out of ignorance or in order to distract from fundamental problems such as the distribution of incentives.
https://news.earn.com/quantifying-decentralization-e39db233c...
"The basic idea is to:
(a) enumerate the essential subsystems of a decentralized system,
(b) determine how many entities one would need to be compromised to control each subsystem, and
(c) then use the minimum of these as a measure of the effective decentralization of the system.
The higher the value of this minimum Nakamoto coefficient, the more decentralized the system is."
If the security of your network is fundamentally based on trust that one central party was honest, then you can't have decentralisation regardless of how many addresses you have, how many miners you have, how many nodes you have, or how many clients you have.
This seems like a deep thought, but it isn't. All this is, a shot fired in coin v coin wars (and this one against Ethereum).
Bitcoin used to be easily minable from a CPU, and was under a dollar for a large point. Any entity could be holding undisclosed amount of coins that were acquired at no cost either.
Bitcoin's mining cost was always a function of hash power, hardware and electricity cost (and competition). If it was cheap, it was because you were early. Satoshi mined from day 1, which is how he amassed coins.
Some forks however played in a developed crypto space and wanted to guarantee themselves a chunk of coins that would not require them to openly compete for them.
Does the difference matter? I dunno.
Satoshi understood the importance of fair distribution, while all ICOs compromised their security in their futile greedy attempts to create "a better Bitcoin." You can't expect to improve a system for which you don't even understand the design rationale.
The fact that he ends the post claiming their fund educates the population about decentralization only reassures me that we're unlikely to have such thing anytime soon.
Though, there's the risk of "well, what if they just also purchase an equal amount of the other side of the bet", but perhaps this way they could still have to pay the cost of their money being locked up until the event resolves, which would at least create some sort of cost of falsely signaling that their belief and/or level of confidence?
What would qualify as skin in the game? Refinancing their home to buy bitcoins? That would be something, for sure.
Also, simply purchasing the coins seem to me to say less than actions in a prediction market could. If someone claims to believe that the price will be above a particular quantity at a particular point in time, buying some of the good doesn't really reflect that exact belief, only that they believe that the price will go up on some timescale. A bet that the price will be above a particularly quantity at a particular time more clearly indicates that they believe that it will be above that quantity at that time, I think.
The history of computing is the cycle between centralization of compute (Mainframe, Cloud) and decentralization (Desktop, Edge) from the users perspective. It's been going on forever and I don't see it stopping because of the computing cost cycles and increasing network speeds.
We're going to push more compute to mobile devices, then back to mesh networks etc...
But to me it seems more likely the flipping has been driven by outside forces. That one approach was genuinely better for a time and then in a subsequent period the opposite was genuinely better.
If you look at the primary platform where velocity is happening in the crypto space (number of transactions or touch points) - coinbase - it's a centralized broker for crypto.
Individuals aren't mining their own coins, they are buying them from more and more centralized miners.
Interestingly there haven't been as many of these monetary cycles in history since the emergence of major centralized banks, which is a good thing. So there is a (strong) argument that currency centralization is a driving force in stability and peace.
Countries that have been at peace a long time, and which have whatever institutions and habits encourage such peace are also ones that can grow big multi-national corporations including banks.
You need a new business model to go along with the new tech. What is that for crypto?
- For Microsoft, it was selling software instead of hardware like IBM. "Open source" hardware complements software. This was hugely profitable.
- For Google, it was selling user attention instead of software like Microsoft. Open source software complements web services. This was hugely profitable.
This analogy breaks down here a bit because there is a strong incentive for "open source" hardware (money for dozens of vendors), but not as much for open source software. It's more of a byproduct of the SaaS movement. I guess you could say the real corporate motivation there is developer attention, e.g. the purchase of Github. So the analogy falls down a bit.
- So now he claims we will have open source data. I would love this, but why? What's the new business model? Companies USED to provide APIs in the 2008 era when they thought it would benefit then, but they've now pulled back -- Twitter, Facebook, Google, etc. Google APIs are in terrible shape.
He said someone proposed a regulation that Twitter be forced to write a better API? Huh?
That would be analogous to the government forcing Apple to open up its hardware like the IBM PC, or the government forcing Google to open source its search algorithms. I see why it sounds nice, but I don't think it will happen.
This is an honest question, not trying to pour water all over this ... what's the new business model, or is there at least a candidate?
I love Steemit's idea (not necessarily implementation) of a decentralized social media platform. All the data is public so now there's a decoupling of the network problem and interface problem. Facebook for example is impossible to compete against due to network effects. If you have a better version of facebook, it's very hard to get people using it because all of their friends are already on Facebook. Steemit decouples the problem by enabling people to create new interfaces to it and everyone with a steemit account can already immediately use it because their account, followers, and the rest is all the same data. Someone connected IPFS to Steemit and created DTube which basically the same thing as Youtube.
Why use a decentralized social media platform? The two biggest reasons are that decoupling the data from the interface creates better and new interfaces and content creators are rewarded with money. Right now, social media companies like Reddit get money for the content you create. With Steemit, the rewards go to those who create content and those who run the network. It seems like a more fairer system. Steemit does have problems though but the idea I think is some version of what we can expect to see more of in the future.
In general I don't think it is possible to create new interfaces to social media sites without the owners consent. It is too easy for them to break you on purpose, and that has happened many times in the past.
Maybe I'm misunderstanding how it works.
Decentralization also seems orthogonal to this issue. If someone could gain users by creating a new UI on top of Facebook, it would have already been done in a centralized fashion.
The thing that makes Steemit unique is that to create your own version of Steemit, you just build an interface on the blockchain. You and every other developer has unlimited access to Steemit users, followers and other data. This allowed projects like DTube (a decentralized Youtube) to immediately have millions of users. If the Steemit website starts placing ads everywhere for example, someone can create a different better interface. The point of this decentralization is exactly what the author mentions. You decouple user data from the interface enabling companies to compete on interfaces instead of network effects.
These businesses profit by being the middlemen in transactions between two users. Bitcoin, for example, competes with traditional banking by cutting out the intermediaries, and their margins.
Ledgers are the simplest application. Next up will be platforms. They often have rather large margins (~20-40%) that are very attractive targets for competition.
This time, instead of one megacandidate, we may have many that are enabled by switching from platforms run by monopolies over to open decentralized platforms. So the new business model is basically any business model that wants the large network from that decentralized platform along with, and here's the business advantage, a fair playing field.
The big reason why those companies pulled back is because they realized that they don't think they will have a monopoly in the long run on being able to monetize the data better than others, every other excuse just seems to be icing on the cake because if they were confident, they open up the data they collect more.
This is something I've been working on for the last few years. I've been working on a selfhosted music server that anyone can setup in there home to stream their collection.
I've had some success building a dependency version that's deployable as an EXE for Windows (or pkg for OSX). The fact is most people aren't going to setup a separate computer just to host a decentralized server. But people are willing to use their current hardware for decentralization. And most people's current hardware is a Windows PC.
The other thing I focussed on was making it dependency free. This way you can just install it and run. Most people don't want to hunt down dependancies and write config files.
The fact is, decentralization won't take off until people can do it effortlessly. The reason centralized services rule is because it requires no effort or cost to the end user. Until that changes, decentralization will always be a niche
Centralization allows for extreme optimization, but decentralization allows a system to keep on chugging along no matter what the conditions (within reason)
This is a very insightful observation.
I guess that makes me slightly pessimistic for favoring decentralized services.
On cryptocurrencies that seems IMO the "real" topic of the article IMO they are not a "free new form of currency" but a test by some bigs of IT that think they are big enough to substitute with their infrastructure the actual banking system. Of course cryptocurrencies are generally free software but due to poor scalability of blockchain operations and it's ever-grown size and the substantial need of connection and hw to operate they are de-facto a controllable ecosystem. And actual hw and firmware's trend is direct in a really dangerous direction.
IMO the problem is not much centralized vs decentralized or distributed but a matter of "design mind", today most people are unable to think free by design solution, being born in an unfree world where web tech are so "affordable&effective" that seems to work like magic.
What is a cryptonetwork and how does it lead to open source data? Wait...lemme guess...a blockchain! I wish that people would be more precise with language. Here's a news flash: crypto already runs the world. And blockchain runs almost nothing. This wholesale substitution of "crypto" for "blockchain" is disingenuous at best and self-serving at worst. It reminds me of substituting "climate change" for "global warming" (or for "pollute less").
> "...regulators...are taking actions to slow the decentralization sector down..."
I thought that regulators were taking actions to slow the centralized sector down (through antitrust, privacy provisions, etc.). Wouldn't this provide fertile ground for the (undefined) decentralization sector. Oh wait, you mean the blockchain sector? And ICOs?
This whole blog post makes me uneasy. The inaccuracies and omissions are one thing, but the fact that VCs now have the power, wealth, and inclination to become lobbyists for failing investment theses by "...spending a lot of time with public servants of all kinds, educating them, imploring them, and desperately trying to get them to understand where we are, why it is an important moment, and why we need to this new technology to succeed..." makes me ill.
These attempts to tilt the regulatory playing field are the stuff that undoes free markets.
The computing primarily is because Moore’s law, i.e., the continuous lowering of computing hardware.
With the dropping costs of hardware, coming the proliferation of software, then at certain generic integration point, naturally some major players pop out: Instruction set Operating system Internet Web
It’s then a matter of who can be the right pint to capature the market opportunity and become the winner.
The article tries to fit this process with open sourcing metaphor. But open source has little matching properties with this process. It’s more like the software abstraction always goes higher and higher therefore the previous winners naturally loses their advantage when the world moving to build new things on a different basis.
Crypto network, how it’s connected to this process is not clear to me.
In the end, there is no competing of centralization vs. decentralization. They always coesist. All the crypto buzzword is just a slogan for financial industry to adopt open data exchange protocols.
And to equate crypto network or blockchain with decentralization is about as deceptive as calling internet an open network for information exchange.
Hear hear.
Which is not to say that any self regulation built into decentralized platforms should closely resemble anything like traditional government.
https://www.slideshare.net/mobile/StuC/ill-see-you-on-the-wr...
These are not new ideas. [User] sovereignty in controlling one’s own view and being able to voluntarily pass control of what you see to someone else are quite literally the founding ideas underlying a democracy.
At the risk of belabouring the point that these are not new, I already have what he describes in production [1], by the way of moderators you can disable and enable as needed, coupled with community mod elections. See the docs here: https://getaether.net/docs/faq/voting_and_elections/
It does make me happy to see I’m not the only person thinking about this, though.
[1] Currently disabled, pending UI work
"But the open PC hardware standard allowed Microsoft to develop an operating system that could run on any computer built to the PC hardware spec and they eventually unseated IBM, only to become a near monopoly themselves."
Microsoft didn't unseat IBM. IBM hired Microsoft to write its operating system, and Microsoft retained ownership of it, and many versions were made for many personal computers. Hooray, write it once, use it everywhere, right? But as more software began using direct calls to hardware (rather than just MS-DOS calls), that software had to be ported to new systems. Most were unable to port successfully, so everyone just started writing software for IBM PCs.
Even though there was this "open standard", the de-facto standard remained "IBM". All personal computer manufacturers were forced within a few years to abandon any proprietary product and sell only IBM PC-compatible machines. It took nearly a decade for the clone makers to reduce the price enough to make it affordable to buying something not IBM. Eventually IBM lost the lions' share of the personal computer market, and went back to its original near-monopoly of mainframes and minicomputers.
"But just as we were wringing our hands about what to do about Microsoft’s monopoly, an open source operating system (Linux), the internet protocols, and the free distribution of the world wide web undid that monopoly and we got Google, Facebook, Amazon, and other big tech platforms."
Microsoft's monopoly wasn't unseated, and the rest sounds like nonsense. There were a variety of operating systems and software running Internet services back then. Linux and the Apache web server unseated them from web hosting and commodity server needs, but Microsoft maintained its consumer-facing monopoly for... ever, and its business-facing network software. I don't know what "the internet protocols" has to do with M$... they tried to market competing protocols, they failed. We could extend this comment into the browser wars, but that's a whole 'nother rant (and the one monopoly they've really given up). As for "the free distribution of the www" ? What does this even mean? The web has always been paid for by somebody, I'm not sure who ever thought it was free. Microsoft was never a threat to the distribution model of the web - if anything, AOL was that threat.
Microsoft's tech platform is still here. Amazon, Google, and Facebook were not created after someone "unseated a monopoly". The web was just there, and these companies built businesses on top of it. Amazon and Google were even created whilst Microsoft was at the height of its power (1995 and 1998).
"And now we are wringing our hands about these near monopolies and their market power and the ability of bad actors to manipulate them. And around the same time, the technology to architect and scale a completely decentralized system emerges."
This is actually an interesting point, though completely by accident. It seems like new markets sometimes spring up while we're distracted looking at some other market. We're distracted right now with Facebook, Google, Amazon, etc. But probably some new business model will emerge that has nothing to do with these, and they'll become less important as we shift our attention somewhere else.
.... However, please note that we still use IBM PC-compatibles, we still use Microsoft platforms, and we will probably keep using Google Search & Maps and Amazon until the sun burns out.