Stuff like polls, recommendations based on subjective constraints, puzzles, webcomics etc. do not belong on the serious main SE sites, where professionals should be considered at work and having just a few spare minutes ("code's compiling") to answer questions, so they should not be distracted by such things. However, I'd also like to have a home for these things still using the SE engine. For the reason I mentioned before, this needs to be a separate place though. Let's call it four.[sitename-here].com.

  • 69
    Hey, while we're at it we could implement listofx.stackoverflow.com, toolocalized.stackoverflow.com and notarealquestion.stackoverflow.com! All problems solved! – Andy E Jul 16 '10 at 10:08
  • 6
    @Andy E's head: one place for all should really suffice. Just a "dump the too subjectives here" – Tobias Kienzler Jul 16 '10 at 10:10
  • 14
    @Tobias: Approaching it from a "dump" perspective isn't going to gain any traction. These posts need a place to live rather than one in which to be thrown away. Personally, I'm not sure why some are trying to force that into the SE engine. – Gnome Oct 4 '10 at 9:07
  • @Roger: I have tuned this feature-request since I wrote that comment. I agree creating a dump is the wrong approach, but for some sites (e.g. gaming) this would really be helpful. – Tobias Kienzler Oct 5 '10 at 7:11
  • Would there be enough polls? – Mateen Ulhaq Nov 3 '10 at 5:40
  • @muntoo: no, and that's why extended this proposal to include recommendations and maybe also puzzles and fun-stuff that would distract when posted on the main SE while it were still nice to have around... – Tobias Kienzler Nov 3 '10 at 6:10
  • 1
    @Tobias As long as there's free cheese along with the puzzles, I vote "YES!" – Mateen Ulhaq Nov 3 '10 at 22:45
  • 2
    I agree...and it would actually be the 5th place, Chat was the 4th - I believe. Regardless, I would like to have another SE area, the 5th dimension or 4th...where more subjective materials can be placed, Polling, comics, and something really big - team recruitment for projects, open source, game work, etc. The sites that have been doing this are changing formats and appear to be dropping the Help Wanted section. This is different than CV job recruitment. – IAbstract Jan 13 '11 at 16:14
  • 1
    @IAbstract Chat was built on the principle of the third place. What did you consider to have been the 3rd place? – Grace Note Jan 18 '11 at 18:15
  • @Grace: I think I was miscounting... – IAbstract Jan 18 '11 at 18:37
  • 1
    @gnat Not really, I'm specifically asking about using the SE format even for the subjective part – Tobias Kienzler Jan 22 '14 at 8:23
  • 1
    @TobiasKienzler feature to retract CV is implemented since Jul 15 '13 – gnat Jan 22 '14 at 10:57
  • 1
    Some alternatives: opinions.[sitename].com, discussion.[sitename].com, subjective.[sitename].com, extended.[sitename].com, alternative.[sitename].com, broader.[sitename].com, relaxed.[sitename].com. – Trilarion Jul 31 '14 at 10:11
  • 2
    Why "four" / "fourth place"? – Geoffrey Hale Sep 18 '15 at 17:35
  • 3
    @GeoffreyHale Back then chat was introduced as "the third place" (see also this comment, so my logical conclusion was that the suggested site would be the fourth, and lacking a better idea for a name (subjective-polling-recomendation-stuff.[site] sounded stupid...) – Tobias Kienzler Sep 20 '15 at 17:17
up vote 39 down vote accepted
+50

I couldn't agree more. In particular, asking for product recommendations, rather than asking about specific products, is a recurring theme in many SE-2.0 sites, and controversial in many of them. Having a dedicated fourth place can be the perfect solution to satisfy both the fiercest opposers and the most ardent supporters for these questions.

Jeff's answer hints that his proposed solution is basically to create a parallel website for each of the existing ones. programmers.se is stackoverflow.com's parallel, and indeed there's also an area51 suggestion for a gaming.se's parallel.

One of the nicest things with creating a parallel site is that questions can be automatically migrated with the off-topic migration option; but on the other hand I think this trend is bad for the community at large, at the very least because it creates a different SE-2.0 site which addresses precisely the same crowd, and that can needlessly fracture the community. I've discussed this at length in a meta answer on gaming.se.

Do we really want to see a semi-subjective parallel SE created for each SE that becomes popular enough? Seems to me that having a specialized 4th place is a far more elegant and healthy solution. It does have its own problems, I admit, at the very least meaning that Google results will start showing semi-subjective questions under the ostensibly-objective stackexchange.com domain; but on the other hand that's exactly what is going on now with programmers.se.

  • 13
    as programmers shows us, even that solution is awkward because we still have rules about what a constructive subjective question is, and the people who support this seem to want a "no rules anything goes" philosophy that we don't believe in. – Jeff Atwood Jan 15 '11 at 8:21
  • @Jeff I don't believe in "anything goes" either, and I don't claim that stackexchange should simply contain anything - there are other websites out there. I understand that some questions are purged from programmers.se as well. But you can't deny that in general programmers is far more subjective than SO. I think that this mythical 4th place should try to be similar to programmers.se - i.e. a place for questions that in essence lack an objective answer, such as requests for recommendations, but still a Q&A site rejecting general discussions, and rejecting "bad subjective" questions. – Oak Jan 15 '11 at 8:31
  • 1
    See also slant.co which is highly relevant – Jeff Atwood Sep 5 '13 at 20:22
  • @JeffAtwood the slant resource is helpful. I do, however, keep finding myself turning to SO and other SEs for highly specific recommendations. It doesn't seem outrageous to ask for recommendations that are long-tailish: discussion on how to integrate system X with system Y on platform Z. Usually I am in this position because there are many resources on systems X,Y,Z, and I don't yet know enough about them to effectively root through the possible results. A response that says google term A, would, in this case, be useful! – Ben Dec 10 '13 at 21:18
  • 4
    I think the real problem with this is that by closing the questions it actually makes them more worthless. Honestly, I don't see why "debatable" is so awful. The whole point of voting on answers is to take something debatable and let the community decide what is correct. What I would like to see is another site that doesn't allow answers to be accepted, since the questions are subjective and often time-sensitive. – twiz Dec 15 '13 at 6:10
  • 1
    @twiz letting the community decide what is correct may actually yield unpleasant results such as creationism and anti vaccine – Tobias Kienzler Jul 1 '15 at 4:44
  • Leave them merged into the main exchange. Mandate a tag of "Opinion-sought" show them shaded in pink on search results so that people's eyes can slide past them. Add search capabilities for not tagged X e.g. -opinion-sought. Mandate requiremets for questions: miminum length, what tried, what really looking for. Mandate requirements for answers: Must compare at least 2 products, must justify answers. A thoughtful opinion is valuable. – Sherwood Botsford Nov 25 '16 at 14:56
  • for the languages SE pools are really important !!! languages have rules but in the end they are modified by communities so sharing opinions about it can help other people out !, like please kill the inclusive language ! – Mike Aug 10 at 13:28

This is a very bad idea, let me explain why...

Story Time

In the beginning people had questions and were looking for answers. So these people went to forums and mailing lists to find these answers from the experts. The system was imperfect though, some might say broken; questions were frequently repeated, it was difficult to know what the answer to a question was, and no one knew who to trust. Now moderators and Karma systems worked hard to correct these problems, but the people cried out: We want a better system!

So a man named Jeff teamed up with a man named Joel and they designed a new system. Where forums had been designed for people having a conversation, Jeff and Joel designed a system for Q&A. Drawing on their knowledge of of Q&A's, search engines and human behavior on the internet, the worked hard to create a system where people could find the answers to their questions. They emphasized good answers "floating," reputation and distinguishing between questions, comments and answers.

And all was good... until some people wanted to start using this software for something other than Q&A.

The Problem

My endearing story aside, the StackExchange Engine is a very bad tool to use for things like: Recommendations and Polling (I'm ignoring comics and chatting for the moment). It suffers from a very large number of problems which have been enumerated before. A well designed recommendation system would profile characteristics and similarities and try to determine the relationships between them. This is very much at odds with a system designed to solve "needle in the hay stack" style problem. Instead of selecting the "right" answer, a recommendation wants to absorb "what" the relationship between A and B is, not the strength. While many people seem to think that the voting system will float the "best" recommendation, the reality is that it will float the "most popular" recommendation, not the one most related. What's worse, all those other recommendations don't gain the knowledge of "how" they are related, only the magnitude of that relationship.

If I were to design a recommendation engine based on crowd sourcing, I wouldn't index based on people looking for recommendations, I'd index based on discrete things (for example, in a video game recommendation engine I'd index on Video Games). From there I'd want users to assign characteristics that the games share: Metal Gear Solid is like Arkham Asylum because both are Stealth games. This way people don't see one and think they're both games about Batman. Finally, I wouldn't want to assign reputation to people based on agreement, but rather have a metric of how close one user's "taste" is to another user. Since the quality of a given person's taste isn't an objective value, similarity in taste is more important than correctness.

Ultimately the StackExchange Engine lacks the qualities of a good Recommendation Engine. It is designed to find the "correct" answer, a meaningless term when considering Recommendations have more than one correct answer and that they are correct in differing ways. Using SE for these things is like using Forums for Q&A, or using Microsoft Word to do a Spreadsheet: Just Plain Wrong.

So let's not repeat the same mistake our predecessors did, and try to abuse the engine into handling this content. I realize when you have a good hammer, how everything looks like nails, but let's try and use the right tool for the job here.

  • @badp I upvoted him initially, but I feel this deserves a more fully explanation than just: No. – tzenes Jan 23 '11 at 1:12
  • 6
    You make some very good points there. I guess you're right - Jeff&Joel made such a good hammer that it'd easily be able to drive screws... It's just, this hammer seems better than all screwdrivers I've seen so far – Tobias Kienzler Jan 24 '11 at 10:37
  • 3
    Interesting point about recommendation engines, but while not ideal, StackExchange still works very well for a lot of those poll questions that are supposedly "low value" – Casebash Apr 7 '12 at 1:06
  • 5
    @Casebash as we've learned that's only true for very small values of "well." The truth is, it's pretty bad for poll questions where the number of answers exceeds a handful (or maybe 2). The answers are too large, and people don't scroll very far; the highest voted answers tend to received the most vote; constant edits can change an answer from what a voter supports to something else; there are no summaries so you can see percentages. It's actually a pretty poor engine for polling questions... – tzenes Apr 10 '12 at 1:04
  • Very good points. Your sentence about having a metric of how close one user's "taste" is to another's reminds me of Netflix movie recommendations, which seem to be surprisingly accurate. They include an "average rating" for the movie, and also their best guess for what your rating will be when you see the movie. But yes, recommendation engines should be custom built according to what is being recommended. I could see a market for a general "recommendations engine" website, community-based like stackexchange...but that's far from the purpose of this site. – Wildcard Dec 30 '15 at 3:45

Stuff like polls, recommendations based on subjective constraints, puzzles, webcomics etc

If this is really what you want, you need to do it on another engine. Ours simply cannot support these kinds of frivolous, pure entertainment based content.

  • 11
    +1 because I believe there should be some limit. Puzzles and webcomics are not exactly Q&A. Asking for recommendations based on personal experience, however, are questions - just subjective ones. – Oak Jan 15 '11 at 8:50
  • 1
    I guess you and @Oak are right, there must be a limit. So what about only recommendations? E.g. hardware.superuser.com and recomendations@gaming.SE. I mean the game recommendation proposal would basically be such a thing, it's just a question of domain naming... – Tobias Kienzler Jan 16 '11 at 9:07
  • 3
    I agree about puzzles and webcomics being off-scope, but many of the polls are far from entertainment based questions and rather some of the most popular and useful questions on the whole site. – Casebash Apr 7 '12 at 1:07
  • I agree that sometimes poll-based questions are some of the most useful on the site. I'd prefer not to see puzzles and webcomics anywhere. – Chris Dutrow Oct 13 '14 at 1:21
  • I don't see how recommendations are inherently frivolous or purely of entertainment value. – Alien Technology Dec 29 '15 at 23:28
  • I don't see how recommendations are out of the scope of this engine. There are already recommendation sites. There were a lot of recommendation polls in SO and they had a lot of upvotes. Users found them useful. Deleting them, instead of just migrating them or enabling such sites (when there were users who found them useful) is opposed to the spirit of a community-driven... community. – Luis Masuelli Dec 30 '15 at 3:48
  • Yes, in the abstract, "another engine" might be the best solution. But the reality is that the SE engine is pretty good (or the "least bad"), compared to alternatives that actually exist. Of course, there is the always popular "other site" ... which no one can ever name ... or, compared to the SE engine, has many flaws. – Ðаn Feb 27 '17 at 15:58

Within the Stack Exchange network there have been some recommendations sites proposed, one in beta and one graduated:

If they had all survived, they would be the places to ask many of the recommendation questions that we see users try to ask on other sites. However it now looks like software and hardware are likely to be the only recommendation sites for the foreseeable future.

I think encouraging users to use the appropriate SE recommendation sites for those questions, and closing recommendation questions on all non Recommendation sites is the correct action. When their askers request migration to the recommendation sites, from the sites that I moderate, I do that too.

The Software Recommendations site works because its users are very strict about how questions can be asked (and answered). Non-recommendations sites struggle to have the patience to propose and apply guidelines that can shape recommendations into a form where they are not routinely closed for being too broad or unclear.

  • 5
    I've used the Software recommendations site. The problem is that it is too broad. If I want recommendations on command line graphics programs, I want to ask people who do graphics. Most of them don't read SR. If I want web log analysis I want to ask webmasters. Most of them don't read SR. If I'm looking for a Mac program, I want to ask on askdifferent. Etc. My signal is your noise. Even on the sites I use on a frequent basis, only a few percent of the posts are relevant. Without that tight focus SR is too noisy to browse. – Sherwood Botsford Nov 25 '16 at 15:02
  • 1
    If we can have a separate site like SR that still uses the SE engine, it doesn't seem like it would be too difficult to get the "high quality eyeballs" by allowing the same questions on SO with a separate tag. – Ðаn Feb 28 '17 at 22:10
  • @Ðаn If you think that, then just try to convince your communit(y/ies) to adopt that as a site-specific protocol. I suspect that you will see questions with such tags attract a lot of downvotes and votes for closing as too broad or unclear, whereas as the same question asked on Software Recommendations or Hardware Recommendations is likely to attract a lot of upvotes (and answers). – PolyGeo Feb 28 '17 at 22:16
  • No, it won't happen (and I find meta.so can be even "worse" than SO). This was as much about Jeff's claim of "you need to do it on another engine" ... if "opinion" (recommendations) can work on their own SE site, then that can (be made to) work on the "main" sites too which would address @SherwoodBotsford's comments. But my comments are moot as that's not "what is wanted", so it won't happen. – Ðаn Feb 28 '17 at 22:22
  • @Ðаn If a community does not want recommendations questions to be in scope (as defined by its Meta) then I think you will have an uphill battle getting them to stay open there. I think it is far better to go with the flow and post them at the site(s) that want them and have designed their scope and guidelines to accommodate them. – PolyGeo Feb 28 '17 at 22:27
  • @PolyGeo no argument. But it kinda flies in the face of the statement(s) (here and elsewhere) that SE can't support such an arrangement because it's not designed to do so; recommendation SE sites prove otherwise. Again, see my comment above. – Ðаn Feb 28 '17 at 22:30
  • @Ðаn You should take that up with Jeff, which I can see you are trying to do. For SE's current attitude to recommendation sites see discuss.area51.stackexchange.com/questions/24050/… which explains why Software (and to a lesser degree Hardware) Recommendations can be considered an exception. If I really wanted a software recommendation, and had sought one unsuccessfully at Software Recommendations, then I would follow my advice at meta.stackexchange.com/a/291573/215590. – PolyGeo Feb 28 '17 at 22:43
  • As I said in a comment for the area51, those are refreshing remarks. Notice the subtleties of "we're focusing our efforts elsewhere" compared to "it won't work." – Ðаn Mar 1 '17 at 14:14

Wouldn't Reddit be the perfect place for things like this? Stack Exchange is for Q&A, not just some other forum.

  • Reddit has things like this. What I'm missing there are the things that make SE so awesome, e.g. edit( suggestion)s such that formerly great answers can be updated by others – Tobias Kienzler Oct 31 '17 at 7:42
  • 1
    @TobiasKienzler If these alternative sites are really for subjective stuff, wouldn't it be sort of rude to edit someone's opinion into something else? – ColleenV Oct 31 '17 at 13:56
  • @ColleenV That's why I said subjective-ish ;) I still wouldn't want to see whole essays for example, where indeed an edit would rather be rude. But it's difficult to draw a line, and I wouldn't say I'm as convinced of such a "fourth place" as I was seven years ago. I still think it'd have its benefits, though... – Tobias Kienzler Oct 31 '17 at 14:51
  • 1
    @TobiasKienzler I get that this is coming back to haunt you a bit, but it's new to me :) I think that it needs to be made more clear in the original post how the SE model would provide a better way to discuss these things than other sites like Reddit. Frankly, I like SE and wouldn't mind if they started up alternatives to YouTube, Twitch, Reddit, et. al. so I could spend almost all of my Internet time here, but it seems like a lot to ask lol. – ColleenV Oct 31 '17 at 16:17
  • @ColleenV Now we're talking - don't forget a (finally) viable alternative to Facebook 😁 Hm, I'll think about it a bit. But please feel free to add your own suggestions as answer. Actually it's quite interesting to be reminded of my own opinions from years ago – Tobias Kienzler Oct 31 '17 at 20:01

I think Chat, the third place, would work fine for this. There just needs to be a way to persist threads and draw people's attention to them. You could use an individual room for each thread, but I think this would get cumbersome. Maybe they could create a way for someone to start a thread, label it with a question, and then have a list on the side of questions that are out there, and a way to filter to the question. Of course, this would be a little complicated and take some work to implement.

  • 4
    Chat? I'd really prefer something more organized and less time-sensitive. Plus I like being able to vote and view sorted by votes... Your modifications suggest part of it, but how different would it then be from the basic SE principle? – Tobias Kienzler Jan 19 '11 at 8:05
  • @Tobias, the problem is that the basic SE principles don't work well for subjectivy stuff. – Lance Roberts Jan 19 '11 at 16:00
  • 2
    true, but something like hardware recommendation has one answer that the OP will have decided on and yet it is subjective. Same goes for "What game shall I play on my LAN party (+details of course)?" etc. And the votes give a nice indication of the communities preferences despite the fact that there may be not one single correct answer – Tobias Kienzler Jan 20 '11 at 8:37
  • 2
    That happens even for technical questions. Imagine this: How can I solve passing non-sensitive Django settings data to my AngularJS config calls? After banging some time my head to figure how to create a good component for that, and failing several times, I could have posted the question in SO (in fact I was about to post that and, instead, I could narrow the problem a bit more). I figured several different ways to do that, and chose one of them. Imagine if users gave me the different possible answers I figured to my problem. I'd choose one of them: that choice would be subjecive. – Luis Masuelli Dec 30 '15 at 22:55

SO is not the place for puzzles or commics. At least SE is not. Maybe 4 would be a good place. Making a separate site for subjective questions seems to be too much. Some subjectivity is not a bad thing.

@Jeff is right, SE sites are not designed for polls or recommendations. However, although imperfect, some subjective questions provide great value. Calling subjective questions frivolous seems a bit harsh.

On StackExchange, any question dealing with pros and cons of different IDEs is squashed with a fury, but they often have tens of thousands of views. Someone apparently thinks that they are helpful. Isn't that the point?

Go ahead and vote me down for thinking that subjective has a place.

Polls would be useful, in particular if they stick to being non-blatantly subjective. Additionally, the questions should not be too broad but have a purpose. Examples:

  • List of frameworks for Websockets.
  • List of alternative HTTP servers or proxies to Apache2.
  • List of development tools which can export/deploy to multi platform mobile devices (e.g. say, Cordova).

Would be useful is those enumerations, however:

  • Become non-dupe. Having two lists of the same topics would be annoying.
  • Have a specific purpose: Enumerating all the existent programming languages would be useless and almost impossible.
  • Avoid using terms like best, worst, ... except when the data is relevant and according to specific criteria (e.g. the link not being so outdated or unmaintained)./

But yes: once upon a time the first 2 questions out of three I enumerated, existed in SO and had a lot of upvotes as Community Wiki. That tells nothing but that users found them useful. And IMO if users find them useful, then they should exist somewhere, although perhaps a special place, in SO.

Yes: I do support every sub-site that can hold content proven to be useful. Upvoted this initiative (at least just for polls).

Edit Yes, Software Recommendations is one example of how this initiative was partially implemented. At least for recommendations, while stuff like Book Recomendations or similars are or were planned to exist, and I still don't understand what's the alleged controversy regarding the SO engine, if a website like that already exists. So please: stop saying NO beforehand. Acting by a priori rejection is not the attitude that brought FLOSS and community-driven Q&A to the top of the search engines.

  • FYI, the SO/SE engine isn't open source. – dorukayhan Feb 18 '17 at 18:08

You must log in to answer this question.

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