It's a lie (intentional or not), and one that does the author a disservice by (a) making him look like a fool to his peers, and (b) obfuscating the actual benef

It's a lie (intentional or not), and one that does the author a disservice by (a) making him look like a fool to his peers, and (b) obfuscating the actual benefits of his computing model (which, from the paper, look to have more to do increasing performance of certain parallel algorithms rather than fault-tolerance).