holdalls, inventory manglement and suspension of belief.

Continuing the discussion from let's play: Inform 10's Basic Inform and Standard Rules (SR: Variables and Rulebooks 4):

I think that this deserve a separate general discussion.

In a certain sense, the holdall is an holdover from the days of not only limited inventories, but also locked out locations (and its content); players tend to move everything can be moved from map/containers to inventory (hence the birth of “red herrings”, in the case serves later (cfr. Rules 3 to 5 of the Bill of Player)

Keyrings are specialised holdall, if one think on; and in general, creating a container with infinite capacity but can take only specific kinds/class is the basic generalisation of the keyring concept. So, multiple specialised carryalls can be implemented, esp. if they make sense in the story (e.g. a trophy case which accept only throphies inside)

Now, today’s stories refrain to cruelties like locking the player out of map section, so an interesting alternative is implementing PC’s memory; if the need of tool raises, and the object is already seen (or even moved in another location), the PC remember where it is; this minimises the need of an holdall.

Narratively, is not that difficult reining indiscriminate grabbing around, giving sensible reasons in the object’s “can’t take” message, which has the benefit of encouraging players to role-play the PC; in this narrative layout, red herrings take a more solid and not gratuitous role, literally offering more food for thought in puzzle-solving.

and so on… the need of an holdall can be removed with an, IMHO little, application of creativity and ingenuity.

Best regards from Italy,
dott. Piergiorgio.

Thank you for making a thread about making IF!

I think you’re right. My feeling is that large inventories have become less common in games (though there are some big, successful exceptions). It does tend to make things more credible (and manageable for players and authors).

In Repeat the Ending, the main character has his clothes, car keys, and wallet, and Marbles—a cat—carries nothing at all.

I do have a protagonist in a WIP that carries a large bag because… he’s just a big bag kind of person. He doesn’t actually put a lot in it.

It’s also true that players probably don’t worry about being locked out as much anymore. That was once business as usual, but it’s fallen out of practice.

My take is that, design-wise, inventory limits served a few very important purposes in early games (entirely separate from mimesis).

One, in “logistical” games (games where you have some fundamental limit, like a lamp with limited battery life, and you have to not just solve the puzzles but optimize your solution to fit within that limit), it’s an easy way to make those logistical puzzles more interesting. You don’t just have to optimize your movement from one puzzle to the next, you have to optimize which things you’re carrying at each point to solve the different puzzles and also cart your treasure back. For all the annoyance, the inventory limit really is a core part of what makes Adventure, mainframe Zork, and Acheton work.

Two, in systems with not-very-good parsers, it’s often easier to say “the player can get past X only if they’re carrying Y” than “the player needs to use Y on X to solve the puzzle”. The Phoenix and Scott Adams games, with their two-word parsers, made heavy use of this. If these games didn’t have an inventory limit, there would be no real thinking about the puzzle—you’d just be carrying around all the relevant objects at all times and get past the puzzles automatically. With a limited inventory, the thinking goes, you’ll need to figure out which item is relevant and bring that one along explicitly. (In practice, many of these puzzles still end up getting solved by accident.)

The third purpose—mimesis—came later, once two-word parsers and lamp timers had fallen out of fashion. And that purpose is really the only one where having a player’s holdall helps. Personally, I think it’s kind of a silly reason; if my carrying capacity is six, that doesn’t mean I’m holding six things in my hands at all times (even while unlocking a door!); presumably I’m putting them in my pockets and such. So why can’t I be putting things in my bag in a similarly implicit way?

(Nowadays we also see a fourth purpose, as featured in Miss Gosling’s Last Case and The Wise-Woman’s Dog: a very small inventory limit, a single object or one object plus one spell, lends itself very naturally to puzzles about getting things where they need to go. The Bat uses a very small inventory limit in a related way, forcing the player to run back and forth across the map to match the player character’s mindset.)

I actually really like keyrings, unlike holdalls, and the reason is that you can still use a key while it’s on a keyring, rather than needing to manually take each object out of a holdall before using it. The main purpose of a keyring is to keep all your keys grouped together in > INVENTORY output, and that’s a purpose I very much support.

Though thinking it through, D (again RTE) did have a hard limit of one “carried” magic, and that limit was important for gathering a lot of the game’s “collectibles”. So maybe it’s more accurate to say he could carry one practical item at a time, and that the limitation was mechanically significant.

I think the original intention of a “holdall” was to hand wave the game convention of the player carrying around a bowling ball, six potion bottles, five books, four loaves of bread, three hula hoops, and a birdcage in their hands. Basically “you have luggage to carry all this, don’t think about it.”

Many games use it as an abstract concept (the coatrack is in the holdall backpack, don’t worry about logistics…) or may make a distinction - there is an Inform 7 “bulky items” extension which sets up the expectation that a player can only pick up an item tagged as “bulky” like a television set but only if they drop everything else first, which is a bit more reasonable acknowledging logistics. There are also extensions that put limits like “you can stash most items in the backpack” but limit the player from putting something like a coatrack or furniture in the holdall.

It depends on the author’s intention. In many cases you don’t want to cause friction and it’s okay if the player is carrying fifty items they wouldn’t normally have enough hands for. For a more complicated game logistics might be a gameplay element.

One unique example is the Jansport backpack in Cragne Manor which has 85 minigames worth of inventory, so the backpack is the game going “here’s how you’re carrying and sorting everything - don’t think about it” and is basically a meta-gameplay concession to the player since the backpack has multiple zippered pockets to organize items - stuff you’ve used but don’t want to drop can go in a specified zipper pocket to separate your inventory into “piles”. Even if you’re storing a crystal ball in a side pocket of a Jansport backpack, it’s a “don’t fret about the absurd inventory, this is to help you” concession.

See, I don’t disagree—this is the mimesis argument—but I think it’s not a good solution to that problem. The main effect of a holdall (as opposed to a larger carrying capacity limit) is that the player will see a lot of “but you’re not holding the X!” messages, since the game automatically shuffles inventory items into the backpack but can’t automatically shuffle them back out again. If the player is manually managing their inventory, they can choose which things to stow and which things to keep in their hands; if the game is doing it automatically, it’ll be an unpleasant surprise every time.

I’m making a stupid IF with survival elements and systems and theres a limited inventory at the start, but early on you find a bag of holding, cause it’s usually not a fun element, even though im jamming in a bunch more frictional possibly not fun elements (hunger, bladder, etc)

when i first played D&D as a first grade the first thing the DM, who a 2nd grader, would do is let us open a treasure chest and get a bag of holding….

What inventory items actually do in game is to increase player ability. You use items to solve puzzles, like keys in locks, or to aid in building the narrative, or both.

Most games restrict player progression by not giving them all the items right away. This is why many games kind of end up being fetch quests: go get the thing, so you can go get the other thing, so you can use the final thing on the whatever. This is entirely how metroidvanias work. If you start with final thing, there is no game. Restricting player inventory is kind of frustrating because you’re actually restricting the player ability: you need 6 items to complete game, but can only carry 3. It becomes an experience somewhat like the farmer with the fox, chicken and bag of feed crossing a river; management becomes a puzzle in itself. In some games, rarely, this can work: I’m preparing for an expedition into this game area, and I know I’ll need a certain collection of items to succeed, likely from previous playthroughs/deaths.

So, does it affect the game’s believability to just have bottomless inventory? Yeah, maybe, but most games aren’t all that believable anyways.

It’s going back a quarter of a century already, but my game Sunset Over Savannah had inventory limits just because it made diegetic sense to me at the time: you’re on a beach wearing bathing trunks and a t-shirt, so you can’t carry much. Technically this made inventory management an element of several puzzles. But more recently I stumbled over the Club Floyd transcript of the game, and a thing that struck me was just how irritating the inventory limit was. I think you can debate mimesis all you want, but at the end of the day the only real question is: is the game fun? Unless it’s part of a logical core game mechanic, or necessary for a puzzle(s), juggling limited inventory really is not fun. So why create friction? The magic holdall is an easy, quickly understood convention. Does it require suspension of disbelief? Sure, but we’re playing games here.

This is easily solved with a smarter parser that knows you can use anything nested in inventory.

True, but sadly Inform’s (where this discussion started) is not that. I wouldn’t be surprised if TADS had that sorted ages ago, though. Dialog just doesn’t support holdalls (or even carrying capacities!) out of the box at all.

Oh, huh. I’m ignorant of how those systems handle these things, and I’d assumed they woulda been addressed, just because this seems like kind of a bedrock issue. The system I’m working on lets the author determine what an item can carry, whether unlimited or limited by height, width, depth, weight, quantity, object class, or even just arbitrarily by object, and the parser understands that anything nested in the player’s inventory is available for immediate use, and even open containers if necessary. (Though to be fair I may be way more hung up on simulationism than most.)

Sunset over Savannah remains one of my favorite IF… but with the proverbial hindsight, being on vacation don’t give also a digetic reason for having a bag somewhere ?

now, an interesting idea worth of experimenting/fooling: what about an early-game puzzle about getting an holdall or expanding an initially “hold-not-so-all” ?

Best regards from Italy,
dott. Piergiorgio.

That can be a good reason to include one, but the puzzle has to be good enough to justify the annoyance of the holdall for the whole rest of the game. How high of a bar that is depends on the game length!

Whenever I (try to) write an adventure I am struck by an urge to make it ‘realistic’ - which inevitably involves bags and containers and so forth - and then I have to row back as I realise it just isn’t fun for the player. As others have said, containers just create unnecessary friction for the player.

In my game, I did include a container, a greatcoat with ‘capacious pockets.’ This provided a way to solve a straightforward puzzle at the start of the game, by concealing an object in it. It doesn’t feel like a great reason to have to deal with a container for the rest of the game, but it does have other uses, so I decided to stick with it.

The whole issue of removing the friction is creating friction for me, I’ve realised, and stopped me getting on with completing the game. This is because I started to wonder if I could implement a system where the interpreter would automatically check if any object the player refers to is ‘in scope’, i.e. here, carried, on a present surface, in a present, open container, or carried by an NPC. I think it’s possible but I haven’t actually cracked it… And it gets complicated. I kinda wish I’d never started on containers :sweat_smile:

Are you making a custom authoring system? What you are talking about is something you already get with existing systems (Inform-6 and 7, Dialog, etc).

I’ve dealt with this in various ways over the years, in my comedy games. In my first game Goldilocks is a FOX! there is no inventory limit as such, but the PC complains bitterly at being asked to carry around a lot of stuff and when overloaded she sometimes accidentally drops something when picking up something new. There’s no inventory limit in Alias ‘The Magpie’, but NPCs will sometimes make comments when they see the PC wandering around with bulky items. To Sea in a Sieve has an inventory limit based on bulk rather than quantity, and I’ve developed this further in the as yet unreleased sequel To the Moon in a Microbus.

I’ve thought about this a lot. It’s always bugged me that in early parser games you had to drop a tiny key in order to pick up a ladder. I think bulk or weight based systems are the best solution, but sometimes I just ignore the whole problem and roll with it. I’ve done this in the new (also unfinished) version of Escape from the Crazy Place, which is choice-based but with an inventory system, and the biggest and most ludicrous collection of items you could wish for.

Oh yes, I know. I’m working in PAW. I have been warned not to try to ‘inform-ise’ the system. That is no doubt wise advice! I think to some extent I was just wondering what it was possible (and practical, given memory limitations on the actual code) to do. I think some of my aims are achievable.
I can’t really devote much time to it at the moment, but it’s still in the back of my mind. Hence my need to chat about it :grin:

In the current WIP with a sack, I wrote my own rules to just automatically do whatever makes sense. That included a rule for the player to put items away if the player doesn’t do anything with them for two turns. That last one is more an aesthetic flourish than anything else.

However, in that game, there aren’t a lot of portable objects anyway, because writing custom rule responses for all of them would be a lot of work. I know that doing so isn’t necessary, but that’s my preference.

It seems like there should be some benefit to the player in adding some structure to the inventory if there is a large number of items to carry around. As long as the game is smart about producing items when you ask for them.