Things Picross devs do that grind my gears

I've seen this happen in incorrectly-programmed picross games where the game checks the state of the grid against the solution the puzzle was generated from rather than the generated hints that the player is using to solve and check the puzzle, which results in puzzles with multiple solutions being impossible to solve because legitimate solutions are rejected if they don't match the arbitrary "one true solution". What some of these devs don't seem to realize is that it's very possible for a nonogram to have more than one solution.

Take this 4x2 puzzle for example:

X X
 X X

The hints generated by this map look like this:

     1 1 1 1
1 1  . . . .
1 1  . . . .

It has two valid solutions:

A.
     1 1 1 1
1 1  X . X .
1 1  . X . X

B.
     1 1 1 1
1 1  . X . X
1 1  X . X .

Both solutions satisfy the win condition (that the grid state matches the row and column hints), but only one matches the source map. A proper picross implementation will accept both A and B. An improper one will reject B. Unfortunately, about half of the 30+ picross games I've played will reject B. Those same ones are usually also keen to make other inappropriate game design choices that are absolutely horrible for picross, like penalty systems, which backfires by, you guessed it, exposing part of the solution. By the way, that's also a way you can instantly tell it's in the "Reject B" category. A game that analyzes that hidden information to tell you when you mess up is also using it to check if your solution matches, because in the 4x2 example, you'd be penalized for marking (0,1), even though technically all 8 of the squares are part of a valid solution. Even if it checks the win condition properly, the penalty system defeats the purpose by enforcing the "one true solution". And as if that wasn't enough, by telling you "NO! THAT'S WRONG!", it's exposing part of the solution. Now you know that the square you clicked is absolutely not part of the solution. So you can mark it with an X on your next run. But it's not information you should have had to begin with. So you're left with a dilemma: either stubbornly pretend that you don't know what's there until you come to a point where the puzzle is grid-locked, or mark it with an X and accept that the puzzle has just been ruined for you. It's shitty design all around, and this is what happens when game designers don't think about game mechanics before they implement them. Sound familiar? Like the vestigial lives systems left over from pay-per-play arcade games that have only JUST begun to finally go away in the recent decade?

Puzzle designers: don't ever do this! The puzzle needs to be checked in a way that the player can realistically double-check, so that there is never a WTF moment. If the player is left scratching their head saying "what did I do wrong? Is the game broken?", then you have done something wrong. Either the error isn't clearly being conveyed, or it doesn't exist and the game is actually broken (which is, in fact, the case if your game rejects B). Either way, it should be fixed.