7
$\begingroup$

I have seen our software rec discussion here: Are software-requests on-topic?

But I am wondering now about general recommendation questions. The question that sparked this is: https://earthscience.stackexchange.com/questions/520 , which strikes me as both off-topic and opinion based. Off-topic because it is a recommendation question and opinion based because this kind of question will tend to illicit many opinions on what to recommend... around in a circle I go.

Before I cast a close vote I wanted to query the community and clarify our position on recommendation questions in general (not just software).

||||||
$\endgroup$
  • 2
    $\begingroup$ I am 50/50 on this, as I feel the field camp gear question could be better answered by the course instructor. However, if there are people who have gone through field camp on this site, it seems a bit harsh to completely reject that question. I am 50/50 on this. $\endgroup$ – Neo Apr 24 '14 at 15:25
  • 4
    $\begingroup$ I personally don't think that type of question is quite on-topic for this site. $\endgroup$ – Azzie Rogers Apr 25 '14 at 0:30
  • 1
    $\begingroup$ To me questions that involve someone's opinion more than real facts should not be encouraged or promoted on this site. $\endgroup$ – tux Apr 25 '14 at 15:48
5
$\begingroup$

Recommendation questions should be off-topic here.

They have no objectively correct answer.

They lose their relevance over time.

They're not science: they're popularity contests, they're shopping trips, and they're frivolous fluff.

These factors make recommendation questions unsuitable for a site that aims to build a canonical body of expert knowledge.

||||||
$\endgroup$
  • $\begingroup$ These rationales are not generally applicable to all questions of this type (the first one might be if you said "no single objectively correct answer", since that's what this meta question is about). It is possible to have a recommendation question with a small set of objectively correct answers that don't lose relevance over long time periods, and that aren't frivolous. In some cases, popularity is a good indicator of the relevance of an answer. I agree that most questions in this category aren't very suitable, but I would rather a guideline than a blanket ban. $\endgroup$ – naught101 May 8 '14 at 2:12
3
$\begingroup$

Straight out recommendation questions should usually be off-topic here. But some recommendations can be salvaged, by being converted into a "what features are important?" question, or even better "what does this specification mean?", and "how do I know what to look for?" questions.

A really good example is this question on superuser, which originally said:

MONSTER BEAT SOLO HD&SOLO HD WIRELESS

I want to buy a monsterbeat solo hd wireless,but i am worried about the sound quality when it's connected to bluetooth,can anybody tells me,is there big sound quality difference when it's connected to bluetooth,even when it's wired is there big sound quality difference between solo hd wireless and solo hd?Thank you very much.

For any stackexchange veteran, that's a painful question to read. But @allquixotic skilfully converted it into a quite useful discussion over the relative merits and quality of bluetooth audio vs 3.5mm wired audio:

3.5mm Analog Audio vs. Bluetooth Audio Quality

I want to buy a wireless (bluetooth) headset, but I am worried about the sound quality when it's connected to Bluetooth. All other things being equal (e.g. the quality of the headphones themselves is the same), is there a noticeable impact on sound quality when it's connected to Bluetooth, compared to an analog audio jack? What about between the a bluetooth headset and a wired headset when the bluetooth headset is wired (plugged into the jack)?

I am currently looking at the Monster Beats Solo HD vs. Solo HD Wireless headsets, but I am looking for knowledge about the pros and cons of bluetooth audio and analog audio in terms of quality.

The new version of the question retains the relevant parts of the original question, but moves the focus to the re-usable information that could be useful to other users trying to make the same kind of decisions.

Not all recommendation text can be salvaged in this way, for example, this question asks for recommendations for online sellers and brands of specific gear. While there could be some useful information in there (the one answer to the question does provide re-usable information), the question is too convoluted to bother, and it would be better to start from scratch and re-ask a cleaner question.

||||||
$\endgroup$
2
$\begingroup$

I am a bit put off by the idea of some group of people deciding what they will 'allow' before it comes along. As I understand it, the point of SE's brilliant system of rep and voting lets the community determine — iteratively and organically — what is and is not interesting and useful. I guess I just don't understand why there needs to be a policy in place on everything.

To put it another way: don't fix problems we don't have.

||||||
$\endgroup$
  • 2
    $\begingroup$ The reason for the post is that site-wide SE does not allow this type of question. However, a few do and as we did have a few rec questions already it is something we should discuss. If you are against discussing it I'll consider this a vote to maintain SE policy and not allow this type of Q. $\endgroup$ – casey Apr 25 '14 at 18:55
  • 1
    $\begingroup$ OK, good to know about SE policy — that's important information. I'm not against discussing it, I was mostly reacting a bit against the "Are we going to allow this" language and the general move to try to control the way the site plays out before it can barely walk. $\endgroup$ – kwinkunks Apr 25 '14 at 18:59
  • 2
    $\begingroup$ Its important to determine what is on-topic and what isn't at this phase of the site when it is easier to set and maintain the tone that new users will see and hopefully use as a guide. $\endgroup$ – casey Apr 25 '14 at 19:30
  • 2
    $\begingroup$ I guess that's what I'm challenging: I don't think it is all that important. It's what voting and rep is for. The community decides, rather than a small group having to decide "for the good of the people". I'm not suggesting a free-for-all, just a light touch. The lightest possible. $\endgroup$ – kwinkunks Apr 25 '14 at 19:35
  • 1
    $\begingroup$ Challenging the idea is fine, but first read what SE has to say about it blog.stackoverflow.com/2010/07/… $\endgroup$ – casey Apr 25 '14 at 19:37
  • 1
    $\begingroup$ Right, fair enough. All good advice. I think I just have a problem with authority! $\endgroup$ – kwinkunks Apr 25 '14 at 19:42
  • $\begingroup$ @Casey: it is not true that these "SE does not allow this type of question" - they are discouraged, sure, but they're not disallowed. There are, for example, a number of good list questions on Stats.SE. $\endgroup$ – naught101 May 8 '14 at 2:14
0
$\begingroup$

I think this is a somewhat nuanced issue. I agree that anything resembling a boat geoscience question should be avoided (e.g. "What should I take when going camping to collect samples"). However, there is a category of question that isn't a straight shopping recommendation, but is asking about specific tools for earth sciences, which I think is reasonable. Yes, the answers will eventually become outdated, but they may be useful to others in the mean time.

Examples of questions that are, arguably, recommendation questions, but that I think are valuable are:

||||||
$\endgroup$
-3
$\begingroup$

I feel like this question is very closely related to Is telling jokes good marketing?.

I think that some specific types of recommendation questions can be beneficial. In particular, questions that recommend the best introductory texts in a field (where the field is reasonably broad - e.g. hydrology, not coastal erosion) are really useful to have on the site as community wiki pages. Pages like these exist on other sites, but often get outdated. Because of the strength of the community at SE, hosting a wiki page for these kind of questions means that they can more easily be kept up to date, and can be reviewed by the team of experts that we have on-hand. However, there should be some strong guidelines for answers: they should go into some depth about the level of prerequisite knowledge, and what is covered in the book, for instance.

The references tag as Stats.SE contains a bunch of good examples of how this specific type of recommendation question can work well and be useful. I think these questions are also quite good at helping the site's search rankings (and I don't think that the idea that they promote poor recommendation questions is a) a real problem, or b) backed up by any evidence).

||||||
$\endgroup$

You must log in to answer this question.

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