From: Halo5 Subject: "Virus" vs. "Trojan Horse" Date: Tue, 05 Aug 1997 15:59:11 -0500 To: boldt@math.ucsb.edu Cahalan's definition of the trojan is not entirely true, either. >With a trojan, there is no "infected binary". The trojan binary _is_ >the original binary. Trojans are usually made by adding evil code to >the source tree of an application and recompiling. Viruses are made by >compiling a virus and then attaching the virus to a normal executable. This has nothing to do with the definition of a trojan. Take for instance code that exhibits the above traits for a virus. If said code is designed to activate on a certain date and erase the entire volume on which it resided, it would be considered a trojan horse (as is Michelangelo). Such details are irrelevant; the fact that the user is executing code that is supposed to perform one task, and the trojan code is sneaking in "trojan style" in order to be executed so that it may perform its task, is what makes it inherent to the definition of a trojan horse. Paul W. Smith -- Who is General Failure and why is he reading my hard disk? Mikrosoft spel chekar for sail, wurx grate! This message is brought to you by Linus Torvalds, NOT Della Croce or Andrew Tanenbaum!