Excelsior, by Arthur DiBianca, pitches itself as very traditional IF – ‘a puzzle adventure’ – but with a limited verb set.
As this suggests, it is a very bare-bones experience: you climb a tower, solving simple puzzles to unlock doors. The writing is utilitarian and minimal, the character is featureless, and there is no story to speak of. There is slightly more scenery than strictly necessary for the puzzles to work, but only barely. The puzzles are of a familiar, Indiana-Jones-derived sort without any innovative twists or narrative justification.
I suppose it might be making the case that old-school, Scott-Adamsish parsers were easier to learn than sophisticated modern parsers, because they were simpler and didn’t promise much. If so, my basic response to this is ‘maybe, I dunno, but it makes for boring-ass games.’ Or it might just be a coding exercise, or an exercise in puzzle design.
The puzzles are of a kind that wouldn’t work in a multiple-choice-based format. The first requires you to infer that you can walk through the wall, and the third that there is a doorway which you can interact with; an explicit noun list or direction list would make them trivial.
I got bored with this extremely quickly, and abandoned it shortly after reaching the fourth level. As far as it goes, it’s not buggy, and it executes the actual things which it intends to execute, and the puzzles are fairly clear, which saves it from a 2.