How should we handle near-duplicates in system recommendation questions?

Question

So, part of the Stack Exchange philosophy is that the question should suggest clear criteria for judging an answer, and ought to be relevant to a real problem the poster is experiencing.

With system recommendation questions, this means specifying what you actually want pretty tightly instead of just casting a wide net that will just lead to everyone just posting their own pet favorites. Because lists are boring while answers to questions are awesome.

But that does mean the question is pretty specific to that poster. What should we do when someone else has similar general requirements but different specific requirements?

Take a look at this question, for example:

What is a good system to play LOTR?

This is a reasonable question as written. It specifies details relevant to what the poster wants, such as "combat-focused" and "low-prop." These are very important for making sure that the original asker's problem is actually addressed.

Buuuut...

Say I find this question by searching for "system for LOTR." What if my criteria don't match the original ones?

Let's say I don't want a combat-focused system (violates OP's requirements), or don't care about system complexity (violates OP's requirements), or really want a system designed for short-form play (not mentioned in the OP's requirements).

Should I ask a new question, also called "What is a good system to play LOTR?" (Will that be closed as a duplicate?) Am I basically out of luck?

Answer 1

Both titles are wrong.

"LOTR" is a hugely broad... thing.

There are questions of explore middle earth with a simple RPG (which is what that other questions should have asked), there are questions of "how can I retrace the steps taken by the hobbits in Lord of the Rings" and there are even questions of "how can I simulate the battle of helm's deep"

Specific questions also extend to their titles.

Answer 2

We should handle near-duplicate sys-recs just like we handle any other near duplicates.

1. If it's so much of a duplicate we think it's really just getting the same answers as a similar question, propose to the OP they close as dupe. Vote to close as dupe if you feel moved that they are really identical.

2. If it's really different, leave it alone - "near duplicate" is not a SE close reason, "exact duplicate" is.

But you are hitting on a more important part of SE philosophy here. A question is not supposed to answer "anyone's" problem - it's supposed to answer the questioner's problem. This is modified with the caveat that it shouldn't be so narrow it'll never help anyone else (even though "too localized" isn't a listed close reason any more, it is still an attribute of a bad question).

In the case of sys-recs, we want to err a lot more onto the specific side of the spectrum. This is because of our experience with bad sys-recs. "Best for LOTR? FATE of course!" cries the FATE fanboy, cut and paste for everyone else. We require more specifics to make the question valid and on-topic for our site. cf. Are game recommendation questions on topic?

In the end, "A system to play LOTR" isn't better or sufficiently more specific than "I want a fantasy game" or "you know, gritty sci-fi." These are not helpful to the OP, or, really, for anyone else (I've read enough "sell me on..." threads on various forums to know I'll never gain valuable insight from them). cf. How to deal with questions that just don't understand the scope of the RPG landscape?

So if you want a LOTR system that's not combat-focused - you post a question with your exact requirements.

If you just want "a more general question for the record" - refrain, as it will be closed as both a bad sys-rec and as a violation of the basic principle of SE, which is that

You should only ask practical, answerable questions based on actual problems that you face.