It depends. A question does not necessarily have to be about one specific software to be considered not broad, as long as it is specific enough in details to avoid five users casting a close vote as 'too broad' (or even 'unclear'). So that is, questions about software can also be closed as 'too broad', or 'unclear'.
Asking about means of accomplishing a specific GIS task (independent of software) is also valid. See for example, my POV in Are software recommendations good candidates for being wiki locked? I have one hypothesis that some people who work only with more standard/developed GIS software has some difficulty to understand that not all GIS (and related fields) tasks can be accomplished in all GIS software.
About the specific question you linked to, If it had been edited after the answer was posted, which was months after that Q remained unanswered, I would not have seen a problem with taken action, i.e., tying question and answer. If it was the case, and I still wanted to post an FME answer anyway, I would say something along the lines "you also was interested using FME, so here is one alternative, etc.". I'd not edit it back if the other answer was accepted, or highly scored (which was also not the case there).
However, it was not like the above; the edit was carried out within minutes after the question was asked. Also, as explained before, don't agree with the justification provided for the edit:
Reduced scope to prevent being closed as too broad
Last, a borderline issue, sometimes I will observe users commenting to OP asking/requesting to state a specific software in question. Sometimes the OP is indeed working with one specific software; sometimes they are not, but give way and pick one; and sometimes OP specifically says he/she is open to work with whichever software if he/she can get an answer that addresses the question. At least the last case should respect OP's will.
And go back to first paragraph, the question would need to be clear, have enough details, etc.
