On Wed, Jun 3, 2009 at 1:06 AM, Edouard A. <edouard@fausse.info> wrote:
Are you guys saying that all algorithms in boost are currently proven?
I'm not aware of any that aren't... But I don't know all of the libraries inside and out.
If I were to submit an I/O library (it's an example, I'm not working on I/O) using somehow a "new" algorithm to manage asynchronous requests, would you ask me to use a more classical approach during the review?
If you don't have a solid mathematical proof for the correctness of your algorithm, then I may instead ask for case studies, etc.
What is the difference between a new algorithm and a new implementation of an old algorithm?
Are you trolling here?
Does boost embrace innovation?
I think I have my answer.
Innovations comes in many forms. It doesn't matter if the algorithm is new or not as long as the author is able to give a certain degree of assurance regarding reliability and correctness, isn't it?
I'm questioning the degree of assurance required for a new algorithm to be unleashed on the unsuspecting masses. Jon