Questions on Your MMO Experiences
This is for the rpg project. I dropped it in those forums, but realized that pretty much no one but testers I already talk with are in there. Use as much experience as you can draw from to answer the questions as extensively as you have time for. Thanks!
What constitutes a good money sink and how can it be fairly applied across the board? (Preferably not repairs)
What makes the combat system enjoyable in past turn-based games you've played, including everything from X-Com to Ultima to Final Fantasy?
Have you played games where level caps were either not enforced or ridiculously difficult to reach? If so, how did this experience compare with games where level caps are set and raised w/ new content? How would it affect a game environment where player-verse-player was important?
What are some of the coolest features you've seen from games that could be applied to a browser-based environment (i.e. can be complex but not enormously graphics intensive)?
What are some of the things that you enjoyed that helped build community and economy in previous games you've played?
What's the most freedom you've seen in a crafting system, and how did that play out in the in-game economy?
I don't have time to address everything, but I'll say something about the crafting system question.
The most freedom I've ever seen in a crafting system was in Star Wars Galaxies, and it was an abominable failure, sadly. The issue was not the freedom itself, so much as the fact that the developers seem to have not thought through the consequences of the system.
Let me explain, roughly, how the system worked. Note that it's been a *long time* since I played SWG, and SWG has changed a lot. As a result, I'm making up specifics in places. The concepts remain the same.
Attributes of the character: A given crafting character in SWG had a level of "experimentation" points based on their level in the crafting profession, and recipes based on which trees they had spent points in so far. The set of recipes was not especially large, and since it came from leveling the profession, it was more or less the same for all characters.
Attributes of the recipe: Each recipe required certain amounts of certain material types. For example, an item might require 5 units of one metal and 10 units of another metal to be created. This means that when crafting the item, you must supply a total of 15 units of metal, five of which are of one type, and 10 of which are of another. This could be 5 metal A and 10 metal B, 5 metal B and 10 metal A, or 15 of metal A or metal B.
More advanced recipes might require specific sub-classes of material. For example, you might need 5 units of aluminum, and 10 units of any metal, and an armor plate. For this recipe, you would only be able to use metals that are "aluminum" for the one slot. The other slot could take any kind of metal (including "aluminum" metals), and the third slot would take the product of another crafting, which itself would be composed of some materials.
Each slot would have certain roles associated with it. For example, in a piece of armor, perhaps one of the metal slots would be to produce a reflective surface, while the other would be to act as a shock absorber. I'll discuss this more below.
Attributes of the materials: every material was randomly generated. Basically, there was a tree of material classes, like "Metals" contains "Ferrous Metals" contains "Iron" contains "Corellian Iron", and then "Gobbledygook" would be a specific kind of Corellian Iron. Different classes of materials had different ranges of possible statistics. For example, all metals might have a conductivity stat, all ferrous metals have a certain range of possible conductivity, and a specific material would have a specific number assigned for conductivity.
Just to make this clear: When a recipe requires 10 units of "aluminum", that means 10 units of a single kind of aluminum. If you have 3 X aluminum and 7 Y aluminum, you can't stack them to give them to the recipe. It has to be 10 of the same specific type.
Gathering materials worked like this: Every X days, old materials would go away, and new materials would spawn. When this happened, you would go out with your survey gear and search out concentrations of materials. You'd then set up a mineral extractor at a location with a high concentration, which would collect the material. You'd need to periodically visit the extractor to reload energy and unload material from the hopper. After a while, that material despawns, and new materials are available to survey and harvest.
Animal materials (leather, etc.) were similar, but you'd skin animals for them. The same "the material is available for a while, then replaced with a different one" mechanic applied.
Each new material had *randomly generated* statistics within the range of possible values for the material.
Okay, so this leads to the question of what the statistics were good for.
The crafting process: As I mentioned above, recipes have various slots that need to be filled, and those slots have roles. Choosing what material to use was based on the role and the material's stats. Perhaps conductivity has a really positive effect on the final energy resistance of a piece of armor when used in the reflective coating slot of the armor recipe, for example.
So, you would place various items in the slots, and be able to see what the final attributes of the armor were tending towards. You could then use your "experimentation points" (based on crafting level) to improve the final attributes of the piece. Every time you experiment, you'd have a chance to improve or degrade that statistic.
When you have finished experimenting, you can either make the item or you can make a crafting template for the item. The idea here is that you experiment several times until you get a really good prototype, then you use it to make a template. You then place a template in a crafting structure, along with raw materials, and you can mass-produce the item with the exact experimented statistics.
When an item required crafted ingredients, mass production then required mass production of the ingredients. So you would experiment and create your template for armor plates, then produce a bunch of armor plates, use some of those plates to experiment and produce a template for the final armor, then mass produce the armor using the mass produced plates.
Likewise for the raw materials, of course, since so much was based on their stats. So once you run out of a material, you won't be able to produce the templated item again. Also, as I recall, there was a maximum number of runs any template could be used for.
Okay, so to the heart of the issue: how did this system break down?
The first breakdown was that low-level crafters could never sell their products. Because experimentation points were based on the crafter's skill level, and since experimentation points lead to being able to produce a superior product, a higher level crafter's products would always be far superior to a novice's. Also, there was not *much* opportunity cost to production, so if an item was worth crafting at all, high-level crafted versions were available.
The second breakdown came when people started min-maxing the material statistics and occasionally finding materials that were really good in one statistic or another. These materials would be harvested by guilds building harvesters that would stretch from horizon to horizon, and stockpiled in massive quantities. So, now you could use the metal with the 997/1000 conductivity in that conductivity-sensitive slot of the recipe, which would give the gun near-optimal max damage, for example.
The final breakdown occurred as this process continued, and major crafters first acquired materials that were nearly perfect in each statistic--they would have the perfect conductivity metal and the perfect ductility metal and the... etc. Then, they started building up stockpiles of materials that were perfect in multiple stats at once. (Just because the conductivity in the A slot of the recipe contributes the most to the recipe doesn't mean that the conductivity in the B slot doesn't help, too.)
So, now people were routinely crafting items which were perfectly experimented upon, using perfect materials, and they had enough materials to go on doing this pretty much forever. They would eventually run out of a given material, but it would take so long that another perfect material could be discovered and stockpiled in the meantime.
Somehow, the developers must have failed to anticipate this development, because the practical effect of this was to make items that were incredibly overpowered compared to the game. With weapons and armor, it was hard to see exactly how overpowered, because there was no real base-line to measure against, except PvE. (Taking down multiple rancors at once in hand-to-hand combat, for example.) The thing that made the breakdown terribly terribly obvious to me was medicines.
To illustrate the breakdown:
Characters had three primary stats: Health, Action, and Mind. Being hit would damage each of these bars to some degree—they were all a combination of HP and mana. Using your own abilities would deplete them to some degree. There were secondary attributes for each. One of which affected regeneration of the stat, the other of which affected how much of the stat your special abilities used.
When I left the game, pretty much every doctor in the world carried around medical buffs that would increase these stats to ludicrous levels. For example, a typical character, unbuffed, might have something like 500 points on each bar. So 500 health, 500 action, 500 mind. When buffed by a doctor, these bars would be *3000* points, and in addition, the regeneration was increased to such a degree that it took seconds to heal to full without doing anything, and the cost-decrease was buffed so that special abilities took almost no points to use.
Now consider that my martial artist character could solo multiple rancors at once *without* one of these buffs.
And that's why SWG became ridiculously un-fun, because of a crafting system that was too flexible for its own good.
The ideas of experimentation and randomly-generated materials were really neat. They were not, however, designed with your typical min-maxing MMO player in mind. There are a lot of ways the breakdown could have been avoided: making sure that even with perfect materials, items did not become ludicrous, for example. Perhaps designing the material spawn system such that perfect materials would burn out almost instantly, so that it would be possible to make truly legendary items, but not en masse. Certainly using a system of diminishing returns would have helped, so that resulting item quality was asymptotic towards an upper bound, instead of having a low hard limit.
On the earlier noticed problem, with the low-level crafters: putting caps on how much experimentation could go into low-level recipes might have helped, so that high-level crafters would have an edge on efficiently making low-level items, and could produce items that were strictly more powerful (but required more powerful characters to use), but wouldn't be able to make low-level items far more powerful than any low-level items made by low-level crafters could possibly be.
I think the key lesson here is: Always, *always* assume that the players will find a way to min-max your system. And second: When creating a novel system, *be ready* to make major changes quickly. If SOE had been willing to nerf the crafting system in SWG into the ground and then buff it back up slowly with fixes, it would have raised outrage among the power-gamers, certainly—but it would not have made the entire game trivial and un-fun.
(End wall of text.)
I read this carefully, Hypatia. The system doesn't sound too far off from what I'm working with, so the concerns you raise are really relevant.
I'd like clarification on one thing. You bring up medical buffs, but I'm not positive on what you're saying. Are you saying:
a) That gear on your character completely trivialized content to where powerful buffs were insignificant;
b) That medical buffs were part of the crafting system, were overpowered, but some classes could entirely bypass the crafting system and be as powerful because of their class concepts?
Short form: I brought up medicines purely as an example of how overpowered crafted items could be. All crafted items were overpowered, and combining the use of many crafted items at once was ludicrously overpowered.
Longer form: Imagine that with these buffs would raise a WoW warrior to have 100,000hp, and regenerate 10,000hp/s in combat, gain 10 rage per second, and all abilities would cost 1 rage to use. On the opposite side, imagine that fully crafted armor and weapons would raise the warrior to 98% avoidance, 98% armor mitigation, maxed out resists for at least half of the magic types at once, and a 500dps weapon.
And at the same time, none of the mobs in the game change at all.
Clearly, this means you just mow through the stuff that you'd normally be fighting. But because in SWG there's not really any established idea of what you should be fighting, the problem was not immediately obvious to the players. Sure, you're soloing Krayt dragons (which on reflection should probably have been a minor raid boss level encounter), but it wasn't obvious just how wrong that was.
In the WoW comparison I made above, I tried to aim such that with just one or the other of the gear or the buffs, you would figure the warrior would still lose when trying to solo BT/MH bosses, but with the combination of both, would have a pretty solid chance.
The crafting system (including the crafted medicines) resulted in a massive increase in the power of characters of all types. And all of these varieties of crafted items were available in quantity for practically free.
(The more I think about this, the more I wonder how the SWG devs could possibly have failed to do anything at all about the problem once it became apparent. I can see how it was missed: SWG tried to do a lot of novel things, perhaps in order to shake off the "Oh, it will just be EQ with wookiees" commentary—and having completely novel system for every portion of the game makes it hard to predict what's going to happen. But I really can't see why they wouldn't step in and make a serious corrective effort.)
I'm going to end up posting too much in this thread, I think. I hope I didn't scare away anybody else who might respond.
On another topic: “What are some of the things that you enjoyed that helped build community and economy in previous games you've played?”
First, another SWG note. There were two things in SWG that were, in my opinion, fantastically successful. One was the emotes: There was an extremely large number of emotes that involved character animation, and this really added a lot of variety to socialization, I think. The other was entertainers. The entertainer classes (dancer, musician) provided services that were required after farming all day. Basically: each of your Health, Action, and Mind bars would take wounds as you fought. Wounds were "black" areas of the bars that would not heal on their own, no matter how long you waited. In order to heal health and action wounds, you would find a doctor, who must use a special medicine on you to heal the wounds instantly.
Mind wounds, on the other hand, were healed by sitting in a cantina and watching dancers and listening to music for a while. This sounds absolutely lame and ludicrous, but it turned out to be very community building. There were of course a number of power gamers who pretty much just wanted to get in, get their wounds healed, and get out. But talented entertainers got good at getting people to open up a bit and chat. Basically, if you were open to socializing at all, you'd get into the mood for it when you dropped by the cantina to heal up your mind wounds.
That all went away once people started using AFK macroing to run dancers for their guilds all over the place—cantinas became very inhospitable places, with the AFK bots shouting "PLEASE TIPz ME!" over and over. But, well, it was interesting and good while it lasted.
A less unusual mechanism that helped build community is something I saw in A Tale in the Desert. ATITD was a brilliant but very flawed game. Brilliant in that it was all about building a society in a game—working together to solve hard problems. Flawed in that it was way too much like work at times.
Anyway, the feature that was really interesting in ATITD is that you could be in multiple guilds at once. In the game, you could chat with people immediately around you physically, you could send private messages to individuals, and you could chat in guild chat. There were no other chat channels available—no zone-wide general channels, no trade channels, no user-created channels, etc.
But because you could create and be a member of multiple guilds, you could use guilds as your chat channels.
Now, here's the thing: each guild had a maximum number of members, based on the size of the guild hall. You could increase the guild hall size, but doing so required resources. And, the required resources increased quadratically.
Surprisingly, this combination of features lead to a lot of really interesting social behaviors. For example, in various regions of the world, "regional guilds" were created. These guilds were used to coordinate regional efforts at supplying materials to universities for research, and so on. Because expanding these guilds would require resources, not everyone from every guild in the region could be a member. Instead, guilds would send one or two members to join the regional guild to act as respresentatives.
This was one example of a general concept known as "meta-guilds". Essentially, people really do have an attachment to the idea of a "guild" in terms of a tribal or family grouping. You have the group of people that you identify with most strongly, that you belong to. That's your guild, in the traditional MMO sense. It was also frequently the first guild you'd be in in ATITD: the guild of your friends.
The meta-guilds were groups that spanned these tight-knit communities and brought together people with common goals or interests. For example, people interested in setting up a currency-based economy would join a meta-guild where people would routinely talk about that topic. And that meta-guild would spawn an actual guild that would own property to make the plans a reality (that was the other thing guilds could do in ATITD: own items in the world, so that only guild members could use those things.)
Anyway, the point of all this is: The combination of being unconstrained in one way (you can be in as many guilds as will have you) and being very very constrained in another different way (the number of members of a guild is limited based on the resources you're willing to expend, with very small guilds being very very cheap and very large guilds being very very expensive, while middling guilds are reasonably priced) lead to an ecosystem of communities, rather than disconnected islands. And this was all without any mechanical support: There was no idea of one guild being different from another in the game. They were all just guilds. There were division between different kinds of guilds. Normal guilds (friend groups, which generally shared property freely), regional guilds (geographically based coordination groups, which would take in property and share it out for major projects), meta guilds (common interest groups, which would frequently own no property, except maybe a tiny amount related to the interest: the moral equivalent of a gaming club's collection of board games), and finally working guilds (which were founded specifically to hold property for a specific non-regional endeavor, like running a bank). But those divisions were the ones created by the players—the devs just sat and watched, after implementing the "member of multiple guilds" idea based on player input.
This system of organization created the most vibrant ecology of communities I've ever seen in a game. It was really something to behold. I think other games could benefit from similar models—maybe not quite as much as ATITD did, since in most games there's not quite the variety of things to do, and it doesn't take eight hours real-time to run from one side of the world to the other. But I think that in any game worth playing, there's enough variety in peoples' goals and interests to make an ecology of guilds do something more than a one-player-one-guild model.
Last edited by Hypatia; 02-06-2008 at 10:58 AM.
Okay. I've come to the conclusion that nobody else is going to post, either because I scared them off or because they don't feel like it.
Anyway, today I will comment on: “What makes the combat system enjoyable in past turn-based games you've played, including everything from X-Com to Ultima to Final Fantasy?”
In my opinion, the truly fun turn-based combat systems are very focused around movement and positioning. I’d point specifically at X-Com, FFT, and even D&D 3E as examples of how these factors can make a game fun.
What this focus suggests is that these games put in a lot of room for tactics. In X-Com you've got manipulation of line-of-sight as a big element—preparing to attack as soon as something becomes visible, etc. In FFT, you have movement to high ground, attacking from flanks/defending flanks, and dealing with certain enemies' vulnerabilities or immunities. In D&D, the attack of opportunity model (while flawed) provided some interesting new ways to hinder an enemy's movement.
One big concern I have about a turn-based combat system is: You definitely need to make sure that the system is survivable. Because of the relatively slow speed of turns, it's tempting to make attacks do large amounts of damage. However, if you combine that with added damage from beneficial position, critical hits, vulnerabilities, etc. you can be in a position where things (including characters) die without having any opportunity to respond. Not such a big deal if it's an occasional unlucky event, and you're controlling a whole team. A much bigger deal if you're controlling a single character.
This is especially true when you have multiple creatures gang up to attack another creature. You get one person out in front, and before his first ally moves, the enemies all move—flanking him, and three hitting him at once. That could do a lot of damage.
On the other side, if you make damage too hard to do, fights can become really drawn out in other poor circumstances. For example, if something prevents you from getting advantages from position, and on top of that the enemy is highly resistant to your main form of ranged damage, the fight could take *forever* even if you're sure of winning.
That brings up the other thing that's really nice: Some way to prepare to respond to an action. Both X-Com and D&D have good mechanisms for this, and it can be pretty important to have some way to do it. For example: Prepare to charge up and engage the next creature that appears. If the creature has a chance to move its full movement before being attacked, it could move a looooong way. Simply delaying after your turn is sufficient for many complex things (like "let's both move up together to avoid exposing our flank!"), but there will always be some options that should be allowed to be reactions.
I guess I'll take this one.
"What constitutes a good money sink?"
Unfortunately, imo nothing. Money sinks (other than repairs) are annoying to me because they tend to feel contrived. I can't think of a money sink as a good thing. If you don't want the players to have as much money then limit the amount that drops or something. Please don't make them spend it on something that only exists to make them spend money.
A good money sink is something that is entertaining, but in actuality provides nothing substantial. (For instance, player housing.)
A status symbol, if you will.
Last edited by Taelas; 02-07-2008 at 01:02 PM.
Dovie'andi se tovya sagain - it's time to roll the dice
Can't control inflation without money sinks or without eliminating a currency system. That's the issue, and that's why every mmo has them.
Originally Posted by Uruz
Yes. I am also looking for something that would be more constant, or alternatively, constantly desirable to a large number of players. Player housing will definitely be a part of this, and may be all of it.
Originally Posted by Norrath
Norrath, do you want your name changed to Threkk on the boards, or stick w/ Norrath?
Without quoting your whole post, yes. That was a really solid outline of the issue in turn-based tactical games. It's precisely what came up in NWN as well, with it's semi-turn based single-player format, and the reason that Baldur's Gate II, NWN2, etc., were handled much more enjoyably.
Reactionary moves sound pretty darn nice, too.
While playing WoW I just thought of a good idea. Gambling, you could give away unique gear to the winners, kind of like a lottery or raffle. Or you could have a system where anyone can challenge someone to a game of poker or dice.
PS: I thought you meant more of a waste of money when you said money sink. I didn't consider that it might be something fun.
A good money sink is something that applies itself weakly to poor people, but sinks heavily the gold of those who accumulated.
Trading money for social status: Player housing is a very good example. Another good example would be bought titles. Or the abilities to add a surname to your character.
Another category is trading money for game advantages that don't impact more gold into the economy. (a real sink).
For example, pay for a lessened hearthstone cooldown. Or pay 10g for a shopkeeper to teleport you somewhere. In old-school MUD's, you could by the ability to be immuny to thirst or hunger.
I've played some MUD's with no level cap. You could level from 1 to 230, then "remort" up to 7 times into another class accumulating the skills of the previous, and relevel all it again. After you did this 7 times, you could scale up a Tier, get a permament increase in some base stat all redo all your 7 lives of 230 levels again. You could then keep tiering yourself forever. It took me 3 months of daily play to level my first char from 1-230.
My experience is that in that MUD without level cap is the same I had in NWN, or WoW. People who started playing earlier than you will always be stronger and will dominate you in any competitive comparison. The think is, there is always some factor that is what clearly gives you 'the edge' above others. In games without level caps, that is getting more levels. In NWN it was to farm more and accumulate more gold and get rarer items. In WoW it seems to be a combination of gold and levelling multiple chars to leverage your professions sinergy and money making capacity.
Hope that helps.
Money sinks are a rough one. Back in SWG, I was always dumbfounded by a couple of things:
1) One money sink in SWG was the cost of transportation from planet to planet and shuttleport to shuttleport on a planet. Here's the thing that I found to stunningly foolish: the cost of a shuttle ticket varied based on use of the route. The more people used a route, the cheaper it became.
2) The biggest money sink in the game was *rent*. This was more interesting than in most games, because there was a much larger variety of player-owned structures than is common. Yes, you had player-owned houses, but the mineral extractors I mentioned above also cost rent to keep running. Let me expand on this segment of the economy a little, because even though crafting destroyed the game, it was kind of interesting.
SWG had no NPC vendors or any sort. All gear was either given to you at character creation time, looted from the bodies of your fallen enemies, or crafted. In addition, gear wore out over time. It could be repaired by using player-crafted repair kits, but each time it was repaired, the durability decreased a little. So, all equipment would eventually wear out completely.
Since all gear must eventually be replaced, and all gear is made by player crafters, this actually forms a very very strong foundation for the economy. By tying in to the crafting life-cycle, you can actually really impact the value of in-game money.
(Oh, quick note: Money came from looting the bodies of your enemies and from "missions", which had pretty large payouts and were pretty easy. They made them a bit harder at one point, but the perfect-materials crisis hit just after that, so nobody really noticed.)
So, here's how things broke down from the crafters to the means of production, and to the main money sinks of the economy.
The first link is the crafters actually selling items to people. Most crafters would set up a house to act as a shop, and put displays up and a vendor to automatically sell items. In order to do this, you'd have to use up between 1 and 3 of your character's ten "plots" to build the house (depending on how fancy you wanted it to be.) In order for the house to not evaporate into thin air, you must pre-load it with credits to pay the rent. Larger, fancier houses also cost more in rent. Also, you must load it with energy to keep equipment inside (like vending machines) running. If you don't pay your rent, the building will begin to take damage until it is destroyed (about a week after it runs out of money). If you don't load energy, your building doesn't work. If you're short on cash, you can pack the building up and stow the deed away, but you need somewhere to store everything that was in it. If a building takes damage, newly-added rent is first used to repair the damage, and then used to pay the rent.
So: Rent, the basic constraint here. What other structures are involved in the crafting process. Let's build up from the most basic level all through the process to the vending area described above:
1) Energy is generated using various energy producing units, from solar power, or rare earths, generally. Running a generator to produce energy costs $x per hour in rent. Larger generators cost more per hour to run, and take up more of your plots, but are more efficient. (So, say, you go from $200/hour to $300/hour in rent costs, go from 1 plot to 2 plots, and go from producing 100 energy/hour to 300 energy/hour.) If rent is not paid, energy is still produced, but the building takes damage.
2) Mineral harvesters, chemical harvesters, and hydroponic farms produce most of the resources in the game. (There were also animal products, produced by hunting, which did not require a building to process or produce.) Each of these buildings requires rent to be paid to avoid taking damage, for as long as the building is in place. In addition, they require energy to operate. Energy is only used while the building is operating (but, since rent is always paid, you'd be foolish to have the building placed and not operating.)
2.5) Once you have all of these materials, you can craft. But...
3) Factories. Factories were the buildings used for economies of scale, mass producing identical items. Without a factory, you could produce good items, but since each one is a one-off, you're likely to lose out periodically on experimentation. If you have a lot of material, you might as well make a bunch of stuff. In addition, some recipes needed things like "2 armor plates" in one slot. Those recipes, you *had* to have mass-produced plates to make at all—so sometimes you had no choice but to go to this level.
Anyway, as you may have guessed, factories take up deed slots, require rent to be paid, and require energy to run.
4) Housing: Houses required deed slots, rent, energy, etc. And some specific things inside (like vending machines) also required independent rent to be paid.
Anyway, here we have a pretty rich way for cash to flow out of players pockets into the ether. Players have houses, and pay rent. They also buy things from crafters, who pay rent on many things, and who use energy, which is obtained by paying rent on yet another building type.
I thought this was pretty good, but there's one very specific way it failed to be all it could be: The rent costs for each type of building were fixed. That means that generating x units of an energy resource always cost the same amount. (Note that energy resources could be better or worse, so some would last longer than others.) Running for a given time would always cost the same. Likewise, if you capped energy costs (and note that people were willing to pay energy suppliers 3 or more times the generation cost for energy, so the costs were pretty fixed), running every other kind of building also had a fixed cost per unit time.
It always struck me that it would have been more interesting if this cost was based on supply and demand. Think of it in terms of a tax rather than a rent. The planetary government would be silly not to increases property taxes when there's an excess demand. If the cost of rent per hour was based on the number of buildings of the same type on the planet, then this would do a few interesting things:
1) It would encourage the players to diversify across planets, to avoid having too many producers of a given type on any planet.
2) When a given resource is in high demand, the production costs for that resource would increase, not just the trade value of that resource.
In short, the game's mechanical economy (the money sink) could have responded to the player economy dynamically, by placing a premium on things that players could see a lot of profit in working with. The market would fluctuate until the rent cost for each resource was enough to let the producers of that resource make a profit, but not a 300% profit.
This also might have slowed the crafting implosion of SWG, since building planet-covering collections of harvesters when really good resources showed up would have been much more cost-prohibitive.
I think this kind of dynamism is interesting, and this shows how it could be done in an economy founded on a money-sink model. Of course, the down side is that repairs are still part of the picture—but I'm not sure you can ever get away from them. If you really want a vibrant economy, it has to be based on player crafters. And if you want player crafters to succeed, you need to have a very good reason for other players to seek out their work. As long as items never need to be replaced, that's not going to happen.