Author Topic: Ethereum & BitShares Partnership?  (Read 58527 times)

0 Members and 1 Guest are viewing this topic.

Offline cass

  • Hero Member
  • *****
  • Posts: 4311
  • /(┬.┬)\
    • View Profile
█║▌║║█  - - -  The quieter you become, the more you are able to hear  - - -  █║▌║║█

Tuck Fheman

  • Guest
This week I am meeting with Vitalik to discuss a partnership between our two projects.  We spent several hours last night and 5 hours to day talking about blockchain technology and challenges we both face.   I am happy to report that we are finding some common ground and are planning more regular collaboration.


Offline BTCjesus

  • Jr. Member
  • **
  • Posts: 21
    • View Profile


IF these two projects can collaborate I think it would be for the best. Both projects stand to gain tremendously by working together than competing against each other. I would love to invest into both projects if they can find a good common bridge they can build from!

Offline yellowecho

We have had some really good discussions that have lead to improvements in both our systems.   The first step is to reach agreement on technical approach and the next step is to work on community buy-in.

Obviously we need to win over Gavin and other Ethereum developers and they are currently not convinced according to V.   Primary argument so far is that DPOS creates 2 classes of users and that it is not "elegant" and is "over specialized for known attacks".

We will be holding a joint meeting with the Eth. devs this week to work through these concerns and attempt to weigh them against the other imperfect alternatives.   

In any event V. and I have plans to meet every couple of weeks for a brainstorming session and hopefully record our discussions as we work through complex issues in the space.

One of the major issues is the need for Merkle trees in chain databases.  There are a lot of trade offs and good discussion that will benefit both of us and the greater community.

This is very exciting news- looking forward to hearing any recordings of the discussions!
696c6f766562726f776e696573

Offline yellowecho

Ethereum does not have and never has had a relationship with Goldman Sachs. It's a malicious rumor spread by people who don't like the project. There were two former Goldman Sachs employees helping ethereum during its early days. Both left the bank decades ago. Let this fucking FUD die man.

Cool, I asked out of ignorance not out of disrespect so thanks for responding.
« Last Edit: August 16, 2014, 05:30:31 am by yellowecho »
696c6f766562726f776e696573

Offline bitmeat

  • Hero Member
  • *****
  • Posts: 1116
    • View Profile
Quote
I haven't evaluated all options, but clearly it needs to be "in-memory" and dynamic.

Redis is a very nice in memory database. What I'm most interested in is the notion of blockchain sharding.

I've been looking at Redis closely and love the direction it has been going.

It would be amazing if someone comes up with a way to run Redis nodes as a crypto. Think self-hosted, indexed blockchain AND various user data (smart contracts/apps/you name it).

You get StorJ/MaidSAFE + network stability.

Offline bitmeat

  • Hero Member
  • *****
  • Posts: 1116
    • View Profile
4) I would want to use a lua scripting engine to validate transactions relative to the SQL database
    a) Ethereum currently runs at 1.3 Mhz with their C++ interpreter, storage access via level db is the most expensive operation right now.
    b) I think the data set should be kept in RAM validated by delegates.   Delegates may need $15K servers at scale, but that should be reasonable.

Good choice! LuaJIT is shown to be faster than C/C++ in various cases:

https://gist.github.com/spion/3049314

charleshoskinson

  • Guest
Quote
I haven't evaluated all options, but clearly it needs to be "in-memory" and dynamic.

Redis is a very nice in memory database. What I'm most interested in is the notion of blockchain sharding.

Offline bytemaster

Quote
I would like to see a full up relational DB with SQL support as the basic abstraction for the block chain.

A blockchain is at its core a functional database with variation over time not place. Something like Datomic seems to be a significantly better fit.  At the very least the entire DB market is moving into the NoSQL realm. Why not chase something like Mongo and BSON style storage?

I haven't evaluated all options, but clearly it needs to be "in-memory" and dynamic.   
For the latest updates checkout my blog: http://bytemaster.bitshares.org
Anything said on these forums does not constitute an intent to create a legal obligation or contract between myself and anyone else.   These are merely my opinions and I reserve the right to change them at any time.

charleshoskinson

  • Guest
Quote
I would like to see a full up relational DB with SQL support as the basic abstraction for the block chain.

A blockchain is at its core a functional database with variation over time not place. Something like Datomic seems to be a significantly better fit.  At the very least the entire DB market is moving into the NoSQL realm. Why not chase something like Mongo and BSON style storage?

Offline bytemaster

One of the things we have discussed is lessons learned from Ethereum and BTSX and ways to generalize things better:

1) We agree that smart contracts are useful
2) We agree that smart contracts + BitUSD are very powerful
3) I would like to see a full up relational DB with SQL support as the basic abstraction for the block chain.
    a)  This would allow constraints to be placed on the tables / rows
    c)  This would make indexes easy to maintain
4) I would want to use a lua scripting engine to validate transactions relative to the SQL database
    a) Ethereum currently runs at 1.3 Mhz with their C++ interpreter, storage access via level db is the most expensive operation right now.
    b) I think the data set should be kept in RAM validated by delegates.   Delegates may need $15K servers at scale, but that should be reasonable.
5) One of the major slowdowns for Eth. is the use of merkel trees to support light-weight proofs.  I would do away with this feature.
    a) On a proof of stake chain you cannot validate block headers independent of block contents because you can not use POW as a proxy for trust.
    b) With DPOS + Bonded validation agents you can get cryptographic proofs good enough for light weight clients.  Delegates can lose their job for lying.
    c) Merk. trees help serve as a "double check" on the "deterministic application of transactions", but otherwise are unnecessary.  The check can be performed independent of the consensus algorithm.

As you can see the idea that Eth. represents can benefit significantly by working with our team and we benefit from their challenges as well.   Whether or not Eth. implements DPOS you can bet that V. and I will probably keep pushing the technology forward for future chains.
 
For the latest updates checkout my blog: http://bytemaster.bitshares.org
Anything said on these forums does not constitute an intent to create a legal obligation or contract between myself and anyone else.   These are merely my opinions and I reserve the right to change them at any time.

Offline Stan

  • Hero Member
  • *****
  • Posts: 2908
  • You need to think BIGGER, Pinky...
    • View Profile
    • Cryptonomex
  • BitShares: Stan
All I know is that when stuff like this starts happening on a Hot August Night in the deserted streets of Raleigh, NC you get cold fusion, flux capacitors, and warp drives...

Anything said on these forums does not constitute an intent to create a legal obligation or contract of any kind.   These are merely my opinions which I reserve the right to change at any time.

Offline bytemaster

Good news !  +5%

Quote
Effectively my goal is to make both projects better and to secure AGS/PTS holders a stake in Ether.

Will the AGS/PTS stake be at least 10% allocation each?
Also, could you elaborate on Ethereum's investment relationship with Goldman Sachs and how/if that would effect future Bitshares projects?

We have not discussed allocations or anything.  It is unlikely they could use our codebase and instead would probably just use the algorithm. 
For the latest updates checkout my blog: http://bytemaster.bitshares.org
Anything said on these forums does not constitute an intent to create a legal obligation or contract between myself and anyone else.   These are merely my opinions and I reserve the right to change them at any time.

Offline bytemaster

We have had some really good discussions that have lead to improvements in both our systems.   The first step is to reach agreement on technical approach and the next step is to work on community buy-in.

Obviously we need to win over Gavin and other Ethereum developers and they are currently not convinced according to V.   Primary argument so far is that DPOS creates 2 classes of users and that it is not "elegant" and is "over specialized for known attacks".

We will be holding a joint meeting with the Eth. devs this week to work through these concerns and attempt to weigh them against the other imperfect alternatives.   

In any event V. and I have plans to meet every couple of weeks for a brainstorming session and hopefully record our discussions as we work through complex issues in the space.

One of the major issues is the need for Merkle trees in chain databases.  There are a lot of trade offs and good discussion that will benefit both of us and the greater community.

For the latest updates checkout my blog: http://bytemaster.bitshares.org
Anything said on these forums does not constitute an intent to create a legal obligation or contract between myself and anyone else.   These are merely my opinions and I reserve the right to change them at any time.

charleshoskinson

  • Guest
Quote
Will the AGS/PTS stake be at least 10% allocation each?
Also, could you elaborate on Ethereum's investment relationship with Goldman Sachs and how/if that would effect future Bitshares projects?

Ethereum does not have and never has had a relationship with Goldman Sachs. It's a malicious rumor spread by people who don't like the project. There were two former Goldman Sachs employees helping ethereum during its early days. Both left the bank decades ago. Let this fucking FUD die man.