...is really boring. Or at least it is in WoW. I haven't really experienced it in any other MMO, though I've read around a bit about it. I know it felt pretty lame in NWN and Dungeon Siege II, though part of that I blame on other things, such as an overdose of possible ingredients, making the only challenge finding them.
I sent Tobold a request to talk about his ideals for a crafting system in MMOs. He promptly obliged. He also included a link to another post by Ixobelle about his own ideas for a better crafting system. I like some of Ixobelle's ideas, though I worry that they could go to far. He mentions A Tale in the Desert having guilds creating 800 billion adobe bricks one by one to build a pyramid, which sounds absolutely insane to me, though I suppose if you're going to build something on that scale, the cost should be on the same scale.
I'd like to list a couple of thoughts and ideas I've gotten from their posts on crafting. It's an interesting topic, since I have such limited knowledge about it. Most of what I'm trying to figure out comes through imagining what the outside effects of various things could be. Forgive me in advance for using WoW references, since it's the game I know the best, and one that most people know well enough as well.
The act of crafting something should definitely be non-trivial. This is so obvious that it shouldn't need to be explained. Basically, this aspect of the game should be treated as an opportunity to build a different way to play for people. I'm thinking of the minor stir that was produced when Popcap Games made a Peggle mod for WoW. Clearly, even implementing a simple but engaging minigame would have a better return than implementing a button and progress bar. At the very least, this would give players something enjoyable to do in game that isn't the same as the normal way of playing, and doesn't bore them.
The other part to this is the whole leveling thing. In WoW, I think the current skill max is 450. When you think about that, doesn't it seem a little ridiculous? 80 levels is quite a ways to level in the game as is. Why do you need to do 6 times that in crafting skills? Because the activity is trivial. If you remove the triviality, it becomes less necessary to repeat the activity so many times. Imagine, instead, only 20 levels to crafting in WoW. For the first level or two, you make practice junk. Then the skill levels from 3-10 involve making increasingly valuable consumables. Things like armor patches, bullets, wizard oils, and better ingrediants for higher-level items. Then for skill levels from 11-15, you make increasingly more useful uncommon, then rare items around 15 or so. Finally, 16-20 lets you make rares and even epics, with 20 perhaps allowing the creation of a legendary.
Along with this shorter leveling curve, it gets actually harder to craft the upper-level items. Making an epic weapon might require 5-10 minutes, or even a half hour of actual engaged play. Remember, this isn't a half hour of watching a progress bar; it's a halfhour long puzzle or something similar. Perhaps it requires a lot of work on making the high level ingrediants required, and you might need to have an excess of ingrediants standing by in case you lose an ingot to a mistake. Imagine a puzzle where each failed attempt at a certain point results in the loss of a single ingrediant. And of course, you can only increase your level by crafting an exceptionally hard item at your level.
Progression in crafting shouldn't require progression in another part of the game. You shouldn't have to raid or pvp to get the best recipes. However, that doesn't mean some recipes can't be acquired through those and other means. But make them tradeable. By making the challenge of crafting meaningful, you no longer need to make the barrier to making high-level items attendance in a raid.
This is difficult to me. First off, how many do you really need? Where's a good balance between realism and enjoyment? Remember that while we want crafting to be a meaningful challenge, we don't want it to be an "elitist" challenge. I'm of the opinion that having a huge number of items and ingrediants and resources to keep track of is similar to having a huge set of stats and mechanics to keep track of and theorycraft about for raiding; it may work, but it also has the potential to be unnecessary fake complexity designed to make things opaque. The more transparency you can have (and still have a fun and challenging game), the better.
Resources are tied into trade and exploration. I'll be focusing on exploration in depth in another post, so I'll address this very briefly. The world needs to be dynamic in it's shape. If roads are sometimes destroyed, and mountain passes change, and forests randomly generated, this will make the search for resources and trade paths more interesting than "will we or won't we be ambushed during this trip?"
Back to numbers, I think it'll partly depend on how many professions are in the game. There'll be progression in resource quality in the type, but I don't like making old things like copper and iron useless after finding adamantium. Certainly we can find uses for them. And even if that seems realistically hard for metals, what about herbs? It's reasonable for some compounds and ingrediants to be useful in all sorts of potions, not just simple ones.
Quality differences are useful though... though a higher quality resource shouldn't simply be available in a higher level zone; that's a fake rarity. Make it actually rare. Then when it's used only for epic items, the price is justified, and part of the challenge in the crafting is in obtaining sufficient ingredients.
One person commented with an interesting point, I think in Tobold's post: make everything consumable to avoid flooding the market with items. I agree that flooding the market is a serious danger. I'm reluctant to make even epic armors and weapons actually disappear after some time, or worse, after some amount of use. That starts impacting the combat game negatively. If it costs me enough to have to replace my epic weapon, I'm not going to use it.
On the other hand, it might keep down gear inflation, which I like. It might make a meaningful decision out of deciding whether to use that giant +20 Sword of Godsmiting when ambushing a simple caravan, or rather using some more common weapon. Still, I worry it could make PvP into a negative-sum game.
So perhaps a time-related degradation in one's gear? And maybe repairing and maintaining epic gear could be a task that doesn't require a master crafter to carry out, nor require as many ingredients. Plus, remember that if you only have to level up 20 times to become a master, you're not going to have to flood the market with items anyway. And most of your economically-beneficial crafting may only involve things that require skill levels around 10.
One last note on this, sort of. Having crafting take a sizable length of time means transactions between players will feel more risky. It's one thing to give ingredients to a player and watch him craft the item on the spot; it's another thing entirely to give them to him and watch him walk away with the expectation of getting the item in a half hour, or even the next day. That takes more trust. This can be addressed two ways that I see: one, you can ignore it, causing advanced crafting to be confined to trusted guildmates more often, and maybe allow players to build reputations. Two, you can implement some sort of contract system that allows you to keep possession of the items even while they're being crafted with, and force the crafter to give you the item or ingredients back somehow. Well, you could create some tools to allow players to check on each other's reputations when it comes to trustworthiness.
In the end, I think the task of implementing crafting can be described as creating a fun and engaging game mechanic, then intertwining it with the rest of the game, without making it require participation in the rest of the game, or having an adverse effect on the rest of the game. It needs to be synergistic, beyond just making the world feel more alive; it needs to make the game easier for the populace, or at least break even in that respect.