From time to time, the best answer will be decent, but not outstanding. It answers the question, but the question could have been answered in much greater depth. That's alright. You can't win them all.

Unfortunately, at other times, the highest answer is nothing but a sorry excuse for an answer. Rather than answering the question, it describes the author's opinion on the question or, even worse, his "best guess" as to what might be the right answer.

I call these pseudo-answers, mainly because calling them low quality answers would be far too generous.

The tell-tale signs of a pseudo-answer are:

  1. The answer is prefaced by "This is not a complete answer, but..."
  2. The whole answer is less than three lines long.
  3. There isn't a single link in the answer or reference, and it's several paragraphs long.
  4. The answer starts with "This is just a guess but", "I think that", "Maybe that, "I would guess that", "In my opinion," "Personally," or any other term that suggest speculation rather than certainty.

Why this matters

As I said in the past, this site's success depends on our ability to continually produce great answers:

High quality answers are the life and blood of our site. If we continually produce high quality content, we'll attract more experts who will want to help fellow skeptics. On the other hand, if low quality dominates the site, an expert who visits the site will conclude this site is a waste of time and never come back.

Pseudo-answers go against at goal in at least three ways:

  1. If a question does not have any answers with a score of at least 1, and has no accepted answer, it will show up in the Unanswered Questions tab. That's right, it's not enough for a question to have answers, it has to have positively voted answers to be considered answered. An upvoted pseudo-answers might be the reason why a question has not yet received a good answer.
  2. Heavily upvoted pseudo-answers gives the impression that this is the kind of content we want.
  3. Heavily upvoted pseudo-answers may reduce the visibility of good answers. If by the time a good answer is posted there are already a few upvoted bad answers, then it's less likely the great answer will be upvoted to the top. It may stay buried under a pile of bad answers, where few will ever see it.

If our goal is high quality answers, and it is, then pseudo-answers are the enemy.

Pseudo-answers do, sometimes, make great contribution to the thread. However, Skeptics is not a forum. We are a question and answer site. Answers are meant to be answers, not replies. If you have something to add and it's not actually an answer, then write it in the comment section.

How can I help?

There are many ways to help us get rid of the pseudo-answers threat:

  1. Write good answers. The best way to avoid the creation of pseudo-answers on a question is to write great answers. If an user learns by reading your answer, it is much less likely that he will feel the need to post an answer of his own. If this is done constantly on each answer, new user will learn that this is the kind of answer his answer will compete with.
  2. Downvote them, at least down to zero. Downvoting pseudo-answers is paving the way for better, more informative answers: it makes sure that unanswered questions show up in the Unanswered Questions tab and gives more visibility to new questions.
  3. Leave comments on infringing answers. Downvoting sends a signal, for sure, but the lesson might not be clear. When downvoting, try to leave a comment explaining why. With luck, the author will persevere and learn from his mistake.
  4. Flag them. If you see a pseudo-answer, please flag it as "low quality." If moderators agree, they might convert the question into a comment, delete it, or downvote it.
  5. Write good questions. A well-written question helps reduce the probability of pseudo-answers being written. It's not a miracle remedy, and sometimes a well-written question will receive several poorly written questions, but putting effort into your questions never hurts!

If we work hard together, we can purge the site of pseudo-answers and the site will be much better for it.

  • 5
    I don't see why a lack of links means it's a pseudo-answer. Commented Mar 29, 2011 at 8:56
  • @LennartRegebro: It's a good indicator that there might be a problem - hence why it's a tell-tale sign - but, yes, there is a chance for false positive.
    – Borror0
    Commented Mar 29, 2011 at 9:04
  • @LennartRegebro: It's an indicator only. Item 4 is a dead giveaway, but the other three are only "smells". Just because something smells, doesn't make it bad. It could be a fine danish blue. Clearly the No Links rule should never be the sole reason for a downvote. Poster could be referencing facts already cited in the thread, or could be putting forth an argument for or against facts not in dispute. Commented Mar 29, 2011 at 12:57
  • 1
    Lots of psuedo-answers taken together can be considered a 'good answer' why discourage partial answers when all it takes is consolidating them into a single 'complete answer' the 'it doesnt show up on unanswered questions' argument is isolating a bug in the software, not a problem with psuedo answers. Just add a 'this question has not been properly answered' button for inclusion in 'unanswered questions' and the problem disappears. Commented Apr 7, 2011 at 17:50
  • "are the enemy" I can't quite say what's wrong with this phrasing, I just know that I dislike it strongly. :-( And it colors (of course) my opinion of anyone using it. Though my opinion doesn't count, of course. Commented Jun 29, 2011 at 17:59
  • 1
    And I think discouraging answers is a bad move, period. Strong connection to the "all answers have to have references" (which isn't the actual wording of that rule, but it's being used as such). Commented Jun 29, 2011 at 18:01
  • 1
    @jae: We believe that quality of answers is more important than quantity of answers. Ultimately, it's the quality that makes the site useful, not the sheer number of answers on one post.
    – Borror0
    Commented Jun 29, 2011 at 18:04

6 Answers 6

  1. The answer is prefaced by "This is not a complete answer, but..."

I like it when people warn me of problems and I don't see it as a problem that people leave partial answers, if anything I think it's a good thing, it's just not the best thing. Leaving well referenced partial answers, is of help to others as long as the answer isn't noise. I don't at all see how it's equivalent to:

  1. The whole answer is less than three lines long.

  2. There isn't a single link in the answer or reference, and it's several paragraphs long.

  3. The answer starts with "This is just a guess but", "I think that", "Maybe that, "I would guess that", "In my opinion," "Personally," or any other term that suggest speculation rather than certainty.

I would also much rather have complete answers for everything, but the main objective should be to leave answers that useful, and not a waste our readers time.

Discouraging partial answers is going to be a huge problem when dealing with questions that have a large scope with lots of research (like say climate change) and seems to me like it's an easy way to remove useful contributions from the site.

Instead let us praise answers with any noise (by up-voting them) that lets us collaborate and build upon each others partial answers, letting everyone chip in to the body of answer that can paint a complete picture even when it's not in community wiki mode.

  • 2
    Don't get me wrong. Partial answers indeed are great contributions: they often make amazing comments. We shouldn't confuse them for good answers though, because they are not. As answers, they decrease the chance of receiving a good answer.
    – Borror0
    Commented Mar 29, 2011 at 6:57
  • 18
    It's not practical to leave partial answers in comments when the character limit is 600 and it offers little in the way of formatting.
    – Kit Sunde
    Commented Mar 29, 2011 at 8:14
  • 8
    well said, Kit. I'd rather leave a partial answer if I've no time to research a complete one, and give others a starting point to expand on it. Also, a lot of things aren't online, so there's no websites to link to.
    – jwenting
    Commented Mar 29, 2011 at 8:17
  • 2
    @KitSunde: No solution is perfect. I personally prefer great answers and poorly formatted partial answers than no great answers and well-formated partial answers.
    – Borror0
    Commented Mar 29, 2011 at 9:36
  • 1
    As long as they aren't upvoted - therefore marking the question as answered - they should be fine, in my opinion. Commented Mar 31, 2011 at 13:55
  • 2
    @AndréParamés: If we lived in a world where we could trust random stranger 15+ reputation to not upvote those answers, I would agree with you. Unfortunately, we do not live in such a world.
    – Borror0
    Commented Mar 31, 2011 at 19:59

I would have a policy of writing better answers on particularly problematic questions with poor answers, and perhaps even coordinating in chat to vote them up -- because they are better, natch -- higher than the poor answer.

I might also support downvoting particularly low quality answers to get them further down the sorting.

  • 1
    Organising up-votes sounds like foul play to me. Your answer should attract up-votes on it's own, without collecting your friends and using subtile peer pressure. You didn't have peer pressure in mind? Yes, that's the problem. ;) We aren't facebook, are we? Commented Jun 30, 2011 at 15:09
  • Note that, while I often try to follow this advice, it is also really too much effort if the users actively policing and taking care of the site are only the mods.
    – Sklivvz
    Commented Jul 24, 2011 at 20:16
  • 2
    @sklivvz that's probably part of the checks and balances.. it should take effort to collude Commented Jul 24, 2011 at 23:44

I think this is an over-reaction.

Heavily upvoted pseudo-answers may reduce the visibility of good answers.

Can you provide examples of this happening?

  • 1
    It's right here. Should I downvote? ;) Commented Apr 11, 2011 at 5:16
  • 2
    @muntoo: No, upvote me instead! :)
    – Golden Cuy
    Commented Apr 11, 2011 at 6:01
  • 1
    If you say so. :) Commented Apr 11, 2011 at 6:02

It seems strange that someone so concerned with having everything on the site perfectly set up would post a "question" that isn't actually a question at all. Assuming you meant to ask "Are pseudo-answers bad?" instead of merely stating your opinion on the matter, the answer is no.

Some information is better than no information. It would be nice if by magic we could make all answers complete. We can't. If an answer provides a partial answer to the question, it deserves an upvote because it has improved the knowledge available. Other answers can build on this information and get upvoted further, eventually reaching a point where we have a good answer. For a contrived example to show the flaws of only allowing complete answers, if my question is "What is the combination on this lock?" and three people each know one number each, the only way to get the complete answer is if they each post their partial answers. This is a good thing and should be encouraged.

If you instead meant to ask "What can be done to improve the site with respect to partial answers?", an answer might be the ability to flag an answer as partial (either as the poster, or as a voter). This could make it still show as unanswered, or be used in the filtering scheme so that even a heavily upvoted partial answer would show up below a less upvoted full answer.

Partial answers can sometimes be posted as comments, but that's not always feasible or the best option.

  • Each number of can be posted in the comments, and then someone can used all that knowledge to write a complete answer. If the first number was posted as an answer and then upvoted, it's highly likely that no one will ever come to post the other two numbers. Perhaps the system should be changed, but until it is we must try to act in the most productive way, which entails not posting partial answers.
    – Borror0
    Commented Apr 20, 2011 at 20:45
  • 1
    and with a less contrived example, where the partial answer actually requires some amount of depth? formatting? charts? More than 600 characters? The comments section is a poor place for in depth responses. Commented Apr 21, 2011 at 17:46
  • When I talk about a partial answer, I mean an answer that does not satisfactorily answer the question. It's not about covering everything. That would be ridiculous. If you write an in-depth post and it still does not answer the question, I would suggest that there's a problem with the question.
    – Borror0
    Commented Apr 21, 2011 at 21:51
  • -1 For verbosely restating things that has already been said in this thread.
    – Kit Sunde
    Commented Apr 22, 2011 at 15:33

It's been suggested that "pseudo-answers" should be comments instead.

Sometimes, I've had the opposite problem on Stack Overflow - I've posted something as a comment, and it's solved their question, and they've asked me to post it as an answer so they can accept it. I can dig up a link if anyone wants me to.

  • Obviously, there two sides to this extreme: comments that belong as answers and answers that belong as comments. Currently, we're seeing too much of the latter, hence this post. the idea is that we are trying to encourage users to write greater quality answers.
    – Borror0
    Commented Mar 31, 2011 at 10:23
  • 8
    resists urge to convert this answer to a comment
    – Shog9
    Commented Mar 31, 2011 at 14:51
  • 1
    "pseudo-answers" as comments are a way for people to give answers that cannot be downvoted but (if on the main question) still get a lot of visibility. I would rather comments be used for clarifications. If an answer is not good enough for voting, it should probably not be given at all. Commented Apr 12, 2011 at 22:24

A far worse problem is people referencing their own answers/posts here and on other sites as authoritive sources of information.

  • That's another problem entirely. If you see those posts, however, please flag them. We will take a look at them.
    – Borror0
    Commented Mar 29, 2011 at 9:09
  • 3
    It's not a problem if the person in question can be considered a legitimate expert in the field.
    – oosterwal
    Commented Mar 29, 2011 at 21:24
  • Why is this a problem? For whom? Commented Jun 30, 2011 at 15:06

You must log in to answer this question.

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