So, you'd think the finer points of "inventory management theory" would have been long hashed out by the games development community, and you'd be wrong.
Gamasutra has two articles; one in 2010 and then another in 2015. That's the opposite of a hot topic.
"Inventory Management Sucks!" is a short reminder on what the player probably expects out of their inventory system. Too short.
"Loot Quest: From Ruminations to Release" gets much more into the issue, from the point of view of the choices one makes building the inventory system, and how it affects the game mechanics. Those guys make explicit and careful choices, because "a streamlined inventory was seen as one of the highest priorities" for their "heroes-with-equipment premise".
There's also "Designing an RPG Inventory System That Fits: Preliminary Steps" which is a good overview of the historical 'categories' of inventory management, and reviews the major systems seen in games over the last decade.
Of course, you can't discuss inventory systems without referring to the grand-daddy of the genre, World of Warcraft. Although more as a cautionary tale, many would say.
Something that's taken for granted with all these 'Inventory' systems is that items are scarce. They are games, after all. Items in games are intended to be rewards, and often consumable. This can lead to entire 'virtual economies' for 'goods' that are really just one entry in a database table somewhere, linking a pre-defined 'item' object into your inventory space.
When you 'craft' an item in these MMOs, what you're doing is collecting a bunch of predefined database tokens which you exchange (via server calls) for another predefined database token. You're not making anything new. Even though it can sometimes feel 'unique' because stats are randomly rolled or cute names are chosen from a big list.
Remember how innovative it seemed when, in Diablo II, you got access to the ability to name one item? With your own name, but still. That mechanic alone resulted in entire generations of characters being created with names that looked good on a sword.
And yet, the amount of time and effort trading these game items has been incredible. Big companies like Blizzard have had to face the question of whether allowing characters to transfer items in-game will create an entire secondary 'trading markets' and whether they allow that. (Or, since you can't stop it, whether they banhammer the players they catch, or try to set up their own market and get a slice of that pie.)
As you might know, I'm writing a game-ish VR environment thingy called "Astromech". Think of it as a level designer. You place geometry, define what the sky looks like, etc.
I found I was building an "Asset Manager" for all the content that goes into a level. And because Astromech is a 'world builder', I like the concept of just picking up 'items' in one level, putting them in your 'inventory', and then dropping them into others.
That means that, in astromech, Items don't come out of a predefined set. They start as directories of asset files, from which visible things in the level are instanced. They are 'program scripts'.
That makes "item crafting" in Astromech a whole new thing. For a start, items can be created by just dragging a filesystem directory into the inventory screen. The directory is rifled through for metadata and assets, and you get a new inventory item! Want a different item, then edit the files and go again! Want to change the icon or title or comment, then do!
|The beginnings of Inventory Management in Astromech. |
Damn, where did I leave that dataset of all known asteroids in the solar system?
That's a concept most games go to extreme lengths to avoid. Because in a competitive system, you don't want other people arbitrarily increasing their power by modifying their tools. But in a co-operative system about building "virtual machines" to solve real-world problems, you actually want the other "players" to be all that they can be.
Because if they build some incredible item that detects bird-song in audio streams, or reconstructs 3D scenes from multiple photos, (using the DSP 'components' that Astromech provides) then that becomes something they can share. Something not originally programmed into the 'game' by me.
How? Just walk up to their avatar in VR and open their 'shop' of public items. All the tropes of inventory trading in MMOs are just the accountancy of database management dressed up in a pikachu suit.
This allows other niceties, such as users storing the items in their own local machines, and doing direct peer-to-peer sharing without central servers. "Duping items" is assumed. Some are just URLs.
There's a LOT of little details to work out, but the paradigm of item sharing in MMOs is far more friendly than file sharing systems like dropbox. I've prototyped the issues now, and I can see where it's all going.
Minecraft introduced "item transforming items" like furnaces and workbenches. I'm going a step further, to "item compilers" which effectively craft unique pieces of functional 'software' from source components. That's what an Astromech item is.
Astromech's real job is to create a space where all these "virtual devices" can play together in a shared 3D environment, and connect with each other, while protecting the user's browser from malicious code, and devices from each other. (Because, this is all in a web page!)
Remember, the secret is to bang the rocks together, guys.
Towards the end, after the Tested team has played most of the mini-games, the table in the hub room fills up with devices they've 'collected' from the various games. Balloon inflaters, bows and arrows... and they begin to delight in the way these tools can be played with and combined, just for their own sakes. And unlike most games, the tools aren't purely destructive.