Yeah - I think there are specific kinds of designs that increase the risk that a frustrated player will adopt a lawnmowering strategy (parser-like choice games and limited-parser ones especially), but it’s definitely a broader issue that implicates clueing, pacing, puzzle structure, and a bunch of other stuff beyond UI.
Honestly, I’d frame this and put it on the wall of this thread and any past/future threads about lawnmowing.
Exactly.
If you clue correctly, inform the player, and maintain the plan-execute loop (no I will never shut up about that, fight me), then you should never have to wonder about what a lawnmower could do to your game.
One game I can think of in which lawnmowering fundamentally doesn’t work is Suspended, notably when preventing the intruders from replacing you.
Winning the game has a lot to do with timing and the correct interleaving of events, rather than determining “what object do I use on what object?” This is hard and I’ll admit my games don’t work on this level. On the other hand, sometimes you get the ordering wrong and need to start from an earlier save, which, as I’ve contended in another thread, is not necessarily a bad thing. But you do sometimes have to
.
* scraps his game about actually mowing a lawn *
You’re absolutely right. Fuck. Back to the drawing board!
In my defense, I didn’t bring up lawnmowing originally. My bigger concern is how not to prematurely present the solution to the player while they’re still in the “exploring to see what the options are” phase.
I too think choice-based games are more prone to lawnmowering. I’ve played some games where I wasn’t even reading them and “finished” them by blind luck and occasional skimming (just to see if I could). I’ve also played some games where it’s near impossible to progress without critical thinking and observation, yet both roads to the finish line were achieved with just mouse clicks.
Some choice-based games are more like CYOA… and some are bona fide puzzles. I remember really liking A Colder Light by Jon Ingold and felt like this was a great balance of choice-based puzzles and an immersive world. If you have the time to check it out, I’d be curious as to your thoughts.
To me, the specific urge that I’d call lawnmowering is specifically a choice-based game thing—it happens when there’s a specific enumerated set of options, and the implication is that I can exhaust all of them, and probably should exhaust all of them, or else I might miss something important. This comes up especially in dialogue trees.
I agree that it’s fundamentally a game-design problem to solve, but I don’t tend to have the same urge in parser games: the combinatorial explosion of verbs and nouns tends to be so huge that no author expects players to try every combination, while I’ve definitely run across games that expect you to explore every nook and cranny of the dialogue tree to find the important piece of information.
Perhaps the parser equivalent would be “EXAMINE every noun mentioned in the room description until you find the one with a clue in it”?
(To be fair, most of the games that expected me to lawnmower aren’t entirely choice-based IF, they’re some other type of game that uses a choice-based-IF-style interface for dialogue. It’s a classic for a reason!)
What I’m wondering is where the term “lawn mowering” came from. iit’s clearly an idiom specific to this community, but the metaphor is not at all obvious to me… though with how esoteric puzzles in old school text adventures can be, that’s kind of on brand.
When you’re mowing a lawn, you’re just trying to hit every square foot of the grass, not really enjoying it or paying attention. It’s a mechanical exercise where the only goal is to touch everything in as little time as possible.
The metaphor feels quite literal in the sort of game where dialogue options are colored differently after you select them. You rapidly start to feel like the goal is “turn all the options grey” rather than “read some good text”.
The typical way around this problem (Emily Short led the way on this) is by giving the player a varying short list of dialogue options every turn, rather than a single big list that can be checked off. Either allow topics to lead to new topics, or randomize the list and have the NPC force the conversation onward after a few choices. Storylet models are good here.
Yes, Lawnmowering is when you get five different choices, such as questions you can ask an NPC, and there’s no reason not to choose them all in sequence to clear them all out - why wouldn’t you get all the information from the NPC they can offer before moving on?
There is also compulsory lawnmowering where the player may be forced to exhaust all choice options before the story can continue.
It’s not always a bad thing. This is a technique in some dynamic fiction works. PaperBlurt wrote several where you might have 3-5 different choices on a page, and you’re intended to click each one and return to the page. When you choose all the links, the story progresses, or a new escape link will appear to move the story on. It’s one technique to break up an infodump.
It is sort of akin to a simple version of “grind” in a QBN (quality based narrative) where you must do something several times to get a new option (“Take me to breakfast three days in a row to become my friend and unlock the romance option or so I’ll like you enough to give you the frob you need to do this other thing.”) but it’s plainly obvious. Grind is not necessarily a bad thing.
Other ways to combat lawnmowering is to include stakes and a bit of gamification - such as bad or less-optimal choices mixed in so the player must choose with care. Perhaps when questioning a suspect you have eight lines of questioning - three of them will give you valuable information, three of them are pointless banter that decreases the NPCs patience level so if you ask all three of those questions first the NPC gets frustrated and leaves, and two of them are personal questions that will anger the NPC causing them to end the conversation immediately even if you haven’t chosen everything else. Or perhaps the player needs to build up trust by asking several casual questions first before they get access to the important questions they need to ask. Lawnmowering isn’t bad if it produces rewards like more dialogue or tweaks a relationship stat.
This is the type of thing that was somewhat ridiculed in the game Mass Effect where the significant choices in character building amounted to “pet the dog/kick the dog” and there was no gray area. Your only options to character build were “teach the village how to agriculture/burn the village to the ground” and it was hilariously binary and obvious that the only character building was “I’m a villain/I’m a paragon”.
It is interesting to think about the path-dependence, for BioWare but also for the greater RPG industry, that their breakout game was Knights of the Old Republic, where binary Light Side/Dark Side morality makes a bit of sense. Based on that experience, and success, you can see how similar mechanics would get imported into Jade Empire and (more influentially) Mass Effect, even as they hilariously tried to argue that the obviously-evil side was more complex than just being a jerk - and it’s interesting too that they recognized this approach was flattening the broader kinds of choices they included in BG1/2, since Dragon Age, their “throwback” franchise, pointedly doesn’t include this sort of thing.
What is the nature of evil? It assumes many forms, but the common denominator is… [checks notes] sarcasm.
Elsewhere: I think it’s valuable to distinguish between “do everything that seems interesting” and “do everything, no matter how uninteresting.” Usually I associate lawnmowering with the latter, since I’ve never seen it used in a complimentary way. In a Sierra game, that might involve trying to use every inventory item, no matter now nonsensical, on every interactive thing. Revisiting every room out of fear of missing a trigger. That kind of thing.
(Some Sierra games were famously frustrating re: their Ballyhoo time modeling)
I checked out The Colder Light, and I think it roughly aligns with what I originally said about “unusual use of object” puzzles in particular not working as well when choices are presented to the player. At one point, you have to attach blades to your shoes to make skates, but I didn’t have any time to think about doing that myself because the option to do that was listed as soon as I had the requisite components. And the game tries to complicate this by saying you need something to tie the blades to the skates with, but again, it was spoiled somewhat by the fact that as soon as you obtain the amulet it immediately suggests removing the leather thong.
So for that puzzle as a whole, while there was some requirement of actively recognizing that now it’s time to do the thing that was suggested earlier, it still kinda felt like the game had basically immediately told me how to solve that particular puzzle.
EDIT: so, I guess to clearly place this in the context of the thread as a whole, I think there’s a problem here that’s different from just “preventing lawnmowering.” But that’s fine, there are plenty of ways to design around it.
Ugh, so then I can also have social anxiety about video game NPCs. Great.
Part of this potentially stems from the player expectation/demand that the “non-good” paths should provide some unique content of their own, whereas the most reasonable/likely ‘grey area action’ would be “leave the village to fend for itself”: moving along, doing nothing and skipping that piece of content altogether. If you allow that too often, the resulting playthroughs are less “good/ evil” and more “good/ generally good but with only half the content”. (And, in a sense, games do often allow that option, if you don’t take or don’t pursue the relevant side quest.)
Also, when the ultimate ending of your overarching narrative is more or less a foregone conclusion (“you heroically save the world (or galaxy or whatever) from the Big Bad”), that kinda limits the believable characterizations of the PC, so I think an evil PC will usually feel out of place in such a scenario.
Mistakenly replied to Mike and apparently can’t edit that after the fact; sorry.
Yeah, handling nuanced morality in a way a computer can reliably track while having all choices be meaningful and lead to unique experiences is tricky to say the least.
And having responsive NPCs is another balancing act. On the one hand, talking to a dozen NPCs with one-liners in search of the one with a real clue as opposed to uninteresting flavor text or navigating a complex dialog tree for the one clue hidden among the flavor text can be dull, but go to far in the direction of making NPCs uncooperative or outright hostile if you pick the wrong thing can get frustrating. Both techniques are valid, but both can be overdone or poorly executed.
Though, regarding the example of using a leather cord from an amulet to tie blades to shoes to make ice skates, perhaps one way around giving away the answer would be for a choice game to implement a robust “combine two or more inventory items” mechanic where some valid combinations aren’t actually useful for any of the puzzles and the player has to think about what to combine to solve a puzzle. Whether that could be well implemented within the choice paradigm is of course, another question.
I really appreciate you checking it out, Avery. (It’s a fairly short game, if anyone else is interested.)
I completely understand where you’re coming from now and agree. There will always be some level of “providing the answer”. (Which is why I’m against allowing USE in parsers as much as possible.) The actual verb that’s performed can be quite different than what you expect, even in that parser game context. A hammer can have many uses, such as pry, hit, dig, wreck, repair, summon lightning… hey, if Thor can do it, just sayin’.
I still think there are ways to reduce the level of providing the idea/solution, but it won’t go much farther than what point-and-click adventures accomplished, in a way.
And now that lawnmowering is related to the actual topic of the thread instead of a tangent—
When Anais and I started writing Loose Ends, we submitted the first version to Introcomp. Neither of us had written choice-based IF before; I had done parser stuff, and they had done static fiction. So we were learning as we went.
One of the biggest complaints we got was that the more parser-inspired scenes (i.e. the ones I wrote) encouraged lawnmowering too much. There was a segment where you get into the crime scene to search for clues, with links to search the ground, search the dumpster, examine the walls, climb a fire escape, try the back door, and so on. All of them gave you different snippets of information, and one of them gave you a plot-critical item.
I wouldn’t have thought twice about this in a parser game—“look around and examine everything” is pretty standard gameplay—but in this format, it was an utter slog. It didn’t feel like we were offering choices, it felt like we’d written a page of details and made you click links one at a time to unlock each paragraph, which was just tedious and not fun. I think every single reviewer mentioned disliking that particular scene.
That’s why I think of lawnmowering as specifically a problem in choice-based games. It’s not that having those choices was any more tedious than typing EXAMINE WALL. EXAMINE GROUND. SEARCH DUMPSTER and so on—it was much less so, since you didn’t have to guess at those commands and type them in. But the experience was much, much worse, because all the choices were presented for you, and (correctly!) gave the impression that you were expected to mindlessly click all of them with no real planning or thought behind it. There’s no reason not to check the fire escape if the game offers that link, after all!
In the full version of Loose Ends we overhauled all my overly-parser-inspired segments, and for the crime scene in particular, we added a timer. There are eight or so things to examine, each of which gives you a different clue about how the murder happened, but you only have three “turns” (opportunities to examine something) before the cop finds you, so you’re by necessity going to be working with incomplete information as you move on to the rest of the game. Now the choice of what to do actually matters, and it’s fun again. (And if you haven’t done the one that gives you the item when time is about to run out, all the other links disappear to ensure you get that. Crude, but functional.)
@Hal: And now I kind of want to make/play a game where the many uses of a hammer are the solution to everything… and the hammer is imbewed with fire and ice magic as well as lightning.