I want to link my Bitcointalk name with BTCJam's. Verification code: 2ff5f562-c092-4efc-9eac-5690b60ffd7f

I suppose they are using many different combinations. That doesn't really matter for most people because in the combinations, they're probably swapping out low-priority transactions (with low fees) because they really want to maximize the fees they get.


What? This is only for the next block, the transactions in the block that they are trying to mine.



Yes. Merchants would pay a few bitcents per block to see what transactions will be included. Then they'll know with a pretty high degree of certainty if someone even attempted a double-spend. If merchants could see this for the mining pools compromising ~80-90% of the hashing power of the network, then they could even accept payments with even 0 confirmations, because for a conflicting transaction to be mined (the main worry for merchants) would require most of the solo-miners (comprising ~15% of the network) to collude and include the conflicting transaction. Since I'm pretty sure that the non-pool 15% is pretty fragmented (no one person has control of more than 0.5% of the network), you'd need to get a lot of solo miners to include the conflicting transaction.

Why don't the 5 biggest mining pools (ghash.io, btc guild, eligius, slush and bitminter) publish what transactions they are including in the block they're mining? (The transactions that will be included in the next block if they mine it) That way you could be relatively sure that your transaction will (probably) be included in the next block. If you're a merchant, you can be reasonably sure there won't be a double spend, if all 5 of them include the correct transaction. (If some of them have a conflicting transaction you know that a double spend was attempted). This seems like a really useful feature. Why don't they provide it? I know that they publish the block headers to connected miners, but not the list of transactions.

You can't. It's completely controlled by Mt.Gox. If you want paper backups and all that stuff, you should store it on your own client (like blockchain.info).

Ah. Thanks.

Sounds like a good idea. So these mini-blocks would only contain the time, the previous block hash, and the nonce? Would they also contain the previous mini-block hash or just the previous normal block hash? I am assuming that you would expect miners to choose the chain with the most mini-blocks in the case of a fork? That makes a lot of sense.

Yes. However, that would mean that the bitcoin address itself would be the same. Funds sent to that bitcoin address could be spent by either private key; the txout script only checks the hash of the pubkey key, not the public key itself.

However, there could be a RIPEMD-160 hash collision of two different ECDSA pubkey point. That would have two different privkeys with one pubkey.

As long as it's less than about 256 bytes, you could include it in a txout or txin script.

Thanks for choosing a name for the attack that doesn't confuse people with other things. 

Because SatoshiDice. 

I am running it uncompiled.
Tried that Path thing (but I have python 3.3 apparently), still gives the same error.
Any other ideas?

Just finished. Took several hours. (I ran brew install openssl pyqt) And guess what?
This happens:
$ python bitmessagemain.py
Loading existing config files from /Users/*my username*/Library/Application support/PyBitmessage/
Adding 66.108.210.240 to knownNodes based on DNS boostrap method
Adding 66.108.210.240 to knownNodes based on DNS boostrap method
Adding 82.72.109.211 to knownNodes based on DNS boostrap method
Adding 82.72.109.211 to knownNodes based on DNS boostrap method
Adding 162.13.42.201 to knownNodes based on DNS boostrap method
Adding 162.13.42.201 to knownNodes based on DNS boostrap method
Adding 70.36.177.246 to knownNodes based on DNS boostrap method
Adding 70.36.177.246 to knownNodes based on DNS boostrap method
reloading keys from keys.dat file
reloading subscriptions...
Database file already exists.
Listening for incoming connections.
PyBitmessage requires PyQt unless you want to run it as a daemon and interact with it using the API. You can download PyQt from http://www.riverbankcomputing.com/software/pyqt/download   or by searching Google for 'PyQt Download'. If you want to run in daemon mode, see https://bitmessage.org/wiki/Daemon
Error message: the sip module implements API v10.0 but the PyQt4.QtCore module requires API v9.2

AGGGGHHH. How do I fix this?

As the OP said, the MAIT would reject new blocks that double-spend transaction they accepted. That's the only way for the fork to occur. If the MAIT has >50%, then it won't be a hard fork. Therefore, the MAIT should only reject double-spending blocks if the MAIT has >50%.

It's still going. How long should it take?

I decided that this wasn't going to work and just reinstall pyqt.
I ran brew install pyqt, and it had to install qt first as a dependency. Grr, but ok. However, how long should it take? It just finished downloading it, and is now running make. A process called "cc1plus" is taking up 76% of my CPU.

But I didn't do -daemon. I just did python bitmessagemain.py

Okay, I may have found the problem. In bitmessagemain.py it says "from PyQt4 import QtCore, QtGui" and do the stuff if there's an error. I just changed it to "from PyQt4 import *" and it worked. But now when I run bitmessagemain.py it fills up my screen with text and won't stop printing stuff like "remoteCommand 'pubkey'  from 195.158.42.109
broadcasting inv with hash: fc900fc608c236a3f3eeeb0230f6f750c01951ccfb42525b7eee23d7bd5fa6dd
ECDSA verify passed (within processpubkey)
within recpubkey, addressVersion: 3 , streamNumber: 1
ripe 0099c9745deded3908a6b4dc41106393428c9c37
publicSigningKey in hex: 045fb4b2d86a21d5950c88eb1d04e54cd4975e5e97952cd2ee69c5474868a0498323b8538874622 5043ce1f6a928924178b36837dd7594ee39d483650a28e1ff80
publicEncryptionKey in hex: 04c2b5f5656ff0418b5463de66fbc9ca667b1186376127aa23a79d26bea07f33fea757f03244688 8a2ee1f5b3d16e52e9ad9822724b25375ca348ced84feabddb3
We have NOT used this pubkey personally. Inserting in database.
We don't need this pub key. We didn't ask for it. Pubkey hash: 0099c9745deded3908a6b4dc41106393428c9c37
Timing attack mitigation: Sleeping for 0.191516828537 seconds.
"
And, for that matter, there's no GUI that shows up.

No, I am running this on my laptop. MacBookPro, snow leopard.

I wanted to install BitMessage. It had a lot of dependencies.
I had to install XCode command line tools, gcc, cc, clang, homebrew, macports, python 2.7.5, qt, pyqt, and openssl. That took about 3 hours. Now I hate makefiles with all my soul and being.
Anyways, I got it all installed, and this happens:

$ python bitmessagemain.py
Loading existing config files from /Users/*my username*/Library/Application support/PyBitmessage/
Adding 66.108.210.240 to knownNodes based on DNS boostrap method
Adding 66.108.210.240 to knownNodes based on DNS boostrap method
Adding 82.72.109.211 to knownNodes based on DNS boostrap method
Adding 82.72.109.211 to knownNodes based on DNS boostrap method
Adding 162.13.42.201 to knownNodes based on DNS boostrap method
Adding 162.13.42.201 to knownNodes based on DNS boostrap method
Adding 70.36.177.246 to knownNodes based on DNS boostrap method
Adding 70.36.177.246 to knownNodes based on DNS boostrap method
reloading keys from keys.dat file
reloading subscriptions...
Database file already exists.
Listening for incoming connections.
 PyBitmessage requires PyQt unless you want to run it as a daemon and interact with it using the API. You can download PyQt from http://www.riverbankcomputing.com/software/pyqt/download   or by searching Google for 'PyQt Download'. If you want to run in daemon mode, see https://bitmessage.org/wiki/Daemon
Error message: dlopen(/opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/PyQt4/QtCore.so, 2): Symbol not found: _qpycore_pickle_protocol
  Referenced from: /opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/PyQt4/QtCore.so
  Expected in: flat namespace
 in /opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/PyQt4/QtCore.so

What does this mean? How do I fix it? I definitely installed PyQt. The file /opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/PyQt4/QtCore.so exists, it has random symbols in it.
*edit* Looking through the QtCore.so file, I can see "qpycore_pickle_protocol" on line 26,653 and 30,960... *edit*