How Hashgraph Works - Dr. Leemon Baird at Thomson Reuters

Поділитися
Вставка
  • Опубліковано 6 вер 2024

КОМЕНТАРІ • 60

  • @user-ly4ht4hk1u
    @user-ly4ht4hk1u 4 місяці тому +5

    Please get Leemon Baird on Lex Fridmans podcast! Both are such great communicators

  • @alfahax
    @alfahax 4 роки тому +29

    One of the best explanations of hashgraph ever...

  • @bhaskar6205
    @bhaskar6205 Рік тому +4

    Dr Leemon Baird is so addictive to listen to. What a genius

  • @slimgringo
    @slimgringo 2 роки тому +1

    This stuff never gets old!!! Had to get my Leemon fix!!! LFG!!!

  • @hassandiallo5326
    @hassandiallo5326 2 роки тому +3

    I never have enough of him.

  • @simoneast7
    @simoneast7 3 роки тому +5

    If nodes each pick another random node to broadcast to, wouldn't you likely end up with massive duplication, whereby a node receives many different copies of the same message? And since it appears that nodes broadcast their message multiple times, how do the nodes know when to stop broadcasting?

    • @Supremax67
      @Supremax67 3 роки тому +6

      Actually, when the nodes sync, they only share what they don't know. This is easily doable by sharing the last thing they know and working down from there. They don't need to share the entire event, just what the event was about. In Hashgraph, they track rounds, you can find more explanations about this in other videos he made.

    • @Iquey
      @Iquey 2 роки тому +3

      @@Supremax67 this is so cool. It really is like neurons almost and how information flows through a network. Anyone who has gotten real time word of mouth gossip in junior high and highschool, especially before social media became big, gets it. Let's say the topic is, who is taking who to the homecoming dance. That's a time stamped event in the future. And there's a finite number of kids in the school who are going to go to this one event. (Afawk, obviously in real life sometimes people bring kids from other schools but for the sake of this example it's mainly kids within the school only going to the dance)
      Kids naturally start to pair up or form carpool groups if they aren't going as couples. As the days get shorter and shorter to when the dance is gonna happen, you learn through the networks you are in, who is going with who. You mainly just remember who told you and if you talk to your friends, you hopefully remember what you told to the next person... The goal and use of this gossip is precisely avoiding repeated information, needing to be reshared as if we had all forgotten every round. Obviously as human beings with a decent memory we don't, because that would make picking dance dates inefficient if guys and gals and all the LGBT folks all kept trying to ask people who were already paired up. Instead if you're a smart kid you will ask people out to the dance who you KNOW or suspect aren't taken yet.
      It very much sucks when you are the lonely kid who is out of the loop and last to know. But even if you are the kid who is last to know, and you still go to the dance alone, you know who is paired up with who because that's the culminating time stamp when the gossip is finished and you can just see with your eyes who is dancing with who.

    • @wilhelm.reeves
      @wilhelm.reeves 2 роки тому +1

      @@Iquey bruh

  • @MrEvertonian20
    @MrEvertonian20 5 місяців тому +1

    Genius.

  • @inthazone2473
    @inthazone2473 Рік тому

    Great explanation.

  • @MrTupalkata
    @MrTupalkata 3 місяці тому

    Can anyone explain why hbar price doesn't take off? ❤

  • @efrainlopez8292
    @efrainlopez8292 3 роки тому +4

    I love the concept. I’m just confused as to what if someone originates a false transaction, how does the network detect that?

    • @learntotradecrypto3470
      @learntotradecrypto3470 3 роки тому +1

      As long as the nodes are 66% honest it will even penalize false transactions.

    • @anonmouse956
      @anonmouse956 2 роки тому

      @@learntotradecrypto3470 So its byzantine fault tolerance is lower than Proof of Work? Then why use it?

    • @Supremax67
      @Supremax67 2 роки тому +1

      @@anonmouse956 -- It is a misconception of a network to claim they are 51% vulnerable. The math theorem states you can't do better than 1/3 malicious.
      Got to be careful where you get your information, if they are not a math professor with a PhD, I wouldn't trust the source.

    • @janielj8691
      @janielj8691 2 роки тому

      @@anonmouse956 POW takes more time which increases both Finality and Transaction cost., Hashgraph overcomes this issue through the Gossip protocol way.

    • @GrgAProduction
      @GrgAProduction Рік тому

      THIS is why the first 39 Master nodes are going to be run with big cooperations with big volumes and on all various parts of the world and in all various categories of interest...
      Then when they all depolyed, others will be able to run nodes themselves..

  • @lleung6755
    @lleung6755 6 років тому +10

    go public already dude

  • @divinesleeper
    @divinesleeper 4 роки тому +1

    but how do you know the person who sent you isn't lying about the last 2 people who sent them?

    • @Supremax67
      @Supremax67 4 роки тому +4

      cryptographically signed, something super computers would take more than a thousand years to fake. So no, you can't "lie" about it.

  • @mishax1
    @mishax1 6 років тому +2

    What if timestamps happen on the Lightning network, that way you have no miner that can re-order them.. wouldn't that make your argument to irrelevant ?

    • @Supremax67
      @Supremax67 4 роки тому +4

      Lightning network, you mean the centralized BS that Bitcoin came up with to fix their shortcomings?

    • @ghostplayz9754
      @ghostplayz9754 2 роки тому

      @@Supremax67 Hey man I have a question about Hedera. How would the nodes know that the first person to “gossip” is not gossiping a fake transaction. (Example: Bob says Alice pays John $1000 but Alice doesn’t consent)

    • @Supremax67
      @Supremax67 2 роки тому +1

      @@ghostplayz9754 -- Every transaction is signed, a signature that requires the knowledge of a private key. So unless you know the private key, you can't fake the transaction.

  • @manasasb536
    @manasasb536 3 роки тому +9

    Screw moon, we are going out of solar system

  • @YoutubeHandleModerator
    @YoutubeHandleModerator Рік тому

    "Virtual votes" is a nice way to throw off malicious intent if one were to do so in the far future.... I mean, just show the algorithm says they would have done XYZ and done.

  • @-Nobody15650
    @-Nobody15650 4 роки тому +5

    7 dollars before 2030

    • @Supremax67
      @Supremax67 4 роки тому +2

      Hello again! 140x by 2030 would imply a 2x every year for 7 years straight. Seems aggressive but I guess we shall see.

    • @Dr.AayushJha
      @Dr.AayushJha 2 роки тому +2

      We might see 7$ be next year.

    • @-Nobody15650
      @-Nobody15650 2 роки тому +1

      @@Supremax67 what's your thoughts about it now

    • @Supremax67
      @Supremax67 2 роки тому +3

      @@-Nobody15650 -- No thoughts. I am holding long term.

    • @MrEvertonian20
      @MrEvertonian20 5 місяців тому

      @@Supremax67still holding?

  • @denjua2234
    @denjua2234 5 років тому

    What happens if Bob hacks Alice and send themself 1000-H, is it possible for Alice to claim the H-bars back from Bob?

    • @Supremax67
      @Supremax67 5 років тому +2

      Hacks, you mean if Bob is able to know Alice's Private key? That same private key that would take the best computer centuries to break? That private key?

    • @ghostplayz9754
      @ghostplayz9754 2 роки тому

      @@Supremax67 Let’s say Alice “gossips” that Bob paid her $1000, how would the system know if the info Alice is communicating is non-fraudulent?

    • @Supremax67
      @Supremax67 2 роки тому

      @@ghostplayz9754 -- Every transaction is signed. That signature cannot be falsified, it would be rejected by the other nodes.
      The only way to fake a signature is by knowing the private key. Unless you gave that information away freely, very unlikely to happen.

  • @thiagofontana7595
    @thiagofontana7595 6 років тому +1

    Your marketing is great. Too bad your hashgraph would not resist at a strong sybil attack...

    • @dwahlstedt
      @dwahlstedt 6 років тому +6

      This is true in it's simplest and most elementary form - as described in this video. But there is nothing to stop anyone implementing this technology from adding any sort of artificial overhead to make a Sybil attack however expensive they want. You could require a micro-fee for each transaction, or you could use proof-of-stake or maybe require a certain amount of computation be done, etc., etc. You can build Sybil protection on top exactly how you want it.

    • @Supremax67
      @Supremax67 6 років тому +5

      This tech is aBFT, resistant to all form of attacks. But when the tech comes out, have at it and try to bring it down. I would love a good laugh.

    • @T4T4
      @T4T4 3 роки тому +1

      @@dwahlstedt Well you got that spot on Dave!

  • @dainthomassida
    @dainthomassida 6 років тому +2

    bla bla bla centralized

    • @Supremax67
      @Supremax67 6 років тому +2

      bla bla bla, you're irrelevant :)

    • @surferboy198
      @surferboy198 3 роки тому +6

      you probably invested in some other shit coin and now you are shitting ur pants XD

    • @drumbottle
      @drumbottle 2 роки тому +2

      Well, not everything has to be decentralised

    • @ghostplayz9754
      @ghostplayz9754 2 роки тому +1

      @@drumbottle centralization is what crypto currencies are out to solve. Why else do you think the dollar is failing ??

    • @rishmithahahaha
      @rishmithahahaha 2 роки тому

      Can somebody explain how this is centralised ?

  • @margrietoregan828
    @margrietoregan828 6 років тому +2

    go public already dude

    • @Supremax67
      @Supremax67 4 роки тому +2

      Done, March 2018

    • @MrEvertonian20
      @MrEvertonian20 Рік тому

      What a difference 5 years makes… here’s to the next 5…