Namecoin Difficulty and the Death of the American Dream
Posted: Wed Aug 10, 2011 1:47 pm
Greetings all,
I am currently acquainting myself with the history and development of namecoin, so I don't purport to know everything, but I am a bit concerned we're in the midst of an easily avoidable problem. IMHO, one failure to act or bad decision can destroy such a nascent system, or hobble it enough that it never grows beyond a small community of die-hards. (whereas U.S. banks can nearly destroy the world economy with 1M bad decisions and they are 'too big to fail')
Currently difficulty is 4x what it should be. Everyone is aware of this, some lament it, and solutions are casually mentioned without any urgency to act. Perhaps i need to lurk longer to get a sense of why, maybe because namecoin is in an alpha state it is not considered detrimental that the network capacity absolutely will not grow until November.
I understand the principle that the system should regulate itself, but the current difficulty rules don't produce reasonable results when the mining input varies rapidly. The rules make sense for a lone network like Bitcoin, where purpose-built machines had little other use than mining that one network. Namecoin, on the other hand, will for the foreseeable future get most of its hashing power *not* from die-hard supporters, but from profit-motivated Bitcoin people who now have two viable networks to choose from. It has a unique problem that Bitcoin does not, so the same code doesn't produce good results.
Doing nothing, waiting until November 13th (as of this writing) for block 20160, will let difficulty drop to maybe 24k. At that point if 1 Terahash jumps on we'll reach block 22186 in about 3 days. All the BTC offers on the exchange will be vacuumed out. Then difficulty will adjust to match the hash rate and the profitability will disappear, so another huge drop and blocks that take 1 day again...Which will force us to make changes to the way difficulty is adjusted anyway.
I'm at work so i got busy and lost my train of thought, but since i'm already pushing tl;dr, i'll just ask, am i missing something? Isn't it smart to address this immediately? What needs to happen? Can a client update for this give more incentive for people to update before the block 24000 changes?
Thanks.
/H.S.Thompson ref in title, btw
I am currently acquainting myself with the history and development of namecoin, so I don't purport to know everything, but I am a bit concerned we're in the midst of an easily avoidable problem. IMHO, one failure to act or bad decision can destroy such a nascent system, or hobble it enough that it never grows beyond a small community of die-hards. (whereas U.S. banks can nearly destroy the world economy with 1M bad decisions and they are 'too big to fail')
Currently difficulty is 4x what it should be. Everyone is aware of this, some lament it, and solutions are casually mentioned without any urgency to act. Perhaps i need to lurk longer to get a sense of why, maybe because namecoin is in an alpha state it is not considered detrimental that the network capacity absolutely will not grow until November.
I understand the principle that the system should regulate itself, but the current difficulty rules don't produce reasonable results when the mining input varies rapidly. The rules make sense for a lone network like Bitcoin, where purpose-built machines had little other use than mining that one network. Namecoin, on the other hand, will for the foreseeable future get most of its hashing power *not* from die-hard supporters, but from profit-motivated Bitcoin people who now have two viable networks to choose from. It has a unique problem that Bitcoin does not, so the same code doesn't produce good results.
Doing nothing, waiting until November 13th (as of this writing) for block 20160, will let difficulty drop to maybe 24k. At that point if 1 Terahash jumps on we'll reach block 22186 in about 3 days. All the BTC offers on the exchange will be vacuumed out. Then difficulty will adjust to match the hash rate and the profitability will disappear, so another huge drop and blocks that take 1 day again...Which will force us to make changes to the way difficulty is adjusted anyway.
I'm at work so i got busy and lost my train of thought, but since i'm already pushing tl;dr, i'll just ask, am i missing something? Isn't it smart to address this immediately? What needs to happen? Can a client update for this give more incentive for people to update before the block 24000 changes?
Thanks.
/H.S.Thompson ref in title, btw