On 6 May 2007, at 14:37, Peter Dimov wrote:
Matthias Troyer wrote:
On 5 May 2007, at 06:16, Peter Dimov wrote:
* I agree that the time(0) seed is unacceptable. A good source of entropy probably deserves its own library. It's also not easy to make it header-only. One compromise could be for the create function to take an Engine argument, but this takes away the simplicity.
How about just adding an additional create function that takes an engine?
The existing create function is dangerous as it doesn't generate reasonably unique identifiers. We can't fix that by adding more overloads.
I fully agree that it is very dangerous, especially on parallel machines. However, in addition to fixing this, it would be good to have the engine be customizable. Matthias