What is full minion bounty sharing? If the support last hit a minion both the marksman and the support would get the gold. If the marksman last hit a minion, the support and the marksman would both get the gold. Of course its not tied to marksmen or supports, any two champions within a certain range of an enemy minion that dies to an ally will receive the bounty.
#Before judging this on the surface please consider the subtle ways it affects the game, I said I'd make a coherent argument so that is what I'm going to do:
The support could go to any lane to support whatever lane needed it the most, roaming from lane to lane not necessarily for a gank, but to actually stay in a lane to help the person who is struggling.
Junglers will still want to jungle in general because it offers untelegraphed ganking power, whereas if you had 5 laners and no jungle you are pretty much telegraphing when the extra laner is going to gank another lane because he has to leave the lane he is already in. Smite is still a good option for objectives, and it is nice to have at least one on your team even though it doesn't do much of anything in the lane, but makes jungling far easier. Thus, even with this change it would still be a winning strategy to have a jungler.
Even if you had something exotic like 1 3 1 and no jungler, it is just a matter of tradeoffs. The other team can afford to send their support up to make it a 1 2 1 + jungler and probably hold the lane in mid and use their jungler to secure advantage on the side lanes.
Support items could be readjusted in price and function so that they cost about the same as everyone else's gear because now they get to have a regular amount of gold. This also gives them more variety and opens up the possibility of roaming plays that can only happen in an environment where supports are encouraged to roam around to the lanes that need their help the most.
For instance, ambushes set up by the junglers to keep the supports afraid, or using your jungler as an escort to transport the support safely to their destination. Some supports would be better at roaming (bard) and others a lot worse at it (raka). This adds an additional axis against which supports can be differentiated.
It makes the game feel more like a team game because you can actually send dedicated help to someone who is really just struggling with his opponent, due to a mistake or an accident early on.
Although I don't really watch esports and I don't care about them, it would make the game more exciting to watch early on since you have so much more roaming action and the regular possibility of having supports cross paths with enemy junglers as they compete with eachother to affect the outcome of the laning phase for each of the lanes. Ambushes, stalking, counter stalking, counterintelligence, all made much more relevant. Jungle + support would be the new "duo lane", and the lane would be the river and surrounding areas.
Frankly the game needs a shake up pretty bad. People are getting bored of the same old same old, and I really sympathize with RIOT that just changing the game for the sake of changing it isn't a good idea if it makes the game worse--- **here's the thing though, I think my suggestion actually makes the game better**.
Lane bullies can be given more contrast between their bullying power and scaling, because now there are supports that can help an ally survive an enemy lane bully (at the cost of not being able to help other lanes while doing so). Champions that typically don't lane because they are very weak to harass or being killed by a more popular pick now have an ace up their sleeve to help them be the best they can be.
It divorces the support and marksman role, so they can be their own separate entities. You could have marksman mid, mage bot, bruiser top, or assassin top, melee carry bot, and tank mid, or whatever assortment you'd like. It doesn't really matter. If you are struggling the support can lend you a hand for a few levels and then Mary Poppins their way to the next town in need.
Because marksmen aren't married to supports, they can be tuned in the same context as every other lane. That means we don't have an apples tooranges problem when balancing bot lane anymore, and it can have the same value as mid lane and top lane. Thus, every lane feels significant and the support is given the agency and variety of play that it needs to be an attractive role to a large enough segment of players to keep the queue times reasonable.
This solves the support agency problem and the bot lane gravity problem at the same time.
#_Lastly this kind of thing CREATES diversity._
With a champion dedicated to being a parachute for failing lanes, people across the highest echelons of play will feel more confident about experimentation in actual games. They might pull out something unexpected to shake up how we think the game should be played, and it makes league of legends more dynamic and exciting.