I understand closing questions as duplicates from a quality POV, but SO doesn't function as a high quality forum of expert users. The plain use case and majority perception of it is as a Q&A site where you can ask questions you're confused about and get the answer; in this case, closing questions as duplicates because they share a premise with another question is purely counter-intuitive.
Let's take for example the Java Infinite Loop question from the article. The asker was an introductory user to Java who made it clear, and asked multiple questions with relation to how to make the program ask the question in an infinite loop. The duplicate marked was purely a "how to do infinite loops in java" question. While for a more comfortable software engineer this suffices since they can use the method for accomplishing the larger concept of an infinite loop in their specific use case, the user was clearly so new that the question should have been approached from a beginner's perspective, i.e. "You'll need to wrap your main question asking code in an infinite while loop like this so that it continues forever." Besides, the secondary question the user asked wasn't answered in the duplicate.
While it makes sense to mark a question that was just "How to do infinite loops in Java" at a high level as a duplicate, this more specific question also requires the answer to identify where the loop should be placed and why it should be placed there. For a site that's supposed to be about helping the community, most of its high reputation users seem to be awfully focused on removing questions, discouraging askers, and the most minor of revisions (i.e. why so many questions I see on SO have been edited by a high rep user to add a line break or the most minor of cosmetic changes).
Moreover, the actual mission they're supposed to accomplish by being a trusted member of the community goes largely unfulfilled. In most SO threads I browse now, the accepted answer is just plainly incorrect or outdated. There's no enforcement at all here, and often times it's been edited in the last couple months as just a cosmetic change instead of actually fixing the glaringly wrong issue. The correct answer usually lies further down below, ironically posted by low rep users who struggle with the incorrect answer, find a better working solution, and then post it with the pure intention of helping others who came across the issue.
An obvious example is in many JS questions nowadays. The OP asks something very simple and direct, and most of the answers either:
- solve the question using jQuery, which is usually unnecessary nowadays and adds bloat to a relatively simple task
- use some random library or module, i.e. "simple, just learn the API for this open source lib instead and use it!"
- contain a solution that doesn't work or works in a very small number of cases (i.e. with most regex questions where there are 10 competing regex answers and you've got to find the right one)
- contain a correct solution but use extremely outdated methods to get there; i.e. using a giant XMLHttpRequest with a callback instead of just using async/await and Fetch, or lacking any sort of ES6 convention (using `var` everywhere)
- contain an almost correct solution except for a few typos, but no one who has spotted the typo and has the right answer also has enough reputation to edit the answer, and so the resolution of the mistake is a buried comment
----
In my honest but probably worthless opinion, Stack Overflow is a Q&A forum with the extremely strict and overbearing rules of a high quality expert's forum, and just encourages this sort of behaviour on the platform to the point that it really just fails to serve its core purpose on many occasions.
Let's take for example the Java Infinite Loop question from the article. The asker was an introductory user to Java who made it clear, and asked multiple questions with relation to how to make the program ask the question in an infinite loop. The duplicate marked was purely a "how to do infinite loops in java" question. While for a more comfortable software engineer this suffices since they can use the method for accomplishing the larger concept of an infinite loop in their specific use case, the user was clearly so new that the question should have been approached from a beginner's perspective, i.e. "You'll need to wrap your main question asking code in an infinite while loop like this so that it continues forever." Besides, the secondary question the user asked wasn't answered in the duplicate.
While it makes sense to mark a question that was just "How to do infinite loops in Java" at a high level as a duplicate, this more specific question also requires the answer to identify where the loop should be placed and why it should be placed there. For a site that's supposed to be about helping the community, most of its high reputation users seem to be awfully focused on removing questions, discouraging askers, and the most minor of revisions (i.e. why so many questions I see on SO have been edited by a high rep user to add a line break or the most minor of cosmetic changes).
Moreover, the actual mission they're supposed to accomplish by being a trusted member of the community goes largely unfulfilled. In most SO threads I browse now, the accepted answer is just plainly incorrect or outdated. There's no enforcement at all here, and often times it's been edited in the last couple months as just a cosmetic change instead of actually fixing the glaringly wrong issue. The correct answer usually lies further down below, ironically posted by low rep users who struggle with the incorrect answer, find a better working solution, and then post it with the pure intention of helping others who came across the issue.
An obvious example is in many JS questions nowadays. The OP asks something very simple and direct, and most of the answers either:
- solve the question using jQuery, which is usually unnecessary nowadays and adds bloat to a relatively simple task - use some random library or module, i.e. "simple, just learn the API for this open source lib instead and use it!" - contain a solution that doesn't work or works in a very small number of cases (i.e. with most regex questions where there are 10 competing regex answers and you've got to find the right one) - contain a correct solution but use extremely outdated methods to get there; i.e. using a giant XMLHttpRequest with a callback instead of just using async/await and Fetch, or lacking any sort of ES6 convention (using `var` everywhere) - contain an almost correct solution except for a few typos, but no one who has spotted the typo and has the right answer also has enough reputation to edit the answer, and so the resolution of the mistake is a buried comment
----
In my honest but probably worthless opinion, Stack Overflow is a Q&A forum with the extremely strict and overbearing rules of a high quality expert's forum, and just encourages this sort of behaviour on the platform to the point that it really just fails to serve its core purpose on many occasions.
reply