Install Steam
login
|
language
简体中文 (Simplified Chinese)
繁體中文 (Traditional Chinese)
日本語 (Japanese)
한국어 (Korean)
ไทย (Thai)
Български (Bulgarian)
Čeština (Czech)
Dansk (Danish)
Deutsch (German)
Español - España (Spanish - Spain)
Español - Latinoamérica (Spanish - Latin America)
Ελληνικά (Greek)
Français (French)
Italiano (Italian)
Bahasa Indonesia (Indonesian)
Magyar (Hungarian)
Nederlands (Dutch)
Norsk (Norwegian)
Polski (Polish)
Português (Portuguese - Portugal)
Português - Brasil (Portuguese - Brazil)
Română (Romanian)
Русский (Russian)
Suomi (Finnish)
Svenska (Swedish)
Türkçe (Turkish)
Tiếng Việt (Vietnamese)
Українська (Ukrainian)
Report a translation problem
Discussions Rules and Guidelines
Report this post
However, "Rock is Rock" teaches you an important tool that comes up frequently in later levels: rules take precedence in the order in which they are formed.
If you form "Rock is Rock" and then "Rock is Flag", the rock will not become a flag because Rock is Rock is an established axiom.
However... [BIG HINT in spoiler]
If you form "Rock is Rock" and "Rock is Flag" in one move, it'll turn into both.
As for Stop and Push...
Stop prevents an object from passing through it. Push prevents an object from passing through it, but allows the tile to be moved. Push inherently possesses Stop, but Stop does not glue a tile into position... the Push rule is stronger.
While it could be said that making "Wall is Push" should cross out "Wall is Stop", since the latter is redundant, this violates the above premise that earlier rules override later rules. It would break a lot of the puzzles.
Just remember this: if something is stop or push, you can't go onto the tile. If you can't go onto the tile, it can't kill you. There's one caveat to this: if an object is both Push and Defeat, and you try to push it when it is stuck, it will still kill you.
Now this is just not true. I tested it with Push vs Stop in the very first level, which is why I was annoyed when I discovered the crossing-out thing was implemented for "Object Is Object." Again, why not for all the other rules? There's clearly a hierarchy of rules at play, because the result does NOT vary with formation order, but the game never tells you the hierarchy. You're left to guess, sometimes suiciding to do so.
In fact, the level you say is supposed to introduce me to this precedence rule actually proves it false. If you break up Rock is Rock, make Rock is Flag, and then re-make Rock is Rock *after* Rock is Flag, the Rock is Flag still gets crossed out even though it's the earlier rule.
And it's not the case that the game "remembers" that there was a Rock is Rock rule first before it got broken up, because in Lake 8 you can make Wall is Jelly and then *later* make Wall is Wall. When you do, Wall is Jelly gets crossed out. It doesn't override Wall is Wall even though it existed first.
Also, that thing you put in spoilers is wrong. I tested it. You need to make a statement using an "AND" to get the effect you describe. Forming them as two separate statements doesn't work, even if the two statements are formed simultaneously. Again, timing has nothing to do with rule precedence.
You can also demonstrate *exactly the same interaction* between Win and Defeat. Go to Lake 8 again. Make both "Jelly is Win" and "Jelly is Defeat." Neither of these rules existed at the start of the level, yet no matter the order you make them in, the jelly always defeats you. It's clear that "X is Defeat" straight-up out-prioritizes "X is Win" in the same way that "X is X" out-prioritizes "X is Y," except that in one of those cases the game is explicit about the priority and in the other your only way to know is though trial and error. Why not be clear in both cases?
Also, how would removing a Stop property while the object is Push break anything? If Push contains Stop, crossing out the Stop rule should have no effect, save for making it clear that it doesn't override Push.
"if an object is both Push and Defeat, and you try to push it when it is stuck, it will still kill you."
GAH! This is even worse. Do you know of a level where I can find this for myself? I checked several but was only able to find ones with both Push and Win, which *doesn't* work like you describe for Defeat. (Trying to run into something that is both Push and Win when it is stuck does *not* win you the level.)
What is the point of this inconsistency? Why does Push prevent both Win and Defeat *except* in a random narrow circumstance where suddenly it lets Defeat through but *not* Win? Does it really break anything to *not* do this? Because this just throws out all predictability. Not only is Push + Defeat randomly different from Push + Win in one random case for no discernible reason, but an object that is Stop and Defeat won't defeat you in this situation, even though in every other scenario Push contains Stop. as you described. So if an object is Push and Stop and Defeat, will it defeat you in this way? And is there any way to *pre*dict the answer, or do I just have to try everything and see what sticks? Because that's not how a puzzle game is supposed to work...
I have beaten the entire game, broke levels by finding unintended solutions and I'm involved with making mods for the game. So some of the things I say might be a bit too eye opening.
---
@Zaq
---
It's a basic principle of a puzzle game to show you a situation that is impossible until you learn something new about it. This is only fair when your options are so limited that you can rule out all of the things you currently know within a few minutes, possibly without even pressing a button or looking at the screen more than once. From there, a player would know to try WALL IS JELLY or JELLY IS WALL given that there are no other options. It leads to a simple "Aha!" moment. Players shouldn't be close minded by this point in the game as they have only just started playing, early enough such that lake could be considered part of the game's 'tutorial'.
The game desperately needs to prepare players for the unexpected and the surprises, and to experiment with almost anything, as the game has quite a learning curve, which it smoothly accommodates with over 200 levels, save for the odd difficulty spike, in which most infamous difficulty spikes occur later in the game. It's partly why the game almost never holds the player's hand and almost never forces them into situations where they physically can't avoid forming a rule.
In other words, the game's tutorial levels consist of: Giving you almost no options so you have to learn something new. I can only think of one level where the game is blatant about the mechanic it's teaching.
---
For lake-7. This sounds like a personal confusion. The DOOR is still STOP, it's just that it's also PUSH. The game never shows that something with the STOP property cannot move, only that something with STOP is tangible and since objects aren't push by default, unlike text, an object that is STOP and not push is an impenetrable WALL and should be interpreted as such from the beginning of the game, as well as the fact that something with STOP can still move via other 'movement functions' such as YOU. (The game expects you to experiment.)
Calling such an object STOP could be a misnomer if you consider TANGIBLE to be a more accurate word, but due to the graphics style, we literally can't change it to that and expect it to be perfectly legible.
When it comes to HOT and PUSH (While BABA IS MELT) it would be even worse to cross one of these out because PUSH doesn't actually overlap with any aspect of HOT or vice versa. If you could stand on it, it would melt you, that's what hot does, but since that's hard to pull off without 'help', you can instead make something else melt, then push the object on top of it, which would cause that object to melt the other object, regardless of any other properties involved.
You can see in lake 8 that a wall being STOP and DEFEAT keeps you from stepping onto it which would defeat baba, meaning that some of the properties activate when you're standing on the object and others activate when you try to make a move onto the object, which is reinforced by the fact that a WIN object always shows you standing on the object in question rather than merely facing it.
... In my opinion, a puzzle game isn't hard enough if you're always certain about what you're doing, that there is never uncertainty. Ideally a puzzle game should have just enough uncertainty at every stage, otherwise it wouldn't be entertaining.
---
@anubiandragon
---
The game never implies this.
You can test this and see that ROCK IS ROCK crosses out all transmutations from rock.
You can do a more complex test to see that if you were to form ROCK IS X then ROCK IS ROCK, before forming Y IS ROCK to turn all Y's into ROCKS, even though ROCK IS X was formed before ROCK IS ROCK, no rocks will turn into X's, showing that the reinforcement of the rule always stops the transmutation.
There aren't 'strengths' to rules. That would be too confusing... STOP simply doesn't apply to the object itself, only objects trying to move onto it, so it stands to reason that if a STOP object could move, it would. PUSH is just one of many movement types in the game.
No. Defeat is an "overlap rule", push is a "move against" rule. It functions as a stop if it can't move, therefore you can never step onto the defeat object.
The way to get a defeat+push object to kill you is to get something else to push the object onto you. There are a lot of ways to do this and this is the actual reason why no rules with backgrounds get crossed out, because 'background rules' never actually override each other.
The intended interpretation is that all of these types of rules coexist and it is up to you to evaluate them and find the distinctions.
---
@Zaq again.
---
Well, there is sort of a hierarchy, but there kind of isn't. It's more that properties don't inherently contain 'disabling features'. Ie; a push and hot object won't melt a meltable baba (assuming it's you), but will melt a meltable flag if you push the object onto the flag. Push disables nothing about stop, stop simply doesn't contain the property of being immobile, just that it's tangible.
The hierarchy is "emergent" rather than "hard coded", so it doesn't literally exist in that sense (though there is a processing order). Through wit, you can feasibly develop an intuition for all of the rules without having to test all possible combinations.
JELLY IS WIN | JELLY IS DEFEAT
For interactions more like these, the game actually 'does' show you the processing order to rules that involve overlapping objects on top of each other. It does it one at a time so you're not overwhelmed, no red herrings so your only choice is to experiment/learn the thing and shouldn't throw multiples of them at once.
WIN is processed after defeat, but it actually still exists and it's possible to use it, so it shouldn't be crossed out. You'll be given tools to do this later on.
I don't see the issue with trial and error in this specific case though. Once you see that things are being processed before win, you'll simply assume that win is always processed last and/or requires baba to be 'safe'/'alive'. That's two ways to interpret it.
Other processing order things try to be as intuitive as possible so it doesn't feel like you're just learning a hierarchy. I could provide an example, but I don't want to spoil later rules.
"What is the point of this inconsistency?" - In case you miss what I say in the previous section, he got it wrong. An object that is push and defeat will get pushed by baba and if it can't be pushed, it'll instead stop baba since push contains the properties of stop and stop also does this. To be defeated by such an object, it has to be moved onto you.
It's exactly like the case with hot and push. You can melt things (that are melt) by pushing the hot+push object onto them and if you're melt, you need something else to move the object onto you.
To see Push not overriding Hot in action, literally just move a Push And Hot object onto something with no collision that Is Melt. It will Melt. You can't actually see Push And Stop directly because Push's effects are cannot be moved onto+can be pushed while stops are cannot be moved onto, but you can observe it indirectly with Not rules. Cancel Stop and the object is still Push, cancel Push and the object is still Stop. Both rules are in effect.
Here's a question. If I make both "Box is Push" and "Box is not Stop," will boxes behave any differently than if I only had "Box is Push"? If the answer is no, then for all intents and purposes Push overrides Stop, and you can let "Box is Push" cross out "Box is Stop" without any worries. Repeat the same question with Defeat and Win.
The idea is to cross out a property statement if that property literally does nothing because of another overriding statement. I get that this will be rare, and it won't happen with things like Push/Death or Win/Stop, but I found two situations where it looks like there is a genuine complete override going on, even if it is just emergent. Why not use the cross-out technique there too?
Also, I'm glad to know there isn't actually an obscure corner case with Defeat/Push. That's a relief.
"It's a basic principle of a puzzle game to show you a situation that is impossible until you learn something new about it."
This is not a "basic principle" of puzzles. Take chess puzzles, for example. You can know all the rules of chess and still get stumped on a chess puzzle. So I'd say it's just the opposite. The core aspect of a puzzle is that you can in principal solve it purely in your head, but the sheer size of that task makes experimentation helpful. Experimenting to figure out the rules can also be fun, but it's an extra thing on top of a puzzle. You can still have a puzzle without it.
Speaking of which, I actually did enjoy the experimentation aspect at first when the levels were more open-ended and there were several paths to victory, but the levels get more restricted as the game progresses, to the point that they've even patched out some unintended solutions. This ruined the experimentation angle for me. Experimenting to find A solution is fun. Experimenting to find THE solution just feels like fake difficulty.
So I guess detailing a hierarchy would be the wrong way to go about it. Maybe something like an in-game lexicon that gives precise information on how the properties work, so one could reason out ahead of time how they'll interact. For example, it is not fully apparent that properties like Defeat and Hot are "overlap" properties and not "contact" properties until you test them alongside Push and Stop.
Honestly, even just showing each interaction clearly in a *simple* tutorial level before involving them in bigger puzzles would help. The issue is when there's only one solution, it's ten steps long, and I have to guess at how a novel interaction will pan out on step seven. This didn't happen super often but it was immensely frustrating when it did.
"From there, a player would know to try WALL IS JELLY or JELLY IS WALL given that there are no other options."
The problem is that the game is set up in a way that will lead many players to conclude that those statements are also not options, not as preconceptions, but as a result of those players' experimentation.
Look, before lake 9, *only* "Object is Property" statements are shown at the start of each level. If a player experiments with other combinations, the *majority* of them (2/3) do nothing. So it's super easy for a player to try two alternative orderings, find that neither of them do anything, and from this conclude that only the combination the game ever shows them is also the only one that works. This is why the existence of transformations needs to be shone explicitly, with Lake 9 before Lake 8. At the very least make Lake 9 *accessible* without beating Lake 8, even if you keep the current numbering.
Now if "Property is Property" and/or "Property is Object" had some effect, then you wouldn't really need a more explicit tutorial on transformations before lake 8, because experimentation would be *much* less likely to lead to the false conclusion of "Object is Property" being the only viable arrangement, but I get that doing that would hugely hamper level design.
This is the idea behind my earlier point about experimenting being fun when there are multiple avenues to success but just fake difficulty when there's only one true way. If you're going to rely on experimentation in a game, you have to set it up so that the experimentation is super likely to yield at least something productive. Otherwise the game will "teach" a substantial number of players that experimentation fails, or other false stuff.
PS: Random idea, but even just adding a useless "X is Y" in the corner somewhere in or before Lake 8 would probably fix the issue. Seeing that statement lit up tells players that "Object is Object" is a viable ordering, which will correct those who were misled through unlucky experimentation. Then Lake 8 could function as a proper tutorial.
If you aren't concerned with how an object would react to other rules being added upon it given its current ruleset, than I'd like to point out Box Is Push and Box Is Not Win functions the same as Box Is Push too, but Push and Win have nothing to do with each other.
The cross-out technique is reserved for when a line of text switches off what another line of text is doing. X IS PUSH doesn't indirectly refer to any other words, so it shouldn't do that. Only reinforcement (X is X) and the NOT word should cross things out because they directly contradict things written by other words, rather than implications.
The game (from what I can recall) doesn't make any weird exceptions like these, so since no other properties overlap, push shouldn't cancel out stop even though they overlap, also because stop does not imply "stuck to the ground", just that it's solid and whenever something is solid and not push, it functions as a wall.
The game should get you into the line of thinking that is centred around questioning how two properties interact and how to get use of one property that's being obstructed by another.
There are multiple types of puzzle games. Unlike snakebird where all of the mechanics are taught at the start and it's up to you to find the combination of moves, baba is you loads its difficulty into a completely different area, you don't know all of the mechanics/properties from the start of the game so the game is all about finding and figuring them out, including what the game allows and what the game doesn't allow. This causes baba is you puzzles to be very easy to execute, but very hard on something called "Insight".
You can later solve baba is you puzzles inside of your head once you get far enough in though and if there was too much certainty, there wouldn't be much experimentation, which is part of what makes baba is you fun... Despite having the ability to solve puzzles inside of your head later on, people mainly seem to complain about later parts of the game rather than the very beginning of the game back in lake.
I don't see any other way for the game to get harder other than having:
- The solution is harder to see.
- Less solutions.
- The solution takes more steps.
- More red herrings.
But the thing is, for a level's solution or solutions to take more steps to solve and be harder to find, unintended solutions 'have' to be patched out, otherwise there would be a shortcut with far fewer steps than the actual solution.
Multiple solutions can be fine, but they can't be unintended ones and they should all be equal to or higher than the number of steps that the level intends, based on how late into the game it is placed.
Experimentation can still reveal the solution though, or even just steps and the best part is, you might discover the steps in the wrong order, only to then put them in the right order in a "late game Aha!" moment.
In my opinion, more frustration comes from red herrings because those massively increase "search space" without really going anywhere. It's where arguably more than half of a certain level's frustration came from, a certain level that got promoted to an extra level and got its red herrings removed and when it comes to levels with too few solutions, those can be circumvented by offering levels in packs so you can solve other levels while you're stuck on other-other levels. All the game needs to do is never throw a bottleneck at you where you're forced to solve very specific levels in order to access the next stretch.
The game doesn't need to. It shows you that some properties are overlap properties, ie: When you win the level, baba doesn't vanish or face the flag, he's standing on it. Other properties are contact properties: You can't move onto a stop. And that contact properties physically obstruct overlap properties, shown in a lake level where a wall is both defeat and stop.
Because it's possible to derive the correct interpretation, straight up giving the correct interpretation seems unnecessary and undermines the amount of testing and experimentation that the players should do all the time.
It would allow players to predict and be far too certain about the outcome of an interaction without learning it themselves or theorising it to be the case based on comparable properties.
I don't recall any levels where this happens. The most nuanced interactions I can think of were introduced in puzzles solely built around them, their key interactions can be tested before you attempt any significant "committing" steps (steps that can't be undone without the undo button), some of which got their "clutter" and "red herrings" removed.
The game should introduce transmutation before reinforcement because the latter is a surprise to anyone who has learnt the former. Since this is part of the game's beginning, many players may misinterpret stuff, but many other players would know to keep experimenting throughout the game because if they don't experience this, then they'll just run into one of the game's numerous other "surprises/curve balls" right when they think the game can't throw such things at them. Levels before lake 8 can be solved with transmutation so there is an avenue for experimentation in this area.
I would settle for a better lake 8 essentially. Anything that shows transmutation without introducing reinforcement.
What I don't want is people to think that the game doesn't throw curve balls because... Honestly, this game does it so often, I wouldn't be surprised if this game was designed with "throwing curve balls" in mind, with padding in between.
Personally, I felt great when I discovered transmutation (and the game's other curve balls) and I wouldn't want the game to straight up tell me that it's possible when I was about to figure it out. If you can think of a way to catch players who miss transmutation without interrupting players who are getting the right idea on the game, I would like to hear it.
There's an even looser way. Have X IS Y be one push away from being formed in a structure where that push is the only way to interact with it. This can be part of solving the puzzle or part of setting up the puzzle... It could be introduced next to lake 8 and function as something to catch players who don't solve lake 8, while everyone who is on the road to solving lake 8 just solves lake 8.
This takes inspiration from your idea of introducing level 9 at the same time as 8 and my idea of trying to postpone reinforcement until after transmutation is discovered by making a different transmutation level.
While I believe the game shouldn't hold your hand, the game also shouldn't bottle neck, so having multiple levels to teach the same thing is perfectly fine. This way it can save some of its surprises without bringing progress to a stand still for more than an hour.
Then it needs to TELL ME HOW THE PROPERTIES INTERACT. I can think about it all I want, but that won't do me any good. I won't know until I try it, and when that means randomly failing what I thought was a solution at step twelve the game becomes tedious.
Interactions are combinatorial. There are literally thousands of property interactions one can build. I don't want to have to find all that crap experimentally, especially not when even finding levels that let you quickly combine the properties is tedious. Like that guy's claim about the odd interaction between Push and Death. I still haven't figured out what level I can use to test that myself, and that's just a single two-property combination.
"You can't move onto a stop"
This is not actually clear from just the Stop property. I know I can't END my movement on a Stop object, but that doesn't mean the game isn't moving me onto it and then immediately bumping my position off the object before it redraws. The only way to tell is to find something that is both Stop and an overlap property, which means possibly killing yourself to test it, and even then you still haven't confirmed whether stop prevents all overlap or just pushes you off the tile before the specific overlap property you tested is checked.
It's entirely possible to code this sort of game in a way where an object that is both Stop and Win can't be used to win while an object that is both Stop and Defeat will defeat you if you run into it. If stop-like properties actually let you onto the tile but push you back before redrawing then they could potentially have different interactions with literally every other property in the game, depending on timing/priority.
"Because it's possible to derive the correct interpretation..."
The problem is that it's only possible to GUESS how an interaction works. If the game actually told me precisely what each property did, then I could DERIVE the interactions, but without details like whether I move onto the tile of a Push object before or after it moves I have no way to actually be sure about how the property will interact with anything else.
I think you ended up making the assumption that properties like Push and Stop never allowed you to overlap with those objects at any point, even between player inputs. It looks like this is true, but it's not the only way they could have coded the game. With this *assumed*, you can derive several of the most important interactions, but please realize that before you actually look at those interactions it is an assumption. Without that assumption literally any priority of properties could be correct. Instead of expecting players to assume this, the game needs to just explain it outright.
"if there was too much certainty, there wouldn't be much experimentation, which is part of what makes baba is you fun..."
I could buy this if they had more alternate solutions and they weren't actively patching the ones they do have. I could also buy it if they had any sort of a "sandbox" level that actually let you experiment freely. But when an hour of searching doesn't yield a way to test a random claim about the Push + Death interaction, I have to say that if the game is supposed to push experimentation and discovery of the rules then doesn't lay the appropriate foundation.
"If you can think of a way to catch players who miss transmutation without interrupting players who are getting the right idea on the game, I would like to hear it."
Multiple solutions is the way to do this. It's honestly a must if you want to build the game around exploration like you describe. If the game is going to require you to discover the rules instead of deriving them, it needs to make discovery a *likely* outcome of exploration, and that means having multiple routes to success. This is how most modern exploration focused games work. Games that claim to contain exploration but only have a single correct path are typically seen as having only fake exploration. Making it seem like there are multiple ways for a player to approach a problem but ha-ha most of them don't work is typically seen as a ♥♥♥♥ move.
"There is a fundamental behavioral difference between an object being Push and an Object being Stop and Push. In order to remove collision from the latter, you'd need to make it Not Push AND Not Stop. Push doesn't turn off Stop. Not Push doesn't turn on Stop. They are separate checks."
Okay, I guess that's a slight error in my question. The way I was thinking of doing the cross-out is that "Box is Stop" would only be crossed out for as long as "Box is Push" is in effect. If you made "Box is not Push" then it would cross out "Box is Push," which would then bring back "Box is Stop."
Basically the idea is to let the player know in advance that Push beats Stop.
Also, there *is* an interaction different between Push and Win versus Push and Not Win, and that's when the object gets pushed onto you. So Push would not cross out Win (or Defeat, or Hot, and so on).
The game has no text tutorials so the best that the game can do is isolate the interaction into a puzzle that consists of those elements and elements that you already understand... There's a reason why the game has hundreds of levels and some of them are excessive.
You don't have to go through 12 steps, just test the step by itself, find a level where you can test it without making "committing moves" or make use of the restart/undo buttons. The game has no speed limit so you can move as quickly as you can tap the arrow keys. (Offers no advantage, just makes the game run much faster.)
The interactions are combinational, but they're possible to categorise as I have outlined before. The game highlights this and will make it clear by allowing you to experiment in relatively small levels, or otherwise make it not relevant until later in the game. The game never requires you to know the exact order to all of the properties anyway, just some of the nuances and the general idea that the properties can be categorised.
Personally, I just don't see any issue with this.
Why can't it just be up to your experimentation? Shouldn't the confusion be part of the experience?
The game gives you the opportunity to test it and make an initial assumption on what exactly STOP does. Including a very early instance of a wall being stop and defeat.
There would be a problem if the properties had a specific set of hard coded properties+interactions everywhere, but they're not, so experimentation, though completely open ended is designed to be as intuitive as possible or offer a very intuitive result.
Without any text based or any form of clear tutorials. The player has no choice but to assume the simplest possibilities first then work his way up... Carefully.
The game is designed with this in mind.
Guessing is the point. Some people will get it, some people won't get it. It happened in the witness where some people got entirely stuck because they got the wrong idea from the rules that the game tried to present without any text.
If it told you what they did, the levels would have to be structured around 'engineering' problems rather than 'intuition' problems. But the thing is, most levels are the latter.
A player doesn't think of the game in terms of code though when he's actually playing through the game. Most people just use intuition and experimentation. I'm not excluded from this group, I didn't start seeing patterns until half way through, yet I didn't have too many struggles before I developed a 'theory' of what was going on because the game doesn't push you to figure that out until later.
I didn't look up any guides either and I avoided looking at letsplayers until after I had total, 100% completion. Though I did look at tons of letsplayers afterwards to see where they struggled and where they didn't struggle.
Finding the solution to the game isn't totally blind or arbitrary, it is possible to figure out, even if you misinterpret something and so players should be given the opportunity to figure it out in exchange for the opportunity to fail. They will try over and over, get new inspiration from other levels or during breaks, until they get it.
Alternate solutions are acceptable and do exist. They're just comparable to current solutions. Some levels are being patched to be easier so it balances out.
Unintended solutions have to be removed though.
If you miss solutions via unintended solutions, then you will fail at levels that rely on you understanding those solutions. Tons of frustration arises from this.
Exploration is going into an unfamiliar area or examining something. It doesn't matter if there is one solution or several because exploration is the process of finding one or more solutions, as well as mechanics in general... Whenever a game does throw one solution level at you, it is about to build on it in later levels. But this is not all baba is you levels.
The witness, a previous game with no text and no tutorial shows that a puzzle game of this type doesn't have to have multiple solutions in all of its levels. The witness is probably even more extreme about it and it was a fine game.
I am curious of one thing though.
What is Hempuli's view on this? Hempuli could settle all of this. Though other views could also offer something interesting.