Carlos E. R. wrote:
The Sunday 2008-07-13 at 23:26 -0700, John Andersen wrote:
On Sun, Jul 13, 2008 at 9:26 PM, Rajko M.
wrote: [1] Security trough obscurity is often criticized as bad practice, but actually it is the only way security can work. ... The patches for the recent DNS security problem were prepared in secret by all distros and OSes. The hole itself has not been publicly explained, as far as I know.
That's a good sample of security by secrecy...
This is stretching the definition of "security by obscurity" to the point of silliness. You're usually better than this, Carlos. In all these cases, the algorithms are publicly known and described, the procedures are publicly specified, the programs are exposed to public scrutiny. All of this has always been known as "security by design". "Security by obscurity" has always been defined as hidden algorithms, hidden procedures, hidden programs. So we have no way of knowing whether a program is secure or not, until either someone cracks it, as is usual in the Microsoft environment, or until the developers and their allies uncover a flaw in the programming, as is usual in the open-source environment. When developers uncover a critical flaw in a critical, widely required component like DNS, it is only prudent to fix the flaw before letting it be known to users _and_possible_evildoers_. The old source has always been available (with the flaw), and the fixed source will be available as soon as is prudent. To put the hiding of keys and passwords in the same class as hiding algorithms, procedures, and programs is simply silly. Really, guys! John Perry -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org