I think that dropping the 1 science per citizen is a good idea. It's the only yield that gets this treatment, and - within the context of civ and history - it doesn't really make sense. There's no 'investment' in science if food essentially equals science because of population. Citizens are currently the best source of science with or without science buildings. That's crazy, if you really stop to think about it.
The more I dwell on this (while on a business trip, mind you!) the more I absolutely adore the idea. It makes a lot of sense, and would really open up strategies quite a bit.
Suddenly the extra science from tile yields, pantheons, early trade routes, and more all become way more valuable. Global literacy rates will stabilize, and the science line of buildings will actually allow for science specialty cities. Science will accrue at rates comparable to the other yields (hundreds, instead of thousands, on average). Academies will be very useful into and past the Renaissance.
There's a lot going for this. Really, all that would be needed would be a soft tweak of the science per citizen values, some stronger scaling on science from specialists, and an ancient-era science producer (a science per citizen bonus on the palace would probably suffice.) Tech costs would also, of course, come down.
I feel like I need to voice the needed other side of the argument here. First of all, this feels extremely drastic. It's not going to lower the value of food at all, getting production is still going to be just as important, the only difference is that you're pretty much forced to rush libraries->universities->public school in every game (a strategy that is already extremely viable). You're also going to be even more reliant on specialists, making that food and the growth as important as ever.
So in essence you're not changing anything other than making the sciencebuildings (which are already strong) and the science on tiles stronger.
This is going to be a total hell to balance, with everything from specific buildings getting too important to Korea's UA to Babylon to Maya. Wonders, improvements, rationalism, tenets beliefs, this is a HUGE project all to make Academies feel more viable?
Do you want to know a secret? Academies have never been and will never be balanced, in vanilla they were completely overpowered for 3-4 city empires, stacking up with all the science-bonuses, and pretty much negligible in large empires for the exact same reason they are negligible in CPP. Only difference here being that building large empires in Vanilla was just not viable.
You could emulate the power of the vanilla Academy simply by having it double its science gain every era if that's what you want, but I really don't think that's the solution to go for. In fact this is pretty much the exact same argument that was had about the customhouses before they were reworked into their current state. The better solution here would just be to figure out some kind of meaningful boost that could be given to the Academy instead, which makes it viable for the first 4 eras before bulbing outscales it.
For my closing argument I would like to quote a certain Mr G, and say:
"Don't reinvent the wheel"
The academy isn't the only concern. Population being too important is also something that is addressed by such a change, and I believe is the main thing Gazebo and others find enticing about removing +1 science per population.
I'd just like to point out again that removing the scaling science from population would not change anything about population being important. Bigger cities are always better, even in Vanilla, if you could instantly get 60 pop cities (and not get crushed by unhappiness) you would do it every time, the ability to work more tiles and more specialists just makes it better.
) - it really depends on what you use your citizens for. The only thing you can assure that a large city has is food (it had to get big somehow). 
