When creating a self-answered, informative post, should we encourage & allow other answers that merely add commentary to the existing self answer?

For example, this answer: https://stackoverflow.com/a/16257082/427309

It far from answers the question in the OP, other than saying "I wouldn't do that myself".

To me, that should have been done instead as a comment, or as a series of comments on my answer (as a good portion of the answer relies upon mine), instead of an answer myself.

However, when I flagged it, it was declined by Brad, with this message:

@RichardJ.RossIII - I don't know that we need to be so strict here. He's added some good information to your self-answered question, so I don't think this qualifies as a broken window to be removed. His answer wouldn't work in a comment, so I have no problems with it remaining. – Brad Larson♦

Personally I don't get this. Isn't the "answers" section for answers to the original question, and not replies to other answers?

  • 2
    It's perhaps not exactly 100% an answer. But it goes beyond a comment and deleting it altogether would seem to mean getting rid of possibly valuable content. Should we actively encourage it? No. Do we need to be terribly strict here? I don't think so. I'm with Brad here. – Bart Apr 28 '13 at 20:35

Firstly, self-answered Q&As are no different from other Q&As. Now that that's out of the way:

It is perfectly OK to write a "dependent"/"supplement" answer

There have been many times where I have written an answer that starts with "In addition to @XYZ's excellent answer, [...]". This means that while I could explain what is explained in XYZ's answer, it is already well written and I would rather make some points unique to my post instead of spending my time duplicating someone else's efforts.

In this case, bbum is adding a supplementary section "why not to do it" to your "this is how to do it" answer. It is meaty and informative enough imho to not be deletion/convert-to-comment worthy. Why delete it? While you may not want to accept it (generally it isn't a good idea to accept supplementary answers), it's perfectly fine to stay as an answer.

Also, Don't do it is a valid answer

  • There is an important difference between self answered and normal QAs, and that's the sorting of the accepted answer. When I accept my answer tomorrow, his response to mine will cover up the 'real' answer, and may cause some people not to see the real answer to the solution. – Richard J. Ross III Apr 28 '13 at 20:44
  • 3
    Then make sure your actual answer is even better and gets a sufficient amount of upvotes. – Bart Apr 28 '13 at 20:46
  • @Bart my answer is better in that its a real answer! I don't want to sound arrogant here, but his answer doesn't really point out anything that wasnt in my original post, he's just re-wording it. – Richard J. Ross III Apr 28 '13 at 20:47
  • Then all you can do is hope to gather more upvotes. That's it. No author of a self-answered question is guaranteed that their own answer will appear on top. It's up to the community to value the contributions. And you'll have to live with that, unless it is decided that the other post doesn't belong. – Bart Apr 28 '13 at 20:49
  • 2
    @RichardJ.RossIII: It's not for you to say which answer is better. The votes say that. And technically his answer acknowledges and #includes yours. This means that a reader of the post will know of the existence of your answer. – Manishearth Apr 28 '13 at 20:53
  • @Manishearth since when do people read context around what they are reading? Tag wikis are a good example. I'm not here to say that my answer is perfect - it isn't, by far, but it at least solves the problem effectively. I can understand why people wouldn't want to do this, but does that really constitute an answer if it doesn't suggest and alternate way (other than what I've already outlined in my question)? I guess it would be one thing if We weren't experts on the topic, but seeing as we are both advanced programmers, "don't do it" seems pointless. Programming is about learning, is it not? – Richard J. Ross III Apr 28 '13 at 20:58
  • @RichardJ.RossIII: People who visit your post wanting to know how to create a block that wraps[...] will see bbum's answer, realize that it references instructions below, scroll down, and read your answer. That's what matters. – Manishearth Apr 28 '13 at 21:06
  • 4
    @RichardJ.RossIII: "Don't do it" is not pointless. You may be an expert. Bbum may be an expert. The visitors are not necessarily experts. – Manishearth Apr 28 '13 at 21:07

Geez, dude, one of the most experienced programmers for the relevant domain in the world took some of his time to encourage you (first sentence of the post) and then pointed out some possible design and technical snags with what you're attempting to do, and you want to delete it because it doesn't directly provide you with a solution? No. Get over yourself a little bit.

More importantly, that post does answer the question.

as a good portion of the answer relies upon mine

That's simply not true. If you read without bias, the majority of the post -- everything except the first and last sentences, in fact -- stands on its own; it could be based only on the contents of your question. It says "from a code readability standpoint, your desired result may not be the best idea". As Manishearth already pointed out, "you shouldn't do that (and here's why)" is an answer. There's no way this needs to be deleted.

  • When you consider his first and last paragraphs are dependent upon my answer, his code snippet is directly copied from my question, there is little left. Yes, it does add to the conversation in him stating his opinion, but aren't opinionated answers not encouraged here? – Richard J. Ross III Apr 29 '13 at 2:17
  • I'm also not going for a personal attack here. My main focus is the quality of the answers on this site, and that answer simply didn't help me. Yes, I understand that the answer is from an expert, but does that mean we should just take his answer without questioning its content? Absolutely not. This post was me attempting to determine the policy for answers like this, and people seem to have misconstrued it to be a personal attack. It is not, bbum is an awesome guy. – Richard J. Ross III Apr 29 '13 at 2:20
  • Many answers are short and opinionated, and yes, they should always be backed by facts or expert testimony. (Some skirt other rules (clarification request).) Neither of these things mean they should be deleted. Your reaction to bbum's post seems overdone; no one else sees it as a big deal, so it looks like you're whining that you have to share the sandbox. If you say that's not so, I take your word for it, but it does seem like you're making a mountain of a molehill. – Josh Caswell Apr 29 '13 at 5:44
  • neither of those answers of mine are reliant of earlier answers - they stand on their own, regardless (plus, tey both consicely answer the question with actual code). Just because I ask a question in the answer means it is the same as what happened there? I don't think we are comparing apple to apples here. – Richard J. Ross III Apr 29 '13 at 11:27

You must log in to answer this question.

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