<oembed><type>rich</type><version>1.0</version><author_name>npub12ftsvulestm9ztkjkgkl7cfurm6u6dau835evhcs3jwxyctn4y6qsk0jt6</author_name><author_url>https://nostr.ae/npub12ftsvulestm9ztkjkgkl7cfurm6u6dau835evhcs3jwxyctn4y6qsk0jt6</author_url><provider_name>njump</provider_name><provider_url>https://nostr.ae</provider_url><html>📅 Original date posted:2016-08-03&#xA;📝 Original message:In light of the recent hack: what does everyone think of the idea of&#xA;creating a new address type that has a reversal key and settlement layer&#xA;that can be used to revoke transactions?&#xA;&#xA;You could specify so that transactions &#34;sent&#34; from these addresses must&#xA;receive N confirmations before they can&#39;t be revoked, after which the&#xA;transaction is &#34;settled&#34; and the coins become redeemable from their&#xA;destination output. A settlement phase would also mean that a transaction&#39;s&#xA;progress was publicly visible so transparent fraud prevention and auditing&#xA;would become possible by anyone.&#xA;&#xA;The reason why I bring this up is existing OP codes and TX types don&#39;t seem&#xA;suitable for a secure clearing mechanism; Nlocktimed TXs won&#39;t work for&#xA;this since you can&#39;t know ahead of time when and where a withdrawal needs&#xA;to be made, plus there&#39;s still the potential for key mismanagement; Similar&#xA;problems with OP_CHECKLOCKTIMEVERIFY apply too – unless you keep a private&#xA;key around on the server which would defeat the purpose. The main use case&#xA;here, would be specifically to improve centralized exchange security by&#xA;making it impossible for a hot wallet to be raided all at once.&#xA;&#xA;Thoughts?&#xA;&#xA;Some existing background:&#xA;&#xA;http://hackingdistributed.com/2016/08/03/how-bitfinex-heist-could-have-been-avoided/&#xA;-- Proposed the basic idea for a time-based clearing house but using&#xA;blockchains directly, this is a much better idea than my own.&#xA;&#xA;roberts.pm/timechain -- My original paper written in 2015 which proposed a&#xA;similar idea for secure wallet design but implemented using time-locked&#xA;ECDSA keys. Obviously a blockchain would work better for this.&#xA;&#xA;Other -- if the idea has already been brought up by other people, I&#xA;apologize.&#xA;-------------- next part --------------&#xA;An HTML attachment was scrubbed...&#xA;URL: &lt;http://lists.linuxfoundation.org/pipermail/bitcoin-dev/attachments/20160804/fe4335fb/attachment.html&gt;</html></oembed>