The concept of 'proof' in mathematics (education) is overloaded with importance/fear/magnificence/awe. What is 'proof' if not just an explanation to that or another claim by a sequence of logical steps? How is it different then 'proofs' in history? Few educators of history would suggest teaching historical facts without understanding and arguing about their cause, effect or importance. Yet, in mathematics education there seems to be more room for debate.
What is a good point to start learning proofs? Well, whenever the student is showing an interest would be fine. Showing an interest means that, at least on some level, they are ready to tackle it (whatever it may be). Then, of course, some care needs to be taken as to the level of maturity the student has in order to decide on the approach to take.
What are pros and cons? Mathematics without proof is, at best, a cook-book recipe for long and boring computations. Not understanding what one does may lead one to mistake mathematics for magical rules to be followed just because somebody said so. It robs the students from the ability to understand what they are doing and how to check their answers make sense. Of course, introducing long and tedious proofs is also dangerous. Before a typical student can appreciate the intricacies of, say, proving the reals exist, one must first come to appreciate the merits of such a proof. Not all things should be proven, it's quite alright to take some things on faith, as long as we are not oversimplifying anything.
Does the whole curriculum needs to be geared towards proofs? Well, it should be geared towards understanding, and that means that it should include explanation. Some people call it proof.