I'm probably the last person to say anything but do have a couple of questions about what "yavoon" experienced that have answers to help us both.
I used to screw-up Monarch level by building (ie. university) where it didn't do any good (ie. 95% commerce corrupt city). I would be paying upkeep for something that didn't justify it's upkeep cost. It sounds like YAVOON might have done this.
But there is the question of what to do with the 1-2 shields per turn that city generates. I still think I don't have the "best" answer for that. Usually I build a worker which is added to another city, works on terrain, or is sold. YAVOON asked what should be done with that 1-2 shields? Well? Inquiring minds want to know!
The other thing I think YAVOON mentioned is the 50+ workers he had for 30 cities during REPUBLIC. It makes me wonder how many total units he had and how much was spent on city improvement maintenance. It sounds like a lot, maybe. I wonder what his research level was compared to income compared to his 30% luxury slider. Maybe within these questions lies the answer to his apparent tech obsolesence compared to AI civs.
If he had maxed sci. research with fewer potentially "useless" city improvements (less maintenance) and fewer workers (maybe a 1-1.25 worker : city ratio) might this have aided his tech research (given what is conventional wisdom about techs being faster to research when one knows several civs already with the tech.)?
I, too, empathize with not being able to keep units upgraded with the most recent tech because of a fast tech advance situation. Generally I respond to it by building new units of the most recent tech before calling in existing units that are currently in defensive positions.
Excess workers can be used to prevent coastal landings by rivals (until marines) so both military units and workers can share the burdens of defensive strategies...to a point. Then, when RR is discovered, the workers can work on RR.
[EDIT: Never mind...didn't realize this thread went on...and on.]