OhGodWhy

The change kind of scares me to be honest. From an administrator's point of view. Batoto is very optimized to be viewed by guests because guests make up roughly 90% of the visitors--and it's much easier to optimize for guests as they require less dynamic actions. When they all become members, many of existing optimizations made over the years become useless. I don't know if our current system can handle the increased load. I don't know if decrease in bot load will be significant enough. I don't know if it'll make more members leave us. I don't know if ad revenues will be enough to support the system once it's changed. There's quite a lot of unknowns.

Those unknowns are exactly the most important things you have to know before making such a huge change. Your objectives become redundant if, after the change, you don't have enough ad revenue to support the site - I also would not advise relying on user donation as a back up if things turn for worse.

If you are going through with it, or any alternative suggested in this thread, you should have a 1 to 2 week trial period so you can at least estimate the unknown variables to a reasonable degree before making a final permanent change (if any).

My opinion on the bot problem is that no matter what you do they will still exist (short of only letting people you approve visit this site). The most you can realistically achieve is have less bots in the site - but is that worth it for a maybe significant decrease in visitors? 

The people behind the bots do it because they make money from it, and they will not think twice in finding a way to get past your new change in order to keep making money.

Have you also considered that the ones that doesn't make an account will start visiting aggregate sites, which makes using bots on batoto more profitable?

Either way, there's way too many unknowns - the changes need to be tested before you go through with it.