This particular SE has the significant problem that many answers will be "soft" in nature; there's no real literature on much of this, making many questions answerable only through "in my experience..." answers. This answer is a fine example of this... the answer may be correct, but only for a certain subset of the interested population, and there may not even be a "definitively correct" answer. That last point is the major problem... many of these questions will not have a definitive answer, but the "soft" answers will often do the job of addressing the question. This question about seminar attendance is similar... there's no real answer, but there are a few conjecture-type answers that seem to satisfactorily address the question.

So, how should we deal with "soft" answers to questions that demand them?

  • What about soft questions? Should we have a tag for them? Coming from StackOverflow, I was surprised when this was upvoted, academia.stackexchange.com/q/13357/8966 (though I had seen a lot of soft questions already). Moreover, should we change the about page? It currently discourages questions which are primarily opinion-based. Commented Oct 12, 2013 at 13:34
  • @Blaisorblade - We generally don't create "meta" tags (i.e., tags conveying information about the question rather than the question's content). However, more to the point here, this question was posted almost two years ago, and we've been operating with this framework for a while. If you want to revisit this (which is fine!), I recommend you ask a new question.
    – eykanal
    Commented Oct 13, 2013 at 2:27

3 Answers 3


There's nothing inherently wrong with questions that seek advice and wisdom from those who have more experience than themselves. And there's nothing wrong with sharing that wisdom here, if you can…

…back up your statements with constructive, sound reasoning.

But where these questions go wrong is when they become so generic as to stop soliciting hard-earned wisdom and veer towards simply polling the community.

The blog post Good Subjective, Bad Subjective should be required reading for this community.

The earmarks of bad subjective is when the answers fill with "I did this", "I did that", and "I did blah blah blah" responses. When answers don't even purport to explain why their solution is better than any other, it goes from being Q&A to just a poll of the community. Overly broad questions are just soliciting a collection of random answers. Folks will vote this stuff up, but it's just not good Q&A.

The Trouble with Popularity

It's a tough sell, but if you want this site to survive, you need a place where people are asking very interesting and challenging questions, not the uninspired poll questions that have all been asked 100 times before on every other site on the subject. If this site wants to rehash the same old conversations found in any random message forum, there's really no point in trying something different here.

If we can avoid questions that are simply asking, "Let's hear what everyone has to say about…", we can maintain the ideals of great Q&A in the face of completely subjective topics. This is especially true early in the beta when the site is new. The earliest questions on a site will set the tone and topic of the site for a long time.

  • When asking subjective questions, would it be useful to include a suggested rubric to be followed in answers? For example, in my question about citation managers, would it be improved by asking sub-questions that get to specific features about each citation manager? I think this might address #1 and #2 in the "Guidelines for Great Subjective Questions."
    – dmahr
    Commented Feb 15, 2012 at 18:16
  • 5
    @dmahr No. I'm sorry, but this is exactly the type of question I'm calling out as a poor fit for this type of Q&A. You're asking everyone to list what software they use. It's a poll. There's no specific expertise involved and just about anybody can answer. The voting does nothing to raise the "best answer" (whatever that means here) to the top. It's just a popularity contest. Adding a bit about how to format the answers does not make the question any better. Commented Feb 15, 2012 at 18:33
  • What is an example of a really good subjective question you've seen so far in this beta? Sorry for pestering, I'm just unsure of how to implement the six guidelines when asking questions.
    – dmahr
    Commented Feb 15, 2012 at 19:57

I know I'm answering my own question, but I want to put forth the argument that these style answers should be acceptable on this SE. In my mind, this SE serves the purpose of transmitting the "lore" of Academia to aspiring graduate students, new graduate students, new faculty, and the like. Much of this lore is not "official", but it is highly useful nonetheless.

That being said, given that the answer to each question is likely going to vary significantly based on the particulars, I think we should customize the use of tags on this site to specify the particulars. Specifically, each question should have a tag that identifies the field of research; if none is there, the poster should be asked to provide one. This will always be relevant to the question; answers to a question about note-taking styles will be different for mathematics and history, for example. Additionally, we may want to consider asking for:

  • BS/MA/PhD/postdoc/professorship
  • country
  • institution (not sure about this one)

I strongly think we should add some text on the "new question" page strongly suggesting that the poster includes tags for each of these, if relevant to the topic.

  • +2: Good question, good answer, nothing more to say :) Commented Feb 15, 2012 at 17:35
  • +1: A very good answer, especially since there's not going to be lots of hard numbers across academia as a whole. I'd like to avoid the Programmers-esq "Are there any studies about some obscure thing?" that are really just fishing for opinions. I find allowing soft answers of "lore" and experience to be far more honest.
    – Fomite
    Commented Feb 16, 2012 at 23:21
  • +1 for excellent tagging suggestions
    – JeffE
    Commented Feb 20, 2012 at 15:59

Many of my answers to advice questions tend to be of the "in my experience" category. For these questions, there is never enough information to be able to give a definitive correct answer. What we can do is give "in that case, I did X and the consequences were Y" type answers - then let the original poster determine which answer better fits their situation. Older folks have found that giving direct answers to questions to younger folks results in the advice getting ignored, so that more circumspect answers actually get listened to. As for your second link, I don't think that young folks understand just how much politics goes on in the real world.

What we do as a "community" will change over time. Some of the other sites here preferred such answers in the past, but now discourage them. This has lead me to quit visiting a number of other stack exchange sites.

You must log in to answer this question.

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