I asked a specific question which might be difficult to answer (link). A few people tried giving me pointers and I improved the whole circuit, which makes the question not relevant anymore.

How should I proceed?

  • Delete the original question
  • Post an answer with "I improved the circuit as mentioned in the comments"
  • Leave it as it is with no answers

I already have a new question based on my current circuit which is in the same category.


3 Answers 3


Good question - thanks for asking it. The object of the site is to collect answers to (on-topic) questions for future use.

I would recommend a variation of one of your options, namely:

  • Post an answer with "I improved the circuit as mentioned in the comments as follows" and include a summary of your improvements in the answer. You can then accept your own answer, to effectively show that the topic is closed.

There is no point in an answer which effectively just says "fixed it" (in different words). But an answer which includes how you fixed it, and especially if you can specify why a particular change fixed it, would have value for the future.

One concern: You say that you have a new question. Before writing an answer saying that changes XYZ solved your problem, we need to make sure that your new problem isn't a different symptom of the same problem, meaning that changes XYZ didn't solve the problem. So is it premature to write an answer to your old question, or are you sure that your improved circuit has really solved that original problem?

Summary: If there is a way to capture knowledge in an answer, which can be used for the future, we should try to do that.

  • 2
    \$\begingroup\$ Thought so. This would also give closure to the people who contributed, they may need to help me in the follow up question. My new question is indirectly related to the old (obsolete) one. My next steps: 1. Formulate my new question 2. Wait for responses\ answers 3. Answer my old question with sourcing the comments and link to the new one \$\endgroup\$
    – v3xX
    Apr 28, 2021 at 12:30
  • \$\begingroup\$ @v3xX - That sounds like a good plan & reasoning to me :-) We can see if anyone else suggests a better one. \$\endgroup\$
    – SamGibson Mod
    Apr 28, 2021 at 16:37
  • \$\begingroup\$ Well, my recent ban made me improve my questioning and also asking in 'meta' before asking vague questions. \$\endgroup\$
    – v3xX
    Apr 28, 2021 at 16:52

This is one of the problems that happens when people start answering the question in the comment section.

There is apparently a user who, instead of telling you all you need to know about debugging your circuit in the answer, wrote three long comments explaining:

  • How to debug problems like this
  • Pros and cons with the different methods
  • Suggests some other workarounds

All of this in comments!? This made you investigate, change+solve your question and now there is no need for an answer even though it would have been a great answer.

However, I'm fed up with the lack of response I get when trying to explain why people need to stop writing "helpful" answers in the comment section, so this will just be a rant. I don't know how to solve it other than reporting repeat offenders.

  • \$\begingroup\$ I think some answers are too 'weak' for a full answer. An answer here seems very grand and permanent. Especially in circuit design or code review most problems need a discussion. One big plus on the other hand is the compact format of 1 question has 1 answer. \$\endgroup\$
    – v3xX
    Apr 28, 2021 at 20:24
  • 5
    \$\begingroup\$ @v3xX If an answer is "too weak" you definitely don't want to post it as a comment where it can not be refuted, discussed, downvoted, or edited. That's the worst place to put a half-baked answer. And stack exchange was designed to take discussions out of the loop - traditional forums is the place for that, and there are many such forums for people who still enjoy discussing such things. \$\endgroup\$
    – pipe
    Apr 29, 2021 at 11:32

I think complicated "please help me debug this" questions lead to this sort of issue. The person posting the comment had no real way of knowing whether this would answer your issue or not. It really was to the effect of "don't even try to debug this until you bring your power to spec". It doesn't answer many aspects of your question, such as "how can I initialize each channel?"

A comment to this effect was fine. A "this helped me out -- can you please post it as an answer" comment would get you the outcome you want, or posting it as an answer yourself would work too, as already pointed out.

In essence, because the question doesn't really match the right answer, this question doesn't match the level of ideal for the archive, as sort of an "XY" question.

I don't really see this as a problem, so much as the nature of the SE beast.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .