Sorry for the long delay… I’ve been gone for a few weeks (Pennsic War, W00T!) and also have been reading Guns, Germs, and Steel (best book evar!!!!) which has given me food for thought…
Oh, I see what you mean now, NW_Kohaku, and what you are getting at! I agree that the life of the average peasant probably didn't change much regardless of the civilization in medieval times until the political, economic, and technological situation came about that gave more social mobility to the lower class. I also have some rough answers for how to translate these "tech upgrades" into DF.
For your example of the power loom, I think the main problem lies in the issue of how skills and economic activities are modeled. If one cannot model the need for spinning thread faster, then there would be no use for a power loom and a hand spindle would work just fine; no tech improvements needed or desired. If time and productivity of certain crafts were modeled better, than the merit of making available certain technologies in game would be apparent. As you mentioned before, many tech advancements are simply just developing ways of doing things that could be done before but better and/or faster. Sometimes these improvements could lead to being able to do novel things ("unlocking" certain activities, i.e. being able to process iron ore after improving furnace design for working with something similar, maybe). So, in order to satisfy your question of how to meaningfully translate tech upgrades into DF, time and material requirements, among other things, of certain activities need to be better modeled. At least, I hope this satisfies it.
Also, with regard to your statement of raw manpower and wealth, that's pretty well put. Economic factors are what ultimately determine the implementation of technology, whether it be by a civ in general or a social class, or an individual. If there isn't the right infrastructure to support or reason to implement, say, a given farming strategy, then even if it's known about, it won't be used. Or, on the social class level, if the commoners of a certain society cannot afford to make or purchase or trade for certain implements, then they won't be used.
As for what I would propose for tech advancement in the game and how to model how knowledge can be transferred and lost... I'm basically going to reiterate what I mentioned in this thread:
http://www.bay12forums.com/smf/index.php?topic=46550.15We need the following things to be in the game in order for my idea to be feasible: writing, the earlier mentioned Bloat27, the ability to interact with other civs/forts/people outside of the fort and for those interactions to actually have an effect on the world. I know that's a tall order... but some if not all of this stuff is planned...
So, for the abstract knowledge system of Bloat27, ideally I think individual bits of knowledge should be able to be "known" by an entity much like (though I don't know specifically how this is done in DF) how legends, individual preferences, and stuff are recorded, except individuals should contain that knowledge if they don’t already.
This knowledge should be separate from associated skills (if applicable) in that one should be able to know about something yet not know how to enact it efficiently (if applicable), like knowing about how to make tables, but not being good at it.
Bits of knowlege should be transferrable through individual interaction (here things like teaching, philosopher, and other related skills could modify the ability of doing so) or through writing. They should also be able to be lost (like if the entity or book that contains the information is killed/destroyed).
Some bits of knowledge should be able to be acquired on one's own if the motivation, inspiration, and appropriate pre-requisite knowledge (if applicable) are there. How motivation, inspiration, and pre-requisite knowledge would be modeled, I don’t know. But I will propose some food for thought.
So, how would an entity come up with a novel bit of knowledge, like a new furnace design for smelting iron, or how to extract poison from GCSs or something? I think it could be as follows: Using furnace design as the example, the entity in question would have to have prerequisite knowledge (in this case furnace operation and building, knowledge of copper smelting, knowledge that pumping air into a fire gets it hotter etc.). The entity would also have to be motivated and have the time to engage this motivation (factors in this case could include presence and access to iron ore, inherent curiosity of the individual, intelligence, etc.) The entity, meeting the prerequisite knowledge and motivation, would spend some of his/her working time devoted to experimenting, building a new furnace design, perhaps with the help of others and perhaps being able to fail at the endeavor. This represents coming up with something through inspiration/motivation and not through accident or a more gradual realization of something. These other kinds of coming up with new ideas could be modeled differently maybe… I dunno.
Individual technologies would have to be something coded in the raws, as I don't think the sophistication of video game AI (as good as it is in DF) is to the level where video game entities have the ability to think things up as is done in real life. But it can be modeled loosely, with developing certain bits of knowledge necessitating certain prerequisite knowledge. So, yeah, a tech tree… This will require a LOT of thought and research to make it a plausible one.
Oh, and regarding books and how it can work with this: books should be a vessel for knowledge in the game, able to pass down information through generations and across cultures. In addition, books should also be vital for tasks such as record keeping of stockpiles, wages, and government related stuff, as I think that's a big part of why writing came about in the first place. Basically, writing should serve as it does in real life as a technology/tool to help one to "remember" something they otherwise couldn't efficiently, like large amounts of mind-boggling numbers in mathematical operations, data during experimentation, code of laws when administering justice, histories of entire civilizations, etc.
As for your (NW_Kohaku’s) previous statement “changes in the farming system are, again, not translatable in DF by some sort of bonus to food production, but would be represented by having a player who better understood how to make the most of his farms, as these advancements were from a top-down rethinking of how to organize agriculture, and public works projects to irrigate it properly”...
Perhaps things that involve “top-down rethinking” may not be able to be modeled for a
player run fort… things like architecture and engineering, in addition to farming as you stated, come to mind. !!!BUT!!! I do think they could be modeled for
npcs and
their forts, though. For example: crop rotation knowledge could allow the AI controlling another fort or city to implement it. The same goes for architectural motifs and engineering principles perhaps. Those NPC forts with masons/engineers with knowledge of, say, arch construction, would possibly have doorways that are arch like (as far as the game can portray that and the physical characteristics of the structure). The knowledge could even be present in the dwarves in your fort, but they wouldn’t matter since you, the player, are in control, and not the computer, for the things where this knowledge applies. The knowledge could be passed through your fort to others like any other knowledge. This would require a lot of tweaking to get right for the AI; one would have to model the motivating factors behind the implementation of these afore-mentioned technology-strategies.
How could Toady make the AI know when it’s appropriate and desirable to utilize a technology? I don’t think this will be at all easy. Let’s use aqueduct technology as an example: Say the AI perceives the need (how this would be done, I dunno) of getting water from point A to point B. What determines the strategies available to the AI for executing this would be the knowledge contained in the fort by its dwarves or whatever. Say the AI’s fort has dwarves in it that have knowledge of arch-building, scaffolding strategies, etc. This could allow the AI to execute a strategy (hard coded?) of building an aqueduct in a particular style (think Roman ones) if doing so would be considered the “best” of the strategies available. What would determine one strategy being used over another would probably involve things like taking stock of available resources, estimated time of completion, etc… Perhaps this would be WAY too difficult to implement and not worth it given how little it would impact the human player, but I’d love to see it done. Then in adventure mode, forts would look more like forts than just mazes of corridors and arbitrarily sized rooms. Blargh.
With some further thinking and refining, I think my suggestions could set up a framework for technology acquisition, transfer, and loss that fits with Toady’s goals of making a game that procedurally generates a fantasy world with a modicum of verisimilitude.
Perhaps people may think this implementation of transferable knowledge and stuff a task not worth pursuing, but I think it would be something that would set this game apart from all others. Modeling tech transfer and development like I mentioned would be not only of interest to me and perhaps some parts of the DF gaming community, but I guarantee you that it would get a LOT of attention from many branches of the scientific community (for whatever that might be worth).
Sorry if this is tl;dr-worthy... my brain is fried now.