In amongst the political and emotional arguments for hard-forking the ETH blockchain or not, here's an interesting examination of DAO code which reveals that in addition to the recursion problem a fat-fingered typo could lead to much more profound problems. http://vessenes.com/deconstructing-thedao-attack-a-brief-code-tour/ If I'm reading it right, it suggests that in addition to the split sending tokens to the attacker's child DAO (which is currently ticking down its 26 days before ETH release) - the event of the 17th people are referring to as "the hack" - the actual token count may be completely out of whack in both the created children and the original DAO now. Which would be an accounting nightmare and lead to further repercussions down the line for the community to solve. mk and insomniasexx - anything tingling your coding senses on this? Edit: Here's a much more forensic, step-by-step examination of the exploit. http://hackingdistributed.com/2016/06/18/analysis-of-the-dao-exploit/ Edit 2: A thoughtful post about how choosing a procedural vs a functional language for the EVM showed naiveté. https://www.reddit.com/r/ethereum/comments/4opjov/the_bug_which_the_dao_hacker_exploited_was_not/ Edit 3: Philip Daian provides a more interesting analysis of the situation and concludes that there may be some very serious issues with Solidity itself (at least when it comes to handling financial issues in the order of millions of dollars). insomniasexx this is worth a read.