I am wondering what was wrong with the following first post that I reviewed as "no action needed":
The question asked for a way to install f77, and the answer answered that, giving the name of a script that simplifies is and demonstrating its usage. I'm not an expert on OSX, so it may well be that the answer is incorrect, but as I have been reminded a few times myself, incorrectness is meant to be moderated through downvotes/comments rather than flags.
Let me clarify: I'm not mad I failed this audit, but I'm wondering about the big picture:
- Given the post above, can this really only be successfully reviewed by someone who has tried and tested this particular method of installing f77 on that particular version of OSX?
- "Incorrect answer" is not an option for flagging an answer, neither is it a reason for deletion when you review low-q answers. This would lead me to believe that the review system is not meant to catch those. (Besides, it seems to suggest that it's not sufficient reason to delete an answer in the first place.)
- If reviews actually require full verification of the answer, is the system still reliable? I'm pretty sure that even the C++ experts don't check C++ answers with more than 2 lines of code - they check for plausibility and accept. (which is exactly what I did for this answer: The description is sensible and plausible, the link is live and links to a page that confirms that Homebrew does exactly what the answer claims.)
- The text guiding you for the reviews says: "Watch for hidden gems, non-answers, and spam." Do inconsistencies make an answer a non-answer?