What may be the desired evolution for a good answer?

Question

From a debate on this question there is some need to clarify how we make our way to write the best possible answer for a given question.

By voting we may find out what answers are better, what answers are not so good. But sometimes we do already have more than one good answer for a question each dealing with a single aspect of the question. It may also be that existing good answers would become even better if some missing details were added.

If we only collected answers that may have more or less votes we don't have a chance for a best "super answer" that may evolve over time. Ideally one single best answer existed for any given question.

When there already are good answers I have the choice to:

1. answer myself to give missing information.
2. commenting to add my additional views.
3. edit the existing answer to make it better.

Unfortunately I have rarerly seen any comments to appear in an edit. Writing an additional answer would need quoting of existing answers to make it complete. Editing an answer may make people feel bad or misunderstood.

What do you think should be done to make our answers better? Is it welcome to quote exisiting answers or comments when writing an additional answer? Should we encourage people to add missing information by editing existing answers?

Answer 1

2.5 answers per question is good, only 1 answer per question needs some work

Think about it. This quote comes from the same people who want (ideally) one answer to rule them all per question.

Finished thinking? Well, the one complete answer to a question is a myth. Yes, I have to admit, that it sounds good in theory, but as so often, reality kicks in.

A complete answer (if such a thing could exists anyway) can be very very long. It can divert so many times that it is of no use for the asker. In many cases three or four concise answers, making their point clear, looking at the problem from different angles, can be much more useful, than a long exhaustive tl;dr answer (I try to emphasize this by the length of this sentence if it got unnoticed).

Still, these three or four answers beat tremendously the habit of non-Q/A-sites like forums. There you can mil through 50 or more answers or snippets of answers to get a clue, what the answer you need may look like.

Sometimes I edit an answer to add the missing part, sometimes I comment, sometimes I add my own answer. It depends on how much is missing, how much time I have and if I think the poster of the answer is familiar with the SE way of doing things.

It doesn't matter with which you come up here, people will break "the rule". I also doubt, that you want to clean up behind all these people who have no editing (or commenting) right. Just leave it as it is.

Sometimes there are these striking exceptions, where you really want to do something about. Do it. That's what exceptions are for.

Answer 2

My five cents on this...

IMHO the best way to give additional information to an already existing answer is by a comment. Because in this case the major contribution is already done by the person who answered the question. Commenting is also a good way if you (partially) disagree on something. Because someone looking for the right answer will see if that answer is generally accepted or not.

If one generally agrees with the given answer but wants to add an all new aspect or information, I think he (or she for that matter) should add it as an answer. If wanted, it could be noted that the person writing the additional answer agrees with the answer of xy. This way the additional information given can also be honoured with upvotes or if it is bad it could be downvoted...

I generally don't like the idea of a "super answer". If someone wants a question answered he will be interested in multiple views. Having multiple answers (that are not summed up in a super answer) shows this views in a better way. Also it will force the person seeking for the answer to read through all of them to get the complete image. This will be better than the shortened version in the super-answer.

This way the helpful answers as well as comments will show up first because of the voting. So someone looking for a summary could just skim-read the answers and the most upvoted comments. Anyone seriously interested in the answer will want to read as much as possible anyway.

I think editing should mainly be done for reasons of formatting, spelling mistakes or inappropriate language. It would be nice discuss major edits with the person who answered the question in the first place.

---

By the way there is only one super-answer: 42. But as long as we don't have the super-quetsion it is useless ;-)

Answer 3

Ich stimme mit Lukas und John Smithers weitgehend überein.

Als Kriterium würde ich nennen: Eine Ergänzung, bei der man recht sicher ist, dass der Autor der Originalantwort sie begrüßt, kann man selbst in die Antwort einpflegen, oder sie im Kommentar vorschlagen.

Letzteres finde ich einen Deut höflicher.

Viele Antworten hier auf GLaU starten schon als Essay. Hier wäre es oft hilfreich die Hauptaussage etwas hervorzuheben - fett, oder mit Bulletpoints¹.

Neue Gedanken gehören in neue Antworten. Man muss ja nicht wiederholen, was andere schrieben, sondern kann auf deren Aussagen verweisen, was insbes. dann gut geht, wenn diese ihre Texte strukturiert haben, evtl. durchnummerierte Beispiele oder Aussagen präsentieren, so dass man gut darauf Bezug nehmen kann.

---

¹) Oft will ich in Absatz 2 schon eine (+)-Wertung abgeben, aber dann kommen noch 5 Absätze und ich schlafe ein. Da die schlechten Gedanken oft am Ende kommen musste ich (+)-Bewertungen schon zurücknehmen, was nur kurze Zeit geht. Also breche ich ohne zu werten ab, wenn jmd. nicht recht zum Punkt kommt.