What would you propose, Guest (london)? Things can only be tested and designed to a point, and then it has to "go live." The computer is an excellent tool, but it cannot simulate everything.

IF you are going to use cosmetics or anything else it has to be tested before you use it, and even then the variability between people can still cause a reaction. Penicillin, for instance, is an excellent antibiotic, but there are still people who are allergic to it -- and if I were going to get a reaction to a drug (still using an example) I rather have it as part of a study than otherwise.

Some things don't need to be tested for lethality -- I don't see why we would need to find the LD50 on lead arsenate, for example. But when the toxicity of a product IS established, I don't see the need for additional testing (unless circumstances require it).

As for experimenting on 10 babies to save 1,000 -- why, yes, I would, if the conditions demanded it. Moreover, I would let one of my children be part of the experimentation IF the conditions demanded it. I'm assuming, of course, that you are talking about conditions like MVSD, infantile AIDS, or a hydrocephalic shunt and not about something trivial.