I guess the difficulties are in collecting the works and collections of science, now largely in electronic form only… and preserving it as a large mass of data… duplicating it many times, and having protectors for it.

The best method might be to approach universities and science libraries to consider and sign on to a science preservation initiative… which develops and stores it’s science in a way designed to survive economic and political turmoil.

I don’t think you would need to collect “science”… let those that have it work toward its preservation… with the aid of well defined challenges, actions, metrics, and of course goals.