Note: This ended up being a bit long, so here's what I'm suggesting in a nutshell:
During world generation, the player picks a complexity level:
Kobold - The newbie mode, with only a few workshops, materials, jobs, everything. If a tutorial ever gets made, this mode hints at how everything works in the other modes as well as this one. The goal is to make most game elements obvious within the first few minutes of gameplay.
Human - The current DF, or the current DF with slightly less features: Perhaps some workshops or other buildings, reaction and materials combined into one (Fishery+Butcher->Butcher?). Or alternatively just what we have now.
Dwarf - The hypothetical future hardcore option, with workshop tools, most items needing several different ingredients, wheelbarrows, more reactions, more ores, more rocks, all of the extra stuff people are suggesting. Maybe even magic?
Before you prepare your embark party you pick a difficulty level:
Sandbox - Infinite embark points, ability to get more dwarves, no skill limit.
Easy - Slightly more embark points, during the game unhappy thoughts are less frequent and have milder effect while happy thoughts have stronger effect, combat is biased in your favor, trade is is biased in your favor, sieges are less frequent and smaller, skills are easier to gain and each skills gives better effects (such as bigger harvest), friendlier and less aggressive wildlife.
Normal - The current DF. (Although, one could argue that the current DF is closer to easy)
Hard - Everything from easy reversed, essentially.
FUN - Everything from hard (except maybe more drastic, such as even more weightier unhappy thoughts), nicer items unavailable at embark screen, possibly less dwarves and meaner skill caps, periodic and frequent floods, volcanic eruptions, animal attacks, army invasions, demon sieges, tunneling monsters, bands of megabeasts... Basically you play in one big map-wide HFS. Building any sizable fort in this mode would probably be similar to ascending in NetHack.
Depending on implementation, these difficulty levels would also work very well with the Adventure mode: Having less items and options to worry about would make it easier to get your bearings (disregarding the interface, which is a wholly different matter), while getting hungry quicker and making combat biased would make it more challenging once you've gotten too good at it.
The complexity level could also, again depending on how it's done, double as a great mod manager since the complexity levels I'm proposing are essentially mods. If it works that way, Toady can even get the community to do the work of actually making the changes once the relevant game elements are moddable.
Admittedly, this might take a bit of work, and might introduce a (hopefully) small amount of extra overhead into future development. But on the upside, we end up getting a lot of difficulty suggestions and problems dealt with, including lowering the entry barrier a great deal without taking anything away from the game itself.
---
Right now, difficulty can be seen as an issue in DF. The problem has several aspects:
- New players are overwhelmed by all the available options, and find themselves paralyzed and unable to deal with any of the myriads of possible dangers.
- Experienced players, even in relatively dangerous locales, can easily set up a robust fortress which is practically impregnable, and slightly... Boring.
There are quite a few suggestions on increasing the difficulty and complexity of the game that aim to fix the second problem, but I didn't see many dealing with the first. However, it is always prevalent because any decision to add complexity or difficulty must consider the tradeoff- making the already quite confused newbie even more lost.
This ties everything up, since making things more interesting for people uphill on the learning curve will make it even less manageable for those downhill, and the only option is less than prefect compromise in the middle. Since DF has a *steep* learning curve, this middle camp is small, and headed in the direction of "experienced player".
To clarify the terms in which I'm thinking, there are basically two kinds of DF players in terms of skill. The new player who is still figuring out how the game works, and the subsequent "skilled" player who essentially knows how to use any given feature offered by the game- that is to say he is able to do anything the game assumes he can do, such as assigning jobs, setting up production chains, trading, building traps, raising armies, dealing with invaders, handling mood issues and so on.
For purposes of this discussion, "skill" does not include actual proficiency in more abstract, strategic capacities. That you know how to dig a channel, how to fill a pit with water, how to make traps and how enemies path, says nothing about whether you can *design* an *efficient* moat, with well placed traps, properly selected defenses and well organized maintenance crews. Knowing how to make dwarfs farm and cook/brew/process the harvest says little about efficient personnel organization and good farm placement. Knowing how to designate stockpiles, workshops, and laborers so that they actually produce has nothing to do with knowing good workshop layouts, being able to adapt ideal layouts to the situation at hand, and getting good efficiency.
In short, there's figuring out how to play, and then there's figuring out how to play *well*.
Now so far, I'm doubtful anyone would disagree much since for the most part these are self evident things. But I'll venture forth something a bit more controversial, and say that the former should minimized, and latter maximized. In other words, games in general, and hence DF as well, should be easy to learn and hard to master.
What does this mean for DF? Here are my ideas:
You won't achieve much if you simply make DF less complex, the complexity *is* a lot of its charm. DF just wouldn't be DF without dozens of rocks, ores, metals, woods with tons of thing to make from them in different quality levels.
Despite that, just blindly increasing complexity will not necessarily make the game more fun. Suspend considerations of computing power and imagine the game had an academic-grade builtin chemistry simulator, you were able to obtain any chemical that is potentially obtainable in the real world with the resources you have in the game, and that worldwide demand for each varies significantly season to season, a lot of things have problematical shelf lives and you happen to be very dependent on trade. There you have it, more complexity than you can shake a ☼glumprong stick☼ at. Would this be very fun? Hardly. It's not just a matter of being a natural born chemical engineer, either: This is just too much for a human being to handle. It's not a good challenge, it's frustrating. You could just pause the game every season and carefully go through your production chains, but you'd more likely just decide to give up in frustration.
Lastly, there isn't any ideal complexity level, because what's too complex for one player may be too simple for another. And then you have people preferring different kinds of complexity in different areas, so... Yeah.
My solution? Make it scalable. One of the things I love about DF is that if you're confused by a feature, you can simply ignore it. Is tracking everyone's material preferences too much hassle for you? Just use whatever material is at hand. Your entire fortress won't descend into a speedy tantrum spiral simply because some lout at the tannery didn't like his left sock being made of silk as opposed to leather. Except if the fort was mismanaged to the brink to begin with so that the sock was the straw that broke the camel's back... But then material preferences aren't the problem, mismanagement is. Also, if anyone wants to care about materials, they can make sure everyone has their preferred items and they will be rewarded with happier dwarfs. It's ok if you don't want to bother with that particular side of the game, but it's not pointless to use it either. It's the perfect gameplay element! I don't think anyone has complained about the material preferences.
Unfortunately, you just can't ignore everything. You can't make workshops randomly and hope they are the right kind, it's just not gonna work like that. There's a fundamental barrier between one or two kinds of generic workshops and what we have now, or anything more complex. You can't seamlessly slide back and forth between the two, the in between is much more ambiguous and chaotic than the material example, and would require too much extra coding anyhow.
But why not give the player a choice? Introduce two difficulty levels, which can be selected at embark, or at world design (in case different difficulty level forts end up being incompatible) or at the first game launch (to be written to init.txt). Make the first one a "DF-Lite", very simple, intended for new players, with very few types of rock, ore, workshops, jobs, and everything, so that you can figure out most of the game at a glance (interface aside). This would also be the perfect place to implement a tutorial, if that ever happens: When you first, say, recruit a dwarf into a "melee soldier" (supposing there are fewer soldier kinds in the easy mode) you get a dialog briefly explaining the simpler military system of the easy mode, and hinting at how the regular mode has more complex soldiers with detailed weapon types and effects and so on. By the time the player is bored with the easy mode, he will be expecting most of the complexity, although in a very general sense and therefore will not be robbed of the experience of figuring everything out.
Creatures, materials and smelter reactions are already moddable, I'm not sure about jobs and workshop types and other things but they shouldn't be that much more difficult. Essentially, the easy mode will be a mod of the game with most stuff merged into single entities. Metalworker(smelting and forging), woodworker(carpentry), stoneworker(furniture, mechanisms), kitchen(cooking, butchering, fish processing, milk, potash) and craftsdwarf (cloths, non-iron/wood/stone crafts, bags, leather, jewellery) doing everything the current workshops do, for example.
If the gap between regular and simple ever gets too much, it can be split further into simple, regular (what we have now), and complex (a hypothetical future version with tools, wheelbarrows, room-workshops, getting food from trees, vassal cities, and everything else)
Of course, this doesn't necessarily fix the "perfect fort" problem. Again, add difficulty levels. But these would affect only things such as happy-unhappy thought balance (and hence likeliness of tantrums), food effect (so that dwarfs get hungry faster), combat multipliers (you lose "mirror matches" on higher difficulties), trading prices, skill increase rates, skill efficiency (less harvest, slower crafting, lower quality), invasion frequency and intensity.
You could just stick the second difficulty option into into the options screen, or make the player select it in the embark screen. It could even modify your embark options. You could even add a sandbox option which lets you take anything you like (including very expensive things and more than 7 dwarfs) or an "impossible mode" which restricts items you can buy.
The complexity option would most logically be decided during map generation.
You could argue, of course, that difficulty is built in in the form of embark sites. But I doubt most newbies are frustrated because skeletal whales walk over to their wagon and eat everybody every other embark. It's not that it's hard to figure out how to defeat a dragon (if anything that's too easy- just build a long corridor of stone fall traps or use any number of the other "dirty tricks"), but it's hard to figure out how to defeat or even attack anything at all. The skeletal whales don't kill you, not being able to figure out how to make the farmer farm and the cook cook in the beginning together with the eventual starvation is what kills you. And kills a lot of the fun for you.
Yes, there is a wiki, but a wiki doesn't make the learning curve magically vanish. It's just a workaround like Dwarf Foreman- it's great to have, but would anyone object to a future update making it unnecessary?