What are the criteria that make for "better" quality questions and answers for story identification questions?

Question

I spend a lot of time patrolling for duplicates in story identification questions. One issue that repeatedly comes up are the differing definitions of "quality" held by various users when deciding which is the "best" duplicate target to choose. Since "quality" is in the eye of the beholder but the site values consensus-oriented policy, this question is being put forth in an attempt to collect group opinion on:

• which factors are important in deciding the quality of an answer, and
• what is the relative weighting of importance for these factors

The idea inspiring this question is that of a grading rubric, such as might be found in any situation where multiple teachers have to grade student essays based on the same standard. While a rubric does not eliminate subjectivity, it does provide a more objective basis by which to rate the closeness of a particular essay to the group-defined ideal. Such a rubric would commonly leave room for hard-to-pin-down aspects such as style and organization, but would still serve to limit the divergence in independent evaluations by reducing the scoring impact of these subjective factors.

It is my hope that something similar can be built here, if for no other purpose than to allow clearer communication between site participants about what they mean by "better" when comparing two posts.

To accomplish this, I am asking that any participating user please post individual answers making a case in favor of each specific factor that he or she deems important. If another answer is posted that more or less agrees with your thinking for a particular factor, please upvote that answer instead of posting a new one. A comment on the answer can be used to make any necessary clarifications or to make a case against that factor. Similarly, if you agree with a particular clarification or disagreement comment, please upvote that comment to express your support.

The hope is that the resulting answers will evolve to an ordered list showing the level of consensus around specific factors, from highest to lowest in group-evaluated importance. (It may also serve as a guide of how to create questions or answers that are broadly deemed as high-quality.) I would prefer it if nobody places any downvotes on any answers, as this will distort the relevant rankings.

I am completely aware that the question of duplicates in general and "quality" in particular have been thorny topics. This is NOT intended to provoke an argument that casts more heat than light. This is a genuine attempt to find out what kind of consensus can be built so that I and others can reduce the friction around the process of closing duplicates.

Note that this question is not about how to ask a good story-identification question so much as it's about what features are considered valuable by the community for an answer or a post as a whole (both question and answer). It's also intended to produce guidance on the relative weighting of the value/importance of each feature.

Answer 1

There is no general rule of thumb as to what makes answer A better than answer B. Answer A might include all of the "good details" but just be very general and not address any of the OP's memories. Whereas answer B might drill down into the work and be very specific to the OP's description. If you use this list here then objectively answer A would probably come out as the "better" one but I'd argue that answer B is actually the better one.

It's also worth noting that these elements alone aren't what makes a story id answer good, it's the correct combination of said elements that make it good. For example, an isfdb link might end up being the highest voted in the list here but if it is pretty much the sole element of the answer is it a good answer? Almost certainly not.

The correct combination of elements will change depending on the specific question and won't apply generally. Some people prefer having book overs shown in the answer but if the question doesn't talk about a book cover you could put them off of the scent because it doesn't match what they remember because of alternate versions. If that ends up being highly voted and so always included it could actually cause a somewhat negative affect with making OP's think it isn't actually what they're looking for because they don't remember that particular cover.

Instead of trying to come up with some semi-policy on which way to close story id duplicates I'd say it doesn't actually matter policy-wise. It's always going to be a gut call on which one is better, answer A might be better but question Y might have a lot more detail than question X. Choose the better overall post and consider it as a post as a whole rather than the question or answer in isolation.

In other words you can't qualify which individual elements are the best ones because it's the combination of those elements and how they relate to the question which makes for a better post.

This is all a very subjective thing and whilst I understand you are trying to objectively come up with a measure here, trying to objectively answer a subjective thing isn't going to work.

Answer 2

Inclusion of the author's name and story title as text (not pictures)

It sometimes occurs that an answer includes a cover image showing the title and author, but the text of the answer does not "spell these out." This makes them invisible for the purpose of searches (either via a search engine or when using SE's own search function). It's better to specifically include these details as part of the text of an answer, even when they are clearly visible as part of a cover image.

Answer 3

Accuracy of included details

It's not uncommon for story identification questions to have misremembered details cited in the question, and it sometimes happens that even the accepted answer includes incorrect information about the story. If one is familiar with the story in question and/or has a copy available to review, it seems reasonable to judge the question and answer with fewer incorrect details as superior.

Answer 4

Inclusion of quotes from the story itself in an answer

These seem especially valuable for confirming that a proposed answer is correct, as they can be correlated directly with the details (at any level) included in the question. Even relatively short quotes can remind a searcher of their own memory of reading a passage, if it illustrates the author's style well.

Answer 5

Inclusion of a link to a legitimate online version of the story

These seem to be extraordinarily valuable to searchers, both the original poster (who can readily confirm whether or not the story is the correct one) and to others viewing the question (who can read the story themselves if they find the premise to be interesting). A legal online version can often be found at sites such as the Internet Archive, various online science fiction magazines such as Lightspeed or Clarkesworld, or Baen's website.

Answer 6

Inclusion of "lower"-level and distinctive details about plot

Here "lower" detail level means more specific and less general. The presence of these in either the question or answer makes it extremely recognizable to searchers, so long as the searcher in question remembers the same details.

Even if not rememebered, distinctive details make it more likely to be found via search queries and also more likely for a candidate ID to be confirmed in some cases (e.g. a particular name).

Answer 7

Inclusion of "higher"-level but distinctive details about plot

Here "higher" detail level means less specific and more general. The presence of these in either the question or answer makes it more easily recognizable to searchers, so long as they are specific enough to separate the story from others that are similar. A broader and more generic description of the whole story is potentially more valuable than a tighter and more specific description of a few particular aspects of the story, especially as a "hub" question that would be linked to the more specific questions closed against it.

Answer 8

Inclusion of a link to an ISFDB or IMDB entry for the story/show/movie

ISFDB is a great resource for science fiction research, and it is particularly valuable for showing in which collection(s) a given short story has been published, as well as the existence of variant titles and translations. IMDB provides a wealth of information about movies and television shows, which may be of interest to searchers.