Welcome to Codidact Meta!
Codidact Meta is the meta-discussion site for the Codidact community network and the Codidact software. Whether you have bug reports or feature requests, support questions or rule discussions that touch the whole network – this is the site for you.
What are duplicate questions?
What does Codidact consider a duplicate question? What practical consequences does that have?
There is no one, over-arching rule across the network. Communities are self-governing (within the limits of what we ca …
1mo ago
I don't think it is necessary to define what a duplicate is for the whole network. Each community might have its own vie …
1mo ago
The most important aspect of a question, is always the answer. We may all have questions, but most questions here are no …
1mo ago
3 answers
You are accessing this answer with a direct link, so it's being shown above all other answers regardless of its score. You can return to the normal view.
I don't think it is necessary to define what a duplicate is for the whole network. Each community might have its own view and criteria when it comes to identifying duplicates and I think Codidact should give each community this freedom.
For example, some communities are more focused on solving problems, for other communities explanations might be more important. This will influence what duplicates are for each community.
1 comment thread
There is no one, over-arching rule across the network. Communities are self-governing (within the limits of what we can host). I agree with this answer.
Personally, I think duplicates should be evaluated question to question, not considering answers. If two completely different questions were to nonetheless have the same solution, that would not make them duplicates. It might make them related (and I've long wanted a better way than comments to indicate related questions), but if a person asking, or seeing, question A cannot read question B and clearly see that yes those are the same question, it's not a duplicate in my opinion. The expectations around cross-community duplicates, like scope boundaries, are worked out by mutual consent of the communities involved.
But that's my opinion. Any community on our network could have stronger or weaker rules. Heck, communities can rewrite, supplement, or even remove the "duplicate" close reason. Here on Meta we added "superseded by" to convey "same issue but go over there instead", because "duplicate" wasn't quite right for those cases. (We also have the off-the-shelf duplicate reason.)
Some communities are here to build a carefully-organized repository of knowledge with clear boundaries and no overlap across questions or answers. They are free to do that. Others are here to provide answers with the context that's needed for the particular question -- answers on two different questions might both say (or link to) "do X" but include explanations about why to do X that are completely different because the questions are different. They are free to do that too. Some communities are here more for the satisfaction of the participants than a goal of building a timeless repository (for example, puzzle/challenge-focused communities). And they, too, are free to do that.
One size doesn't fit all. I want to help each of our communities best serve its goals and its participants.
0 comment threads
The most important aspect of a question, is always the answer. We may all have questions, but most questions here are not of the philosophical or rhetorical kind, hence without an answer, the question is not particularly useful.
This is the practical answer to why we enforce duplicate closure on Codidact; it's better to gather the answers in one location, and send everyone there.
Questions do not have to be the exact same in order to be considered a duplicate. Different questions can sometimes have the exact same answer. If that is the case, they are considered duplicates. In the same way, very similar questions aren't necessarily duplicates, because they may have different answers.
Cross-community duplicates
Sometimes a community receives a question which is already answered elsewhere on Codidact. The scope of whether or not a question is a duplicate, is not just limited to a single community. However, it's important to take into account the community in which a question was posted in, when judging if it's a cross-community duplicate. Sometimes, a rather similar question will have a different answer based on which community it belongs to. For example, a question asking for an interpretation of a specific religious text, will have a different answer based on which religious community it was posted in, as it's looking for the viewpoint or interpretation with that specific religion's aspects in mind. Therefore, not all questions are cross-community duplicates despite their similarity.
If an answer already exists in another community, but has an insufficient explanation for the cross-community question, it's reasonable to not close it as a duplicate, but instead provide an answer which links back to that original answer, and additionally expands on it with an explanation suitable for the question in the current community. For example, an answer on Software Development is likely unreasonable for somebody seeking answers in Power Users to understand, or apply to their own case. As such, it's better to write a new answer more suitable for the different audience.
With another example in mind, the same request for a new feature may be posted in multiple communities' meta categories, as well as on Meta Codidact. That is definitely a cross-community duplicate.
It is the curators in each community that best knows if a question they receive, is a cross-community duplicate. Their knowledge of the community they curate, is best fit to judge if the question is this type of duplicate.
We do not yet have the functionality on Codidact to close posts as duplicates across communities. That is something to be implemented down the road.
Signposts
Exact duplicates are often not very useful to keep around. Low-quality duplicates will suffer the same fate. We eventually delete them.
On the other hand, some questions provide a different entry to the same answer. Others simply present a different view of the same problem. Together, they all form useful paths toward the same set of answers. Therefore, we consider these questions signposts. We still close them as duplicates, but we do not delete them, because they are valuable to keep around.
Is it necessary to close duplicate questions?
Yes. In order to prevent fragmentation of knowledge, and to make it easily accessible and discoverable, it is beneficial to ensure that duplicate questions link back to the location in which the answer is already found, as well as prevent answers from being posted to the new duplicate question. It would be very hard and confusing for both curators and answer seekers if the knowledge is scattered all around Codidact. Additionally, it will promote repeatedly posting the same question from different users, which bloats our knowledge repository with no added value.
Question incorrectly closed
We all make mistakes. That includes the curators and moderators on Codidact. Luckily, a question closure is not the end of the world. If you are confident that a question was marked as a duplicate in error, flag to reopen it. It may also be useful to edit the question to clarify it, or post a comment explaining the case.
The duplicate question and answers are outdated
Use the "outdated" reaction on the posts. There is currently no system in place to handle these, but we'll build one down the line.
1 comment thread
0 comment threads