BitcoinEXpress, I'd say if it was not intentional then it's not an attack. But if it was not intentional, why don't you show some good faith and return the coins to their original owners before the reorg? If you send the coins to me, I will make sure they get to their rightful owners. If you do that, people won't be as upset at you.

FYI, If you click on the title of the actual post, you can get a direct link to the post:
https://bitcointalk.org/index.php?topic=47135.msg561292#msg561292

Honestly, I rather have the code lock down most of the 7.7 million fund for your laundry use than for only 4 million to be locked down and you have 3.7 million that you will give to faucet and developers at your discretion. But that's just me. I would suggest you to create a new poll and see what people think.

The nice thing about this solution is that it will prevent both scenarios: Lolcust goes insane AND hacker steals laundry fund.

I'm going to assume that all exchanges will be running on laundry checking code, since it would be insanely stupid for them to not do that. So if the attack happens, the attacker would not be able to unload their stolen coins to an exchange because the exchange will not consider those transactions as valid. So in that sense, the attacker would not be able to do much with those stolen coins.

I don't quite understand what you mean by "the exchange being the one to pick up the slack when reorg happens (I bet exchanges would verily not like such a prospect)."

There's quite a difference between laundry gets hacked and coins sent to laundry stolen VERSUS 7.7 million laundry fund gets stolen and dumped on the market. The former will only hurt the future of your laundry business. The latter will kill Tenebrix altogether. So this solution should prevent the latter from happening but it does nothing for the former. You still have to protect your laundry service from hackers. But if they do somehow manage to hack into your service, at least the network prevents them from touching any of the 7.7 million coins in the laundry fund.

The act of choosing which miner (the current one or the one with the laundry checking code) is a vote in itself. So I suggest we release the client with the laundry checking code. If you care more about making sure the laundry fund stays intact, then you run the new miner client. If you are more worried about not wanting to be tied to a coin laundry service, then you run the original miner client.

If in the future, there is a transaction that steals money from the laundry fund, then the chain will fork. And those people on the wrong side of the fork will likely not want to stay on their fork where the laundry fund is being misused, so they will naturally want to switch to the other fork. I think this will work itself out nicely.

On a reorg, how does the code handle your unspent coins? Does it do a rescan? I assume it can be done the same way. Basically it could treat these 2 addresses the same as it treats other addresses in your wallet.

If a reorg causes some transactions to be undone and those transactions became "illegal" due to the reorg, then they won't be spent until they become "legal" again. Unfortunately, I don't know enough about this to know if there are edge case issues with this implementation.

Should be easy to keep track of a running total, right? When it initially load the blocks, it calculates the coins in those addresses the same way it calculates your unspent coins in your addresses. So at each block, you always know how many coins those 2 addresses contains. It will be similar to how the client won't let you spend coins you don't own. Should be simple enough. And it should be simple enough to test it also. Once the code is out there, Lolcust can create a transaction that tries to spend coins that violates this rule and people can test to make sure that the transaction is indeed rejected by the code.

Not really. Someone send 2000 coins to the deposit address. And at random times after that, 20 different 100 coins transactions are sent from the clean buffer to 20 different addresses all belonging to the original sender. Then let's say there are 100 people laundering coins around the same time. There's really no way anyone can associate which clean coins came from which dirty coins.

Basically people are wary about the 7.7 million premined coins. It takes 3 years for the network the mine another 7 million coins. So for the foreseeable future, Lolcust has absolute control of the market. And if he decides to unload part of his laundry fund in the future, no one can stop him. Or if some malicious user manages to hack into his computer to steal his coins, they can easily crash the market. And that's troublesome.

Here's a suggestion I made to Lolcust on IRC. Basically have the Tenebrix code enforce that the laundry fund can never be stolen by Lolcust or anyone else for that matter. The laundry can be implemented using 2 buffers: clean buffer and deposit buffer. These buffers are just 2 tenebrix addresses that contain coins. The clean buffer will be where clean coins come out of. The deposit buffer will be where people send their coins that they want cleaned. After a random amount of time, the same amount (minus fees) will be sent to them from the clean buffer. Initially, the clean buffer will contain the 7 million coins and the deposit buffer will have no coins. After 7 million coins are laundered, the clean buffer will run out, and the 2 buffers will swap roles.

Given this implementation, we know that if the laundry is running correctly, the sum of the 2 buffers will always be greater than 7 million coins. If at any point in time there are less than 7 million coins in the 2 buffers, then either there's a mistake in the laundry code or someone is deliberately stealing from the laundry. My proposal is that we change the code to prevent this from ever happening. All you need to do is add an extra check in the transaction checking code. If the transaction involves taking money out of either of these buffer, just make sure that the 2 buffer still sums up to at least 7 million. If it doesn't then the transaction will be considered illegal and will not be added to a block and confirmed. And if a rogue miner tries to add an illegal transaction to a block it found, that block will be rejected by other miners. With this code, people can be sure that this 7 million premined coins cannot be used for anything other than the laundry and will be forever stuck in those two addresses. And in the future, if Lolcust turns rogue and releases code without this check, people will see that and refuse to update their client to that one.

I offered to help Lolcust code this up. What do people think?

Yup, already been discussed. Lolcust premined 7.7 million coins. He's supposed to use about half of that for development work and faucet. The other half will be kept for his yet to be implemented laundry service. And he claims that he will never unload those coins on the market. So far, Lolcust has been doing an awesome job in supporting this chain with pool and exchange.

If the premine coins really bother you, there's Fairbrix: https://bitcointalk.org/index.php?topic=46528.0
Fairbrix is pretty much an exact clone of Tenebrix just without the premined coins. Pool and exchange should be coming pretty soon.

I cleaned up the first post. Good enough?

Ok idiot, tell me which chain was forked with an exchange set up immediately.

It's not ok, but in all fairness, ~35 thousands coins is a lot less than your 7.7 million premined coins.

Not sure what you mean by a rebrand. The relaunch of Fairbrix had a hiccup due to the ~1400 blocks getting rewritten in a chain reorg. Other than that, people are mining happily away. I expect to see a pool soon and eventually, we'll get an exchange.

Well, I'm interested in doing this for Fairbrix. I don't think the community would hate this. With a new currency, you only send coins to the exchanges anyways. So it's not unreasonable for the exchange's view of the chain to be the "right" one in order to prevent an attacker from screwing over the exchange. I'm sure we can come up with things to limit what the exchange can do.

Here are some possible bad things the exchange can do:
- Run its own miners and only accept blocks produce by its miners, thereby reaping all the mined coins
- Do a double spend against its own users. After a user withdraws FBX, the exchange can lock in an different block and reverse that pay out
- Totally kill the chain by locking in a bad block

Somethings we can do is to have the ability to undo a lock. Maybe a kill switch for this block locking functionality. And if the exchange really misbehaves, people will stop using it, and I can always release a new client without this code.

Well for those that are still mining Fairbrix and want to run the client as a daemon, here's the code:
https://github.com/coblee/Fairbrix-daemon

Thanks to Lolcust for the Tenebrix-daemon-exp code that this is based on.

Fairbrix daemon is done: https://github.com/coblee/Fairbrix-daemon

What do you mean? Other than the chain reorg at the beginning due to an attack, things seem to be working normally right now. What issue are you seeing?

I'm not saying tbx price won't go up. Just saying mining might not be worth the effort and electricity right now. My machine does 11 khash/s. So instead of mining for 0.11 btc each day, I mind ask just buy 0.11 btc worth of tbx on the exchange.... or more if I believe in the coin.

Difficulty just increased to 0.09258932 at block 12096. A bit more than double the previous difficulty of 0.04195601.

By my calculation, if you have 10 khash/s, it takes about 11 hours to find a block. You will make about 51.6 tbx per day. At the current exchange rate of 0.002 btc per tbx, that's about 0.10 btc or about $0.50. So basically about 5 cents a day for each khash/s. In terms of making money, it's probably not worth it to mine tenebrix anymore. Mind as well buy them on the exchange.