You're right, I could do it myself, but I have a deadline and need to push other code first. This can be delegated. 

LMK if you know anyone

It's for several things

- Classroom simulation and learning
- An experimental general ledger I'm experimenting with

I don't envision it being too hard, but I haven't read the original source.

Because this code will be reviewed by PhDs and such, I need it to feel like a professional wrote this code (Unit tests, etc)

Freelancer posting: https://www.freelancer.com/projects/project-10687503/#/details

I need a C# implementation of the Ripple Consensus Algorithm as described below.  In other words I only need a way to build consensus among peers as described in this video:

Consensus Video: https://www.youtube.com/watch?v=pj1QVb1vlC0

What are they making consensus on? For my use case, it's not "ripples", instead, all nodes have to agree that random number X in the range of 1 - 10,000 equals random number Y.

NO Cryptography is needed in this project.  
NO Networking code is needed in this project.  

Instead of networking, the code will spawn a new thread that simulates each node class.
- There will be 100 classes of type node in memory (configurable).  
- As each node starts, the constructor will use a lock to add a reference to itself in the static dictionary
- A random wait time from 1-30 seconds will occur
- Each "node" will locate 5 random other nodes from a C# static dictionary.

At this point we have 100 nodes in RAM that create a mesh network.

The Main program will have a console prompt that asks me to type in a node number (in this case I type in a number from 1 to 100).  
The Main program grabs an object reference from the static dictionary.

Next the console application will ask me what variable "X" is.  I input an integer.  
Next the console application will ask me what variable "Y" is.  I input an integer.  

It's up to you to implement the consensus algorithm in a way the ensures all in-memory nodes get updated.

Part two of this project is to randomly poison the in memory nodes so that they
- Don't replicate the value (ignores it)
- Replicate a wrong value
- Delay the response in a FIFO queue


Docs:         https://ripple.com/knowledge_center/the-ripple-ledger-consensus-process/
WhitePaper:  https://ripple.com/files/ripple_consensus_whitepaper.pdf
Ripple Wiki:   https://wiki.ripple.com/Consensus  
Source: https://github.com/ripple/rippled/tree/develop/src/ripple

Unspent outputs are only a transactionID and an output number.   In other words , any spending of a Bitcoin requires this sha256 hash and a number referring to the output and the script therein.

Bitcoins dont exist. Inputs and outputs do.

What a rolling root should be (based on my understanding so far) is a pointer to that transaction.  

So how about this formula:

Agree on a confirmation count that provides zero trust guarantees.  Say 100,000

Find the lowest target in the blockchain (remember difficulty can decrease due to a split or other reasons).  If that is less than 100,000 (selected above) then choose the candidate purge point to start at current-100,000 blocks.  If it is more than 100,000 blocks away from the current block, candidate purge point starts at that location.

Change the block version number to "4" (versions 3 are already assigned) and include a special spot for old unspent transactions.  Transactions in this special part of the block would look like double spends if placed in the regular part of the block therefore special handling is needed.  Miners and full nodes will need to validate that the full txid hash ( no matter what normalized form it's in) is identical to the old tx.   

The transaction will then live in two blocks, but with the same txid and same outputs.  During the sunset period all clients and miners should verify that the old tx and new tx do have matching ids and content.  This protects from an exploit in sha256 as the current Tx is added to the trusted merkle tree.

After a sunset period of 100,000 blocks the old block can be purged.  

If someone attempts to modify the transaction, the SHA256 hash will be different.  All full nodes who see an altered transaction should reject the Tx, and if it's included in a block, reject the block.


Yes someone probably thought of this. Yes there are probably errors and inefficiencies but it seems logical.

Just curious what do you plan on doing with the FPGA? 

How much do you want to spend for it?

I'm thinking of selling my September 18th spot on BFL's order.  Order number #8061.

PM me with what you are thinking of offering.

For anyone interested I am selling a few of my preorders.

List of all sales
http://www.ebay.com/sch/makerofthings7/m.html?_nkw=&_armrs=1&_from=&_ipg=25&_trksid=p3686

1 x 60 GH/s Bitcoin Miner BitForce Single SC ASIC, Butterfly Labs, Avalon
http://www.ebay.com/itm/1-x-60-GH-s-Bitcoin-Miner-BitForce-Single-SC-ASIC-Butterfly-Labs-Avalon-/230952324907

2 Bitcoin Miners: 1 x 60 GH/s BFL ASIC and 1 x 800 MH/s FPGA
http://www.ebay.com/itm/2-Bitcoin-Miners-1-x-60-GH-s-BFL-ASIC-and-1-x-800-MH-s-FPGA-/230951973570

Date Paid      Order Number     Trade-In (Y/N)     Qty SC Jalapeno     Qty SC Little Single     Qty SC Single     Qty SC Mini Rig

Sept 13 '12      7937   (FPGA)
Sept 19 '12      8061                     N                          0                    0                             0                      1
Sept 18 '12      8590                     Y                          0                    0                             1                      0
Jan   15 '12     16609                    N                           0                    0                            4                      0

User
mot7

Has anything interesting happend on TestNet as a result of this change?

Here are my 2 cents

* Any regulation that applies to Money Orders will eventually apply to Bitcoin.

* This includes businesses that accept or resell Money Orders  ( http://www.fincen.gov/financial_institutions/msb/materials/en/prevention_guide.html )

* What explicitly isn't mentioned is that merchants (people who accept BTC) are required to register.  This seems to be partially covered by this sentence



If that sentence means that anyone who provides a service for BTC needs to register, then it's possible that miners, who are accepting BTC for securing the network, needs to register.

Aside:

I don't like saying mining "creates units of this convertible virtual currency" like the notice indicates. I think the actual creation occurred in 2009 when the Bitcoin Protocol was launched. I would argue that the protocol "creates" the coins, and USA based miners are just claiming them.

looks like it's fixed.  The volume is now showing a whole number for the last n trades

John we're talking about two different things.  "Volume" is the count of trades over a period of time.  It should always be a whole number.

Example:

3 trades for 0.5 BTC is a volume of 3. 

If "volume" is a non-fraction and an integer how can "volume" logically not be a whole number.

The last few trades on MtGox's website are infinite repeating digits.  This is very atypical and to have this occur several times in a row, along with a price jump makes me suspicious.

Can anyone validate whats going on?

For example the volume indicator says the quantity of trades ends in .3333333   .6666666 etc.

How can "volume" include a fractional trade.  A trade is a whole number no?

rPZhwqx7kQdssJD6i3cnEpd9kQ1Tofzjkx

What GPU technology will work with an ASUS A8N-SLI MOBO?

I'm a little out of date on CrossFire vs SLI, and am not sure if I can even use AMD technology in this system.

Doe anyone know what's best?

My goal is to mine LiteCoin and VanityGen... possibly do some BTC mining if it ever becomes profitable to do so.

Please note: If you see "Jr. Member" by your name, you have already met the 5 posts and 4 hour requirements and don't need to post here. Additionally, after you meet the requirements, it can take up to 10 minutes for your account to be automatically upgraded.


[/quote]

So I need 5 posts... OK this is #4

I would like to be whitelisted... My StackExchange profile is here:

bitcoin.stackexchange.com/users/1878/makerofthings7?tab=activity

There is irony for me having to spam a thread I don't care about in order to write a sensible reply to another part of this forum.

Why are there so many newbie restrictions?  This is a pain.