| |||||||||||||
Choosing between portability and innovationChoosing between portability and innovationPosted Mar 3, 2011 17:05 UTC (Thu) by nix (subscriber, #2304)In reply to: Choosing between portability and innovation by mezcalero Parent article: Choosing between portability and innovation
If you have ten implementations you have ten different sets of bugs: nobody will be hit by all of them at once.
Your implicit claim that a single well-reviewed implementation can somehow be free of security holes, or indeed any kind of bug, is laughable on its face. I don't know of any software product of any kind that this has ever been true of (even TeX).
(Log in to post comments)
Choosing between portability and innovation Posted Mar 3, 2011 17:45 UTC (Thu) by martinfick (subscriber, #4455) [Link]
10 implementation may have 10 sets of bugs. But nothing prevents a bug from being in all 10 sets. Remember ping of death?
Choosing between portability and innovation Posted Mar 3, 2011 18:06 UTC (Thu) by nix (subscriber, #2304) [Link]
Well, yes, but that was in all descendants of a single implementation, wasn't it? (More relevant perhaps is cases where buggy algorithms have been implemented out of books into lots of unrelated programs.)
Choosing between portability and innovation Posted Mar 3, 2011 18:45 UTC (Thu) by martinfick (subscriber, #4455) [Link]
If windows inherited this bug from unix, I would say that there is just as good a chance that free unix implementations will inherit bugs from each other, if not a much greater one.
Choosing between portability and innovation Posted Mar 3, 2011 22:58 UTC (Thu) by nix (subscriber, #2304) [Link]
Linux, almost uniquely, didn't use the BSD TCP stack. Windows did (for a long time, if not anymore).
So, no, unless it was an algorithmic error Linux would not have inherited the ping of death (at least not *that* ping of death).
|
|||||||||||||