I've seen loads of articles on blockchain, but usually they suck in explaining anything in a way that even my mother could understand it. So it's now time to explain the Hashgraph / blockchain problem and solution in a really simple way that everyone should get. I won't get into the complex bits and math because that's not the goal here, the goal is to just explain the basics as well as to provide you some great resources if you want to understand Hashgraph in more detail.
Below is an excellent video by Mike Maloney on the topic that I highly encourage you watch, but I will also summarize most of what Mike gives in the video as well as additional insights that I think he didn't emphasize enough.
The basics for newbies that might not already understand this stuff (ignore if you aren't a newbie)
At the heart of it, a cryptocurrency is nothing more than a shared ledger of transactions (called a Distributed Ledger Technology or a DLT) with a unit of value (usually called a "coin" or "token") that it is transacted via shared recording and verification of transactions on the ledger. At it's essence, every cryptocurrency is simply a ledger of transactions in units of value called "coins" or "tokens". That's all it is at its core.
A blockchain is a specific type of algorithm / data structure utilized to manage a shared ledger. Just about any algorithm / data structure could be used to manage a shared ledger, but when you have millions or billions of machines accessing a ledger, it becomes a really difficult problem to solve. Blockchain was the first solution devised to solve this problem and was implemented in Bitcoin.
The most difficult problem is how to make sure you can ensure reliability of entries with people you don't know or trust. So if you have millions of computers and some are malicious, how can you ensure they aren't trying to deceive the community and add wrong entries in the ledger to steal from or sabotage the ledger? The solution should provide something known as "Asynchronous Byzantine Fault Tolerance." There's an explanation of this is in the video above by Mike Maloney as well as in the Nasdaq article here on Byzantine Fault Tolerance.
Bitcoin developed the idea of "mining", where miners are responsible to solve the problem of coming to consensus on transactions and thereby safeguarding the ledger. It is done in a way that ensures that the network is probabilistically correct through solving complex hashes and through multiple confirmations of transactions. It is an insanely expensive computational exercise that is totally unnecessary based upon newer algorithms / data structures.
Mining always drives the incentive to create larger and bigger pools that could then manipulate the blockchain. The Bitcoin blockchain can be "taken over" if more than 50% of the miners are all owned by the same entity. They could they swap out the blockchain with a different version of the blockchain that could allow the theft of all Bitcoins. (There is also a 34% attack that can happen with blockchains as well, but we'll come to that later.)
Some blockchains (like Bitcoin) even have been designed to go slow - Huh? Why would they do this? This is because as the chain of blocks is being built, they can't have the competing chains grow faster than they can verify the transactions, otherwise they would lose control of the growth of the chains and never catch up. The incredibly complex and expensive mining process slows the blockchain down enough so that there is time to reach consensus before chains are built further. Talk about a worthless waste of time and energy. (See Leemon Baird's Harvard presentation below for more on this.)
A "wallet" is thoroughly misnamed. It is an application to hold your public and private keys which allows you to read as well as transact on the shared ledger. No coins or tokens are ever contained in your "wallet"
Enter Hashgraph and Directed Acyclic Graph (DAG) algorithms / data structures
Again, I won't go into all the details of Hashgraph and DAGs, but will leave you resources to understand them in detail such as the three videos above. Below are some of the key points that makes Hashgraph unique as a truly inexpensive, secure, and fully peer-to-peer technology:
Hashgraph is both an algorithm and data structure that doesn't use blocks or wait for blocks to be confirmed. There are no blocks of transactions, but instead streams of events that can be segregated into clear transactions. Since there is no waiting for blocks, Hashgraph can do hundreds of thousands of transactions per second. In comparison, Bitcoin can do 7 transactions per second and Ethereum 15 transactions per second, but even these rates are at their best performance and not their worst.
Hashgraph requires no miners and is computationally very inexpensive. Hashgraph uses something called the "gossip protocol." The gossip protocol is extremely efficient in sharing information between computers since basically it's just what it says it is. Each machine tells each other machine it can "see" all it knows about the specific transactional events they are concerned with. This is in turn used to come up with a consensus image of all transactions. It is so efficient that even a smartphone can easily participate in updating the hashgraph. No insanely expensive mining rigs are required. No electricity to power a small country is required either.
Hashgraph consensus is mathematically guaranteed vs. Bitcoin which is only probabilistic. Hashgraph is unique in that it uses a protocol called "Gossip about Gossip" where it adds and sends additional information about the "gossip" history to each participant. This provides enough information, along with something called "virtual voting", so that assured consensus (Asynchronous Byzantine Fault Tolerance) can be reached by mathematical proof within seconds. Bitcoin can never achieve this and therefore requires numerous confirmations to get closer and closer to being sure that consensus is achieved on any transaction that has occurred. This incredibly improves speed, efficiency, security, fairness, and even stops hard forks. It basically blows away any other technology out there for consensus purposes.
Hashgraph is completely fair - So the immediate question that arises here is, "What is "fairness?" Fairness in a distributed ledger environment is where the ordering of the transactions is correct and therefore "fair". If someone could alter the order of transactions in an auction or a securities exchange, then it's not fair since someone was given preferential treatment and this was not based upon the timing of their bids and/or asks. Hashgraph can ensure that this never happens.
Hashgraph is not the same as other Directed Acyclic Graphs (DAGs). The Hashgraph protocol of using "Gossip about Gossip" is patented and unique. I've heard many people compare IOTA to Hashgraph, but it's nowhere near Hashgraph in performance. Other DAGs that exist (e.g. IOTA, Byteball, RaiBlocks, etc.) have the advantage as well that there are no miners and no use of blocks, but this is where the similarities basically stop. Hashgraph can run truly peer to peer, but the other DAGs use nodes and frankly, IOTA's performance is known by many as a complete disaster. It has been a highly questionable marketing campaign that has sold IOTA as being something similar to Hashgraph in terms of performance and security. It's simply not anywhere close.
Hashgraph is encryption agnostic - Any encryption / privacy method can be added on top of Hashgraph for a cryptocurrency solution. That can even be quantum computer resistant encryption and privacy protocols. This makes Hashgraph incredibly safe as well.
There is only one major weakness to Hashgraph and that there is something called a 34% attack. When more than one third of the nodes are controlled by a single actor, then the same style of attack as the the Bitcoin 51% attack mentioned earlier can occur on Hashgraph. What most people fail to realize (and this is pointed out by Leemon Baird in the 1st video above) is that Bitcoin can also suffer from a 34% as well as a 51% attack. Therefore Hashgraph is just as safe as any other blockchain solution out there. The advantage with Hashgraph over a blockchain like Bitcoin is that because it is computationally cheap, you don't get the same concentration risk that you do with miners (think Chinese Bitcoin miners), however, it's also possible that a state actor could add insane amounts of cheap nodes on the network to hijack the network as well. Yet, this potential problem also exists with other blockchains. The Hashgraph team is working on solutions for such issues as well, but right now they are at least as good as any blockchain out there in this regard.
The vision of Hashgraph
The best person to give the vision of Hashgraph is it's inventor, Leemon Baird. He outlines it in the video above and the vision is extremely compelling. The key points to this vision are as follows:
Using true peer-to-peer technology to disintermediate everything! - The history of the control of humanity is based upon key monopolies and oligopolies that control all the key facilities of humanity. They then can manipulate and control these to gain all the wealth and power in society. These monopolies include money creation, securities trading, ledgers of ownership, communications channels, etc. By having a completely fair and fully peer-to-peer network for these activities (which is what Hashgraph enables and blockchains can't), then no select actors can monopolize, manipulate, hijack, or control these channels. This means all the traditional intermediaries such as banks, stock exchanges, government organisations, etc. are taken out of the picture. They are no longer needed and therefore can no longer hijack the systems that humanity uses for their own interests. Blockchain still can't yet fully guarantee this in it's current technological state. Hashgraph can.
New social communities / worlds - Social interactions with likeminded people that are totally decentralized and secure will provide the basis for a radical transformation of society to totally free expression, interaction, and even new legal and business models within these communities. Hashgraph looks to enable that anyone can develop these newly at any time. Since Hashgraph is incredibly cheap to use, this becomes a real possibility. Disintermediation of systems like Facebook is the goal here.
No more Google as a search engine intermediary - Disintermediate Google with a massive distributed service directory that publishes the latest on every information and service available.
Hopefully you can agree that Hashgraph is an extremely compelling model. I very much look forward to seeing public solutions such as new cryptocurrencies based upon it in the near future.
Please do contribute you questions and comments below. I will be sure to update the above post accordingly as it is still available for editing in Steemit.
This is super awesome, and my mind is reeling with the possibilities.
This isn't actually correct. BTC currently can suffer a 34% attack, which I also did not know prior to seeing Baird's talk at the business school, but his explanation is clear, and probative.
Thanks very much!
Thanks very much and excellent point about the 34% attack issue. Thanks for calling it to my attention. I've corrected it in the above post.
What will happen to Bitcoin if Blockchain technology will be replaced by Hashgraph? Will it evolve adapting the technology or will it lose it's value and it's better to buy other coins/tokens that supports Hashgraph?
My take is that it would launch Bitcoin upwards in value for a very, very long time, but the chances of that happening are slim to none. The Bitcoin community is filled with so much opposition to any changes that it will be almost impossible to fundamentally change the core of the Bitcoin technology - and especially not to something like Hashgraph that uses no mining.
What a fantastic, detailed, but understandable discussion of these competing technologies. I am a noob, but learning and excited about this field. I appreciate your taking your time and sharing your wisdom and opinions.
Thanks a lot! I'm really glad to hear that's the post has been helpful. Thanks also for letting me know!
Thanks for sharing all of this information. Wow, it's so incredible what is happening with technology. This is amazing and it will be so interesting to watch what develops. I'd love to see Leemon and Larimer on the stage together having a discussion about Hashgraph and EOS. Can you imagine? As far as I can tell, they each purport to do the same thing, which is introducing increases in speed and security that are magnitudes greater than existing standards. Hashgraph also has this fairness and virtual voting. Wow.
The gossip about gossip protocol seems like a complete gamechanger. I feel like I have more understanding of Hashgraph than I do of EOS at this point after seeing this video, and I would love to hear how EOS could respond to the underlying faults that Leemon finds in blockchains in general.
Or, is EOS using the term "blockchain" in the manner that Leemon describes in this video, that is, as a substitute for "distributed ledger"? Is EOS also a whole new beast that really doesn't fit the traditional explanation of blockchain? What do you say, @dan?
Great stuff, what exciting times we are living in!
I think the point that Leemon is getting at is that Hashgraph is a completely different algorithm / data structure than blockchain at the very foundation of the distributed ledger applications that are being built. EOS is a distributed ledger application and is therefore far more than just a blockchain. What Leemon suggests is that if these applications like EOS are modular enough, then you could swap out the blockchain algorithm at the foundation for a Hashgraph algorithm. This is already happening with Lykke. They are implementing Hashgraph at the core of their architecture as far as I'm aware. Regarding EOS, I have some separate concerns about EOS, that you can read here.
I appreciate your prompt response and insight. Do you have a good source to read more about this Lykke-Hashgraph marriage?
Looking forward to digging into your EOS concerns also. I've been high on it, but want to stay tempered and educated on both sides of the coin.
Here is the only link that I have seen on it from Lykke. I don't know anything else at this point beyond what they've described in their short summary.
Wow that a whole days watching saved and will explore thanks for sharing !
Sure.. thanks for the positive feedback and upvote!
I'll have to reread this material and watch the videos more carefully. Thanks for putting up this information! Resteemed, for what it's worth (from a minnow).
Thanks very much - and for resteeming as well. By the way, I'm also still a minnow. Every bit makes a difference, so thanks!
We seem to be entering a time where decentralized autonomous organizations can be created by almost anyone and the ability to do that is fast becoming democratized this being so it's easy to anticipate or coming period Where there is great deal of new creative innovation especially in the software + app and bot space what skills are needed to build a DAO and how do you go about creating something like that in an open source way?
I think the technology is still quite immature for this to be honest. There is a lot more work to be done. What we are talking about is distributed databases and applications that aren't run by anyone specifically, but right now this is still difficult to do in a lot of cases. It currently only works in cases where there is certainty in all aspects of the business transaction.
Let's take insurance as an example. There is still loads of uncertainty in the risk level assessed, the various features that will be covered in the policy, the corresponding cost of the policy, the claims assessments on the level of damage after an event. You can't do this as a simple distributed ledger. This requires human interaction and some serious process management. Most businesses have end-to-end processes like these. It's only fully discrete business models that can currently be enabled by distributed ledger in a clean and straightforward way.
Yeah, I found out about this recently. It's really too bad there's no way to invest in it currently.
I'm hoping that the Hashgraph developers get their tech into some cryptos very soon. With the right team and business model, they could rock the crypto market and create amazing solutions for everyone at the same time
Yeah, definitely.
Been thinking about that. Hashgraph can be currency agnostic, just as it can be encryption agnostic.
This can enable it to become a decentralized exchange, where people can exchange cryptos between themselves in a secure and open ledger type environment. I don't NOT want to see Hashgraph based currency, but that can also be amongst those in such an exchange platform.
The possibilities are really vast.
You are 100% correct. It's the perfect distributed exchange technology. I would be happy to see it in a crypto as well, but it's properties would be fantastic for an exchange. You could scale globally without an issue.
Hello N_&_V.
Excellent!
Once Again, you've surprised me with your insight.
This time, regarding 'comprehension' of revolutionary milestones in Cybernetics.
Kudos!
I actually came across Mike Maloney's #8 video only a few hours after it was released.
I listened to the first 15 minutes and bookmarked it for future listening.
As we know Mike doesn't take up hashgraph until ½-way in the presentation.
Then ZH-Zerohedge had a headline the following day:
http://www.zerohedge.com/search/apachesolr_search/
Having Christmas ' & Family 'intervening' in my available time the past few days put my own 'investigation' of hashgraph on the back-burner.
So.
What a surprise to find you're on it.
Will checkout all of the videos remaining as priority #1.
Thanks a lot!
X-
Thanks very much X- ! I'm very glad that you've found it interesting and helpful. I think Hashgraph is fantastic technology for enabling true peer-to-peer interactions and businesses (and hopefully taking out all those historically dangerous middlemen at some point). After realizing that this was the case, I thought I should try to make it a bit easier to understand for everyone regarding what it really can do.
I hope you've had a great Christmas and I wish all the best to you and your family & friends in the New Year. Thanks again for the positive feedback and enjoy the videos. I think you'll find them very interesting. - N&V
An excellent post @newsandviews and goes at least some way in explaining to someone as technologically unaware and incompetent as myself how blockchain tech works. I have seen for a while the potential dangers in blockchain technology and wonder at the choice and use of the word itself, as with so many others in life. Hashgraph definitely sounds a more secure and less controllable route in future, despite the 34% statistic.
I'm really slow at all the techno mumbo jumbo. My brain isn't wired that bway but I try to keep up as best as I can and am curious to find out more having read your very interesting article.
Thanks, my friend. I hope your day is going well. :)
Thanks a lot tonyr! I'm very glad to hear that the article has been helpful. I really want to make sure that it easy to understand as well, so thanks very much for the thumbs up. Considering the potential dangers in manipulated and controlled systems, I think it's really important to make it easy to understand the differences between the technologies and what they offer vs. others.
I believe that Hashgraph is fundamentally different and is going to be a game changer for ensuring fair, fast, and secure peer-to-peer networks that can't be controlled by the historical monopolies and oligopolies that we have gotten used to in the past.
Unfortunately, I was wrong about the 34% statistic in the sense that all blockchains actually have the same potential limitation. So Hashgraph is no worse than any other shared ledger solution out there. I've corrected this in the article above.
I hope you're having a great Christmas and I wish you a great start to the New Year. All the best to you and your family and friends. - N&V
Thank you for the summary. I very seldom watch videos.
How would HashGraph compare to Steem?
Steem run on a blockchain, albeit one of the fastest ones out there at the moment. Dan Larimer has developed a blockchain called Graphene the underlies Steem, Steemit, and BitShares. Although Graphene is still not as powerful as Hashgraph, it blows away most other blockchains on the market in terms of performance.
Dan is now working on EOS which should allow even greater underlying blockchain performance as well as a whole lot of other framework capabilities that then don't need to be rebuilt each time someone creates a new blockchain. It wouldn't surprise me for a minute if at some point Steem, Steemit, and BitShares are migrated on top of EOS when it is ready. This would be a very logical future decision.
Since Hashgraph is only a consensus mechanism technology and does not cover all the other bells and whistles built into Steemit or EOS, Hashgraph could easily be used to swap out the underlying Graphene blockchain in Steem or whatever the blockchain in EOS is. This however is assuming that the design of Steemit and EOS is modular enough to allow this. Think of Hashgraph as just the shared ledger / database and all the other features as part of an application sitting on top of it. That's probably the easiest way to approach it.
I hope this answers your question, just let me know if doesn't.
Definitely seems like the next step! thanks for sharing 💯🐒
Thanks for letting me know that you found it interesting.
Very much so! thanks 💯🐒