Bitcoin Forum

Here's a conceptual approach for merging two or more block chains that are created as a result of lack of network connectivity between parts of a network.

http://bitcoinery.tumblr.com/post/6731628999/how-to-merge-bitcoin-block-chains (http://bitcoinery.tumblr.com/post/6731628999/how-to-merge-bitcoin-block-chains)

Essentially, the suggestion is to not simply discard smaller chains, but to merge the differences back into the longest chain. This maintains all existing properties of bitcoin.

I can't read this link, but what does it propose that doesn't already happen? 

Neither are they a feature of Bitcoin - Transactions from invalid blocks become pending again already. This Blog post solves an issue that is already solved.

Thanks for looking at it. I'll update the article to say that this is what is currently happening.

If this is already happening how does it handle long delays i.e. one month.
What happens to bitcoins created and spent in this case? Obviously the transactions are voided, but the economic activity has already taken place and the money would have been distributed through the network as wages, purchases or gifts.

Seems that for areas of frequent connectivity dropouts it may be better to have their own currencies and block chains with an exchange rate.

Is there any documentation for how block chain management works?

The same way that it handles short delays.


If coins are created during the split, and the split lasts longer than 100 blocks, then those coins and all the transactions that depended upon them in the minority chain simply cease to exist after the merger.  There is no other way to do it, unfortunately.  If the minority blocks were permited to exist, block splits would become a profit vector in their own right, breaking the 21 million coin limit with ease.


If you have frequent connectivity dropouts, you probably shouldn't be trying to mine at all.  But blockchain splits should never be an issue, as it's trivial for a watchdog process to detect a blockchain split, particularly when it's on the minority side of the break.  As large as the Internet is, a near even network split, resulting in a near even block discovery rate on both sides of the divide, is astronomicly unlikely.  If the rate of block discovery, from your perspectives, drops by more than half then users should be suspending the acceptance of new coins created since the division anyway.


Have you read the white paper?

I'm surprised how many people think that mining (i.e. doing it just to get the coins) is the most important thing.

100 blocks take about 17 hours to generate. I'm talking about longer dropouts due to natural disasters, etc. It's easy to say "should suspend the acceptance of new coins created", but hard when you're faced with economic reality of supply and demand in a 24/7 economy of a busy city/country.

I don't mine and I don't intend to mine.
I'm beginning to believe that bitcoin reward for mining is a bottleneck that will inhibit the growth and performance of the block chain and will be the cause of inefficiencies and complexity that is bordering on insane considering how hashing power is growing daily. I don't have a solution for this, but I'm actively thinking about it.

Yes, I read the white paper (Satoshi's PDF) and it doesn't talk about the economic reality. It's more like a spherical horse in a vacuum - technical proposal suited for ideal situations.

You don't have to be a miner to notice the decreasing rate of block discovery.


Hard, perhaps.  Still something that should be taken under consideration if you find that your entire country is isolated from the rest of the Internet, and not just in the case of Bitcoin.


I never have, either.  Mining is still a necessary process whoever does it.



Yes it does.