Acknowledgment. I first read of the possibility of such a Trojan horse in an Air Force critique [4] of the security of an early implementation of Multics. I cannot find a more specific reference to this document. I would appreciate it if anyone who can supply this reference would let me know.
The date of this article is 1984, and by this time the aforementioned Air Force paper was effectively already lost. It saddens me to think of all of the stuff of this type we're going to have to relearn (possibly badly) because we failed to keep/retain/organize the information.
Archive.org retaining so much stuff and Google searching it is initially exciting, but there's so much still behind closed doors, whether they're those of the US DoD or just a simple commercial pay-to-read interface.
The ability to distill something as seemingly boring as a product (early/beta Multics) security review into an important and generally applicable treatise on software trust definitely earns my admiration (aside from the important obvious work that he's done on UNIX), and I would be delighted if I could ever contribute something so important to the world of Computer Science.
The date of this article is 1984, and by this time the aforementioned Air Force paper was effectively already lost. It saddens me to think of all of the stuff of this type we're going to have to relearn (possibly badly) because we failed to keep/retain/organize the information.
Archive.org retaining so much stuff and Google searching it is initially exciting, but there's so much still behind closed doors, whether they're those of the US DoD or just a simple commercial pay-to-read interface.
The ability to distill something as seemingly boring as a product (early/beta Multics) security review into an important and generally applicable treatise on software trust definitely earns my admiration (aside from the important obvious work that he's done on UNIX), and I would be delighted if I could ever contribute something so important to the world of Computer Science.