I doubt this will (ever) be implemented, at least in this mod, but I can't help myself; it seems like a really cool idea!
The basic premise is figuring out how to integrate (in real-world terms) the 'Great Man' theory of history with the 'social environment' theory. On the one hand, for instance, yes, calculus was certainly 'in the air' during the early 1700's, but it still took Isaac Newton and Gottfried Leibnitz -- two of the great minds of the age, if not of history -- to actually discover it.
My proposal is this: disregard all the usual rules for technological advancement. Everything that currently is allotted to 'beakers' for research goes into what is now classified as 'great person points.' This includes libraries, monasteries, universities... but also markets, banks, temples, forges, theaters, etc. (I have no idea how this might mesh with the current system for commerce, but give me time and I'll think of something).
[...]
A "distinguished person" theory.
Don't use random chance. Well, work against it like this: Put them on different counters, or, if you must group them, then put them all the same, but make it so that when the meter fills and fires, it only removes the points for the GP that was made. Heck, if the meter was coprime to 2,3,... # of GP types, there'd never be ties and you could deterministically fire the greater accumulation, but probability would here be appropriately tamed so that the unlikely great people are -very- unlikely to spam your spawns, and occur with compensatingly more frequency than other GPs if they yet do.
This would be a lot faster, way more than even Distinguished People should be showing up, so now instead of taking the GPPs from every source at all times, only allow the best source of each given type to contribute. (the city with the highest gpppt) But to guarantee that (most) every city still aspires to produce a Distinguished Thinker or distinguished whoevers, to avoid having gamey minmaxxing of giving up on some city ever having distinguished Scientists, now you've got a reason to combine the otherwise valuable, even indispensable, bonuses of specialist slots or whatever, with the compulsory GP points that can sometimes make life difficult.
It's amazing coming across someone proposing almost exactly what I thought up for Civilization when looking at modding Civ V. Only I wanted to combine Distinguished people with Great People, one having mixed counters one being all separated ('all upside').
It is way too dangerous having probabilistic appearance of great people though, outright per turn. You need to use points to smooth that over.
Attention everyone, I have a bunch of serious suggestions. I propose we nerf the Wealth and Research (Culture too I guess) processes, as imo those should be a last resort after you run out of things to build, not an oftentimes superior alternative to actually building commerce multiplier buildings. Similarly nerf FailGold. The idea of it is to give you a compensation if you actually fail to complete a wonder, so investing hammers into wonders just to get gold should be a niche strategy at best. I suggest that for the commerce processes half of the production output is turned into the respective commerce type, not all of it. Only in a fully industrialized city, that is one with 100% hammer modifiers, should you be able to achieve a 1 Hammer = 1 Commerce conversion rate. FailGold should be nerfed even harder to compensate for stuff like Marble and Stone modifiers. You should only receive a quarter of production back in the form of gold, thus FailGold is never more efficient than building Wealth directly unless you have a boosting resource and the Organized Religion modifier.
God yes please.
Please have Meager Wealth process from turn 1, maybe meager culture. Like 10% to the yield. As you say, 1-1 conversion should be a privilege of a very specialized , equipped, and accomplished civ and city. The ability to even make the kind of choice, of building commerce in place of a particular thing, should not show up except to some civ who has completed high-value tech, maybe a special building or perhaps an exclusive one (world wonder). Between these extremes, the understood technologies unlock the 33%-50% conversion, as "something hammers can do".
Here's something for failgold:
Failgold works for the first 20% of the wonder's hammers, and itself pays off at the Meager Wealth exchange rate.
The next 30% of the hammers of the wonder are rewarded with culture. At the height of all half of the wonder, you should get maybe 4 turns worth of the culture that wonder provides.
If you got halfway there it's still really bad, actually desperately so, because building half is way more like you just guessed wrong than any alternative situation. Working more hammers and failing here should get you relatively worse.
Ideally if it could be made, at 66% or 75%, you receive a building in place of the wonder. Maybe the building is unique, but of course worse than the wonder. Has no special effects. I'm not sure whether it should be worse or better than typical buildings of its era.
You win some failgold and culture for amounts past that threshold (or all is silent for the last stretch...?)
Possibly there's a prize for 95% completion, a world marvel only constructible by failing to get the wonder, so instead the previous tier is just a completely normal building, and this tier is the unique but not wondrous building described.
Still have to avoid making wonders a way to get gold, and you can't console with hammer overflow either.
I'm sure the key is this graduated prize program. For even attempting a wonder to get a consolation is critical for I think game reasons of encouraging the attempt (it's too punishing to try otherwise!), for immersion reasons of explaining what happens when you -don't- build that thing you were all pretty sure you were building 10 years ago. The trick is the consolation must -never- be a decent way of trying to get the consolation resource, which has to be done by mixing up different ones with brutal caps.
Room for more complexity, failing the wonder puts the unique building in the queue, but it may decay, or maybe it's not even selectable and won't stick in the queue. So you gotta 'see the project through'. Especially neat if the hammers required are equal or greater than the run to the end of the wonder missed, but then definitely the consolation has to at least provide culture per turn.