The title says it all. I have posted two different Tribond puzzles (puzzles that ask what three disparate elements have in common, with the answer often containing some little trick). The puzzles are here and here. One of the puzzles was well received, the other was not. The poorly received one elicited a comment or two about how this type of puzzle should not be on topic. So, I'm trying to get a consensus: are Tribond puzzles on topic?
To provide the opposing view to Devousi (which is actually an underlying agreement), I think that tribonds are most definitely on-topic...
...however, because you're effectively limiting the design space of the puzzle to three unique pieces of information, and single words at that, it's going to be very hard to produce one that escapes being too broad. In other words, a tribond is a valid puzzle form, but the quality barrier is probably going to be extremely high for an individual tribond to be considered acceptable for the site, compared to other puzzle forms.
As a very concrete example, you discovered the issue yourself when you commented on one solution to your second question as being "a valid response, but not the one I was thinking of". This is the essence of "too broad", and I think it's going to be very hard to have a tribond where there aren't several different, legitimately correct answers.
It seems that the only ways to prevent a tribond being too broad is to:
- Include a bunch of additional caveats to artificially restrict the solution, which saps the simple beauty of the puzzle by effectively stopping it from being a tribond
- Come up with an exceptionally clever tribond, where one answer is clearly better than any possible alternatives
I'd say they're highly discouraged if not off-topic altogether: they are nearly always too broad and they are low-effort to produce. They can be used well as components in a bigger puzzle (eg several layers of tribonds with some thematic final answer) but by themselves I'd say they're unacceptable.