Decentralisation In Blockchains - Let's Talk About It
These talks are about layer one only for web 3 applications to be built on top.
Blockchain solves an important requirement for decentralization, which is the ability to incentive "miners" or infrastructure providers to secure the network in a trustless & autonomous way. Someone somewhere needs to store your data on a server, and that server & operational cost to keep infrastructure up and running isn't free.
Without a way to properly incentivise open and collaborative distribution of the network's infrastructure, it becomes centralised.
The only ones who'd be willing to store your data for "free" would be because they would find a way to profit from your data. If very few stores your data, you rely on them not to censor or even delete your account.
While people can use self-host, and we encourage as many as possible to do so using the 3Speak Desktop App we also realize in practice that many cannot self-host.
In some places, self-hosting can become a liability, especially if you're hosting controversial content in your country. Aside from the fact that relying only on self-hosting means you, as in your hardware, internet, & surroundings, become a single point of failure, and if any one of them goes down, so does your content. In addition, increasingly, more people are using mobile devices, especially in developing countries where desktop computers are just not applicable at scale. Moreso, many cannot self-host due to resources or using mobile devices.
Incentivised, distributed technology encourages decentralisation to form.
Decentralisation
Decentralisation is akin to a scarce, natural resource; this means it cannot be manufactured; it comes of its own. Therefore, one can only encourage the atmosphere in which decentralisation can grow, similar to creating the right ecosystem for a living organism to grow.
You can create the best technology possible that promotes decentralisation. Still, it only matters the community ethos and willingness to join that community voluntarily.
Communities are raw, spontaneous, free-willed, and naturally against being bought out or artificial circumstances. Therefore, the most talented developers, amazing artists, & overall independent people flock to authenticity.
Why is decentralisation so hard to manifest for coins?
To create an atmosphere in which decentralisation can form & sustain, it takes a benevolent founder or circumstance, IE an alien invasion-like scenario where the whole "society" bands together; therefore, everyone is encouraged to join forces to survive.
For the first circumstance to work, the founder must find a way to remain anonymous. Regardless of that figurehead's stake, having a figurehead provides a potential centralising force whereby the majority have a "obedience to authority" and blindly listen to the founder. No person is perfect, and any upgrades to the protocol thereafter should be on equal merit. Decentralisation is to relinquish all power as much as possible.
The founder(s) must not seek any credit or financial gain from their invention. This means you're asking an entity to build something that cost them time, energy, & money to then give it away for free without any strings attached. Moreover, the conditions in which the technology is given away must be such that the coin is not highly desirable at birth.
The coin distribution must be bottom-up, community first, which is the opposite of the business class and general investing culture practice.
Suppose the coin is highly desirable at birth; even if the founder is anonymous and relinquishes all power, venture capitalists will easily accumulate the supply, thus creating a centralising coin distribution.
At birth, coins are usually at their cheapest or near their cheapest in perceived value; however, as time has gone by, we have seen that to be increasingly the opposite. Coins at birth have started to come out at extremely high evaluations. Large venture funds are hungrily awaiting to buy up the supply, whereas smaller participants cannot compete. So, even if a benevolent entity were to arise and gift us with amazing technology for free, there would still be a need for the right circumstances that are outside anyone's control.
If decentralisation cannot be forced into reality, the next best option would be to join what decentralisation exists today in web 3.
Even centralised coin distributions with known founders with large premines can have a community that rallies around the benefit that technology brings. While centralised due to coin distribution, communities can "break away" due to the nature of open-source software; communities can easily "fork" and start anew.
Community forks are encouraged to manifest when a outside force attacks, aka "money attack," where the attacker gains enough resources to brute force rule the network or a hostile founder who retained power over the network via a premine decides to harm the network. This is rare because it's counterintuitive to want to destroy your own investment by attacking it; compound this with the need for a community that is strong & united enough to break away from centralising rule & sustain itself without any centralised source of funding. If the founder is liked and acting benevolently, the community under the founder is generally okay living in centralised conditions due to trust in the founder.
In the event of a community fork of open-source software, the participants in the attack on the network can have their coins excluded on the new blockchain, leaving, in theory, a balanced coin distribution whereby no one entity has a controlling stake.
Web 3 Needs
Organization
Since reward incentives are needed to encourage a foundation of decentralisation to occur, the most important aspect of any network is how well distributed the coin is.
Key aspects of a coin's "birth" are needed to ensure decentralisation is encouraged.
- Anonymous founder or better yet, no founder (community created)
- No premine or token sale of any kind.
- Fair & open access to all for a chance to earn the coin from start to finish of the coin's distribution.
- Coins need to be distributed widely across many countries and cultures.
The key part to note here is, you can't buy a community—quite the opposite. The coin should be seen as "valueless" for as long as possible; this ensures it's not gobbled up by opportunist venture capitalists early on. The coin should also have a clever distribution method, one that encourages distribution far & wide.
The best time to ever launch a crypto network is first or early on. Early crypto coins can circulate without much value or none, go relatively undetected for vast amounts of time, and create a network of early enthusiasts who like the technology over it being seen as some investment. Going undetected also allows the benefits of a strong community to form, any bugs can be fixed, and the governance mechanisms can mature without strong attacks from opposing forces.
Unfortunately, we have reached a time where even if one were to launch a coin in a "stealth" way, if it were a great project, it would quickly be discovered. Venture capitalists suck up the supply, or it would be easily squashed by red tape regulations, attacks & web 2 hostile takeover attempts.
The idea for a strong network effect is to have a way to get the coin in as many hands as possible, and those hands can't be there solely as an easy way to make money.
I believe the network effects that will carry web 3 have already been formed.
I do not believe a new coin will come around and generate a better community than what is already here today. And the communities already here are deeply embedded into the foundations they have helped shape over the years.
Web 3 is still a tiny tree, which will grow over time as we nurture it. Planting more trees does not make anyone tree grow quicker; it takes patience.
The best way to get a network effect is to join one that already exists today and help grow it.
Scalability
This includes performance & decentralization, therefore requires a nice balance between the two.
The same principles of PoW vs. Stake-based networks apply, as in money attacks, are feasible on both. In PoW networks, the more hardware & energy you use on the network, the more control over the network you gain. The only difference in PoW vs. Stake-based systems is if a PoW network were ever taken over, the community would not be able to “fork out” the attackers in a meaningful way. The attackers can deploy the same attack on the community again. Since the community does not benefit from hardware sales outside the system, each successful attack leaves the community weaker than the last.
Whereby on Stake based systems, the underlining asset must be purchased in a money attack which then enriches the community of coin holders being attacked. The successful attack on a Stake-based network leaves the community stronger than the last. The PoW network (PoW competes vs. other PoW, therefore more likely to consolidate into one) that reaches critical mass would be the hardest to take over but cannot pivot as easily if ever corrupted. However, a stake-based system may, to some degree, be more venerable to a brute force money attack vs. a PoW network; it can more easily pivot and survive in the face of overwhelming sustained hostile takeover attempts of the network.
Therefore, I will focus primarily on web 3 governance models that can scale to handle applications to be built atop in this talk.
Node Count
When a country or company is formed, you don't have 100,000 presidents that all have to agree before they could act. It's because having fewer people in charge is much more efficient. To achieve scalability here, we would need to have 100,000 people vote on one person they believe can do the job the best.
Therefore, it would be more efficient to have stakeholders vote on those they trust the most to secure the network. However, while having everyone vote on one person to run the network would be most efficient, it would also be the most vulnerable. So, the network should decide on at least a dozen or more nodes that reach "consensus" together.
For example, you can have 20 nodes of the network and 100's of backups ready if one of the 20 goes down, all voted on by potentially every stakeholder.
When talking about staked-based systems, node count becomes much less important, if not irrelevant to an extent. For example, if someone needs 51% + of all voting coins to take over the network, then both elected and non-elected systems have the same point of failure. Moreover, if both elected and non-elected systems have a balanced coin distribution in which no single entity has a controlling share, both are arguably close in censorship resistance.
The primary way to takeover a stake-based network is to have a controlling stake. There is virtually no circumstance where 20 elected nodes needing 2/3rds majority to pass new changes on the network could conspire to take over the network because they are voted in by many different stakeholders. There are many real-world examples of nodes with a smaller stake being elected in consensus by larger stakeholders. Therefore, assuming a correct balanced coin distribution, if any attempt by elected nodes to collude or be coerced into harming the network, both stakeholders and backup nodes (naturally incentivised into holding consensus to account) would instantly vote out corrupt nodes.
However, on the contrary, in a unelected system, many nodes can collude behind the scenes and cannot be voted off. This means harmful nodes can lurk in "shadows" and strike more unpredictably and can't be held to as much of an account. Compared to an elected node who must earn the reputation of being a good node to be voted in, which is hard to earn but easy to lose.
Contrary to unelected systems where nodes can pop up like whack a mole and are essentially anonymous to reputation. Elected systems are anonymous, meaning no ID is needed to run the network, yet reputation can be built using an anonymous persona. The ability to gain reputation yet remain anonymous means you can be held to account yet, don't become liable to the network via a known point of failure. While "doxed" nodes have an edge to gain a better reputation and become overall more trustworthy, it does come with more risk—freedom of choice.
Moreso, the ability to coordinate, organize and "move as one" is essentially impossible on unelected networks. Elected networks force strangers and friends to collaborate and work together much quicker, and speed is paramount in any hostile takeover event.
Avoid mob rule (1 person one vote); stake-based votes can be used. To avoid the rich, get richer, the stakeholders vote on network nodes instead of just sitting back and collecting the network fees themselves. While stakeholders can vote for themselves, they will require more outside votes to become consensus nodes; thus, they will need to show the other stakeholders that they are trustworthy and useful/consistent.
Elected node systems provide the foundation for scalability.
Social Layer
The most important layer to sustain decentralisation is layer zero. If layer zero is made of humans, we need to make sure communication is a top priority. Having "in-house" communication, where one can post permissionless any time of day without being censored, is especially important.
Web "3"
I believe any web 3 community needs three things built into the scalable technology.
- Account(s) Ownership - This includes all content, supporter, community list, and coin balance. All of this MUST be decentralized and secured.
- Decentralized Exchange (Dex) - This includes the ability to create liquidity pools and Dex "skins," which tie into an unbiased technology in which the dex skin can set the fees. 100% customizable dex gateways for each community are tied into a larger dex that is stake-based controlled.
- Social Layer - This includes the ability to create breakaway communities. The ability to communicate with all stakeholders in a censorship-resistant, permissionless way 24/7, 365, rain sleet, or hellstorm is critical to the survivability of any ecosystem.
You need to own your account, your exchange, and the ways you communicate. This means that the community is 100% self-sufficient and relies on no outside centralised technology. If your tech stack is missing any of the above, you've left yourself a major attack vector that will end up being a major liability in the future.
For instance, without having onchain communication, we would be vulnerable to being silenced, siloed, and overall fragmented. There is no way you can organize a emergency fork on an international 24/7 network if you're not completely organized. DPOS comprises stakeholders; therefore, every stakeholder needs access to what is being communicated by the witnesses and other stakeholders. Communication blackouts for even a few days could be catastrophic to a network under attack.
Moreover, you're unable to transfer value from one person to another in a trustless way; you severely handicap yourself. When crypto assets become more prominent, more widely accepted, there will be a day you won't need to sell your crypto. You could potentially live 100% inside of web 3, earning interest on your stable coins, the ability to spend on goods & services, all without relying on centralised payment processors.
To be continued...
Amazing post.
What happens when the rich are the ones running the nodes? e.g. blocktrades, gtg, themarkymark, therealwolf etc...
Not criticizing anyone, I love them all, but we have to realize that witnesses on hive have massive stake, and seem to be only ones able to earn from the DAO easily, as they have a good track record.
Once the price of hive goes higher so that witness rewards are large enough, we should have a rule banning witnesses from submitting proposals to the DAO. This will help avoid the rich get richer problem.
Although anonymous witnesses could always create two accounts and game the system.
If an attack on the network is successful and we need to organize to hardfork, the attacker can easily shut down communication and censor users, as he can hardfork every 2 minutes with dummy witnesses.
So even though hive is censorship resistant, it won't be when successfully attacked. Am I missing something here?
Not to point anyone witness out, but since you mentioned Him, therealwolf has powered down most of their stake and is not considered a large stakeholder anymore, yet is still in consensus. I'll name one more (there are several) - Luke Stokes, who was a consensus witness for years and was not a whale.
Sure, you'll have whales that become consensus witnesses, which comes hand in hand with being a hardcore hiver and the means to back up what you believe. Large staked users have the most to lose as well; thus is more highly desirable to be in consensus due to trust in one's self interest to be upheld. The diff between elected vs. non-elected systems is that non-elected whales can sit back and soak up supply, where I can name several large whales who are not in consensus and never have been (myself included). Thus we have not been sitting back collected inflation due to solely our stake. In elected systems, in theory, the ones getting inflation are those you want to get it. In elected systems, you're only as strong as your network of "good" voters; we need as many good Hivers with as much HP as possible, evenly spread, of course.
While nothing is perfect, elected systems do run into the issue of set it and forget it, meaning you find a group of great nodes; nature dictates that you're either getting better or worse. Thus these nodes overall have gotten better in terms of more stake, more time on their reputations, and a better understanding of the network. So in a sense, the most reputable in elected systems get richer, and to offset this, we have PoB inflation distribution, which I believe has been the backbone of our network effect.
On Hive, for instance, we can see a money attack coming with the 1-month powerup. We hard forked rather quickly vs. Steem, and that involved a lot more changes due to wanting to get Steem out of the code name. If we were to fork again, we would likely just keep tAhe name Hive (sue us, lol) and leave out the attacker's balance. So as you saw, when Justin finally got control over the network, by that time, Hive was up, and we could communicate over there. There is no way to "instantly" take over Hive, and we would have plenty of time to set up shop elsewhere minus the attacker's coins. A
Very good point forgot about that one. Although someone who really wants to attack hive could create many accounts and power up less 1 million hive in each account. No one could see the attack coming if he powers up at a random frequency in many accounts during months.
Then he could suddenly censor key community leaders from the chain and communications would still disappear on chain. The good thing is that it's impossible to block communication between a community offline, it will never be too hard to talk outside of hive.
We could hardfork even without on chain communication imo, or with limited on chain communication. So that's not such a big problem.
I completely agree with that and I welcome it. Although it means that it is possible we fall in the rich get richer problem, if those people start making bad decisions. They have a good track record for now and definitely don't seem greedy.
They put hive before their own interests, or at least it appears that way to me.
The ultimate solution to the rich get richer is forking whales out, or forking to a new blockchain where concerning whales get only half the coins they did before. Forking solves everything.
Hopefully the threat alone will always keep whales in check.
I don't see a point tbh. If the proposal directly benefits the community, and if it demands a lot of work I wouldn't exclude to vote for compensation.
As you said, in this case anyone can game the system.
I like to think that witnesses have to maintain their integrity, whereby demanding a compensation for 6 lines of code would be their political suicide :)
And being top witness when hive hits 20$ will be quite lucrative deal.
Can't we just run the old code? I doubt anyone could censor everything before the community reacts. That's the main strength of DPoS and Hive.
Yeah, maybe there's no point in that. However, I would definitely prefer voting for witnesses that use the witness income not just to run nodes, but also contribute to hive in other ways. Obviously at this price it's not possible.
No, because an attacker with majority stake will just vote 20 dummy witnesses and run whatever codes he wants to run.
We'll have to hard fork without being able to use hive (if the attacker censors people on hive).
That's not a huge problem though, people can still talk over many places on the internet and reorganize.
To me this is a well-thought-out, comprehensive, (reasonably) understandable look at how to create grow a robust, prosperous, scalable decentralized system. :)
There may be some nuances that I'd like to dig into someday, or some philosophical points that may ruffle a few people's feathers, but overall, this was a fantastic read.
Impressive thoughts and writing here, @theycallmedan , I appreciate you sharing, and look forward to part 2. Wishing you a great day. 🙏
@cynshineonline I'm not sure if you'll grasp all of this easily, but it's worth a read and I can help explain if unclear.
thank You so much for the tag. You were correct lol it wasn't an easy thing to grasp but You explained it quite well and I feel like I get it now. I agree that it was impressive writing.
Thank you for posting it @theycallmedan it's mostly understandable and what I didn't get Jay helped me with. ❤️
Props to you for tackling the read, even as a layperson! 😁🙏
🙏❤️
I think you are right about this.
True decentralization is really hard to achieve (anonymous founder or community rebellion), but once achieved is very resiliant.
Hive may well be the seedling that will grow into the entire tree of Web 3.0.
@podping is certainly proving that Hive can generate enormous value as a decentralised central global notification / signaling system with very little resource demands on Hive.
That's a perfect explanation of why PoS has benefits over PoW. Like you, I also do believe in the need for a strong community around a blockchain. A community that is able to communicate with each other. And also I do believe that any crypto project that wants to survive the length of time, benefits from creating the community first whilst keeping the speculators away from buying the token. That said, in our HIVE ecosystem we may have a 'large' community, however also see two things that may be far from optimal: 1) only a small part of the users are active users and 2) distribution of power is not that large looking at the distribution of 60 to 80% of the HP. I somehow like projects such as 'Pi', which gives all miners a truly equal playing field (tap or finger mining as I call it). With almost 20 Million users mining, we can definitely talk about massive decentralisation.
With the Speak network you guys are planning, you plan to distribute tokens to the community. That in itself is wonderful. However, the (stake) distribution of the tokens will be relatively small, not better than HIVE itself. What is your view to further distribute stake in the Speak network?
ps Since last week 3Speak is becoming almost impossible to access. The Desktop App is not loading any content, and the online version is coming up with nothing other than an error page. Are you guys planning to post something in eg the @threespeak channel to inform us what is happening and when you plan to solve the issues?
-Speak network distribution.
The initial start will be the 1 for 1 drop to Hivers, so the initial supply will be the same as Hives at the time of the snapshot. From there, Hivers can power up those miners and mine for SPK. SPK's token system will live on Hive (one day), and thus you cannot "Justin Sun" attack the network by purchasing SPK. You'd need to attack Hive itself to take over the token system. In any protocol, giving the ability to self-host is the nuclear option for content. However, with a token system, you 100% need a distributed system with active participants. Therefore the SPK tokens will be as safe as Hive one day, in theory. Of course, this will require a layer 1 change to Hive, something we can only build the take for, and then try and convince the community why it's ideal to vote on such an upgrade. But I'm pretty sure most around here want SMT's :)
SMT's has been talked about for a very very very ling time, but still no progress in that direction as far as I know. Others want Smart Contract support instead of SMT's. Maybe we need both, I don't know. What I do know is that we need to progress with either or, or both. HIVE is falling behind the market. One of the reasons why the last original HIVE (Steem) users from back in 2016 when all started, are powering down and leaving. I know of one of such users, a mini-whale, he lost confidence. This guy is in crypto for the quality projects and was hoping with HIVE we could become better than what we were before. Recently he told me: I lost my last bits of hope and started to power down. Such a shame!
EDIT: I just received the link to the packaged version (https://github.com/3speaknetwork/3Speak-app/releases). Maybe the link at the bottom of the spknetwork website may be changed to this one? Still not seeing any videos in the app though. IPFS is having more than 500 peers and some considerable traffic out is going. I'll be checking further with the tech team at Discord to try and solve this.
The desktop app isn't loading any contnet? That must mean those creators did not post the content to IPFS. We plan to bridge the web app with the desktop app more and encourage users to use all the features more. Right now the web app is experiencing an issue with anti-virus and browsers, the site is fine its just a certificate. Right now if you want to upload, you can using the desktop app. The web app should be resolved soon
The Desktop App worked before, when release 11 was out. This has been replaced by a new release. Unfortunately no install file for Mac is provided. Last 4 hours I tried to manually package the release 12. Had to install Xcode (for N-1 MacOS release this is not straightforward), then Homebrew... and then the commands as per 3Speak readme at Github resulting in errors with one of the packages: 'electron'.
I wonder why for Mac users an install package is not provided for? Since the .tv service is not working, Mac users don't seem to be able to connect to the 3Speak service anymore, not online, not through Desktop App. Its not for me to upload video's, but for my daily curation in the music section of HIVE in which quite a few users use 3Speak. The unfortunate effect is: I can't curate any of the 3Speak posts anymore. Very unfortunate for those who use the service last days.
I do realise devs need to do things to progress the service. But what I find truly annoying is the fact a service that HIVE users do really on, is not available anymore. Added to that, one need to be a true dev to make it work. This is harming the use of this service, and at the same time it is harming HIVE. Any none-HIVE user trying to open a 3Speak video when browsing HIVE posts, will potentially be a lost user for 3Speak and for HIVE.
Mind-blowing article @theycallmedan!
'Social Layer' is the reason why I am here! Communication is the most important aspect of any community to thrive. It's the reason why we are able to teach so many people about so many things, especially Blockchain & Crypto. The best way to have an open dialogue in a decentralized manner.
A little dense but great philosophical points as they relate to decentralization and governance.
It's crazy how 99% of the people here don't realize how lucky they are to be here at the ground floor.
Grass is always greener on the other side, I suppose.
I am going to hypothesize you are going somewhere with this. 😁
This is a new way of thinking that ppl in the decentralised world need to adjust to. It also defends against ur token being listed as a security and helps influential investors ensure their projects are collaborative, voluntary and have the beat chances of organically decentralising in such a way that benefits the token in the long run, instead of in the pump and dump short run. Dan probably could issue a check list in the bnext version to help anyone trying to set up a fair, distributed, decentralised project / system to have the best chance of achieving that with the above principles in its foundation. A project following these foundational principles is set up for success in the long run
I think in a healthy market people always take profits. That's also a way how token disrepute in new hands. I would expect with the first big FULL TIME influencer, that comes from web2 to web3, we will see a shift.
Well explained
Awesome post.
Man I wish there was some sort of Discord with a Web3 HIVE backend.
in time
Heck yea I !LUV hearing that.
Great and direct and easy to undertand, thank you for this information, I learned a lot
Great content ❤. Learning new things everyday.
@starkerz this seems fitting considering our call today... :)
Yes. I merely skimmed the surface of this. Trying to bring u in slowly. But the info in this piece is super great even for old school blockchain geeks! Dan really pushing the philosophy (and movement) forward with content like this
You gave me a tons of great info, help, and understandings, and you were considerate to not overwhelm me, I appreciate it. Without your explanations, I wouldn't have understood a word of this wonderful post :)
Decentralization is the next big thing, it is a great way to manage and boost the economy. Unfortunately, it is not readily accepted in most parts of the world right now. But when it does, its bound to bring about a revolution just like the internet did.
Congratulations @theycallmedan! You have completed the following achievement on the Hive blockchain and have been rewarded with new badge(s) :
Your next payout target is 408000 HP.
The unit is Hive Power equivalent because your rewards can be split into HP and HBD
You can view your badges on your board and compare yourself to others in the Ranking
If you no longer want to receive notifications, reply to this comment with the word
STOP~~~ embed:1416059258869846016 twitter metadata:ZmVkYWFzaGlraGF8fGh0dHBzOi8vdHdpdHRlci5jb20vZmVkYWFzaGlraGEvc3RhdHVzLzE0MTYwNTkyNTg4Njk4NDYwMTZ8 ~~~
The rewards earned on this comment will go directly to the person sharing the post on Twitter as long as they are registered with @poshtoken. Sign up at https://hiveposh.com.