In order to resolve our doubts concerning [Lamport and Melliar-Smith's JACM 1/85] analysis of ICA [Interactive convergence algorithm for Byzantine failure clock synchronization], we embarked on a formal verification … [ref]. … [p. 7] In total, we found that four of … and that the main induction was incorrect. … We found the formal verification of ICA to … [p. 12] Note that our corrections merely dot the … [in the corrected proof] is due to Lamport and Melliar-Smith.
Google-1
Google-2
Copyright clearance needed for quotation.