Matthew Devander has an excellent
video that goes into detail about "honest" level design (more particularly puzzle design). The word honesty here essentially reflects an idea that the level designer's job isn't working to inconvenience the player. The designer doesn't create puzzle solutions from contrivance, that is to say, your levels should not be arbitrary assemblages of obstacles to overcome. This is common in old-school point&click adventure games, where often the puzzles essentially boil down to "a series of keys to unlock a series of doors".
Here's
another talk by Jonathan Blow that dives deeper into the role of a designer. I really like his concept of the designer as an explorer of sorts, or a captain of the vessel, rather than strictly the creator. Say for example, you have an interesting and unique idea for a platformer mechanic. There are surely some compelling level designs you could make from that mechanic, but you don't know what they are yet. Rather than trying to "force" a level into existence, you are trying to "discover" it by playing around, fiddling a bit, trying things out to see if you strike gold. Another useful tool is to remember "less is more" - remove extraneous clutter from your levels and try to really hone in on one specific idea. Obstacles that aren't core to the level concept serve only to decrease its clarity or increase that "arbitrary inconvenience" factor.
Great game design is quite the rabbit hole and a very difficult skill. Most discussions on these forums are programming or engineering related, which is certainly important, but I think we developers sometimes forget that every great game needs thoughtful and skillful design principles behind it. I've certainly been guilty of building a game that rested on very little foundation - just a compilation of systems and mechanics I thought were cool without any deliberate design goals or intentionality.
TLDR; Principles I try to keep in mind:
- The designer is an 'explorer' in the space of your game systems -
discover what is compelling about it, rather than trying to force something compelling by artifice.
- What is the player supposed to
feel while experiencing your game/level/story/mechanic? How can you incite that emotion?
- Don't expect the player to find the fun on their own, help them see where it is (without over-tutorializing).
- The simpler and more focused a game/level concept is, usually the better. Simpler does not
necessarily mean less systems/mechanics (though it often does), but less concepts competing for the player's attention.