13

We have a problem. Three of our currently highest scoring questions are no longer acceptable by current site rules, and may never never have been acceptable to begin with. It's likely the only reason they still exist today is that they each contain excellent answers, and valuable information.

The questions I am referring to are as follows:

If these questions had been asked today, they would be closed as Not a Real Question since they are both 1) Too Broad and 2) Vague. However, for some reason, they were never closed when they first posted, and then due to the excellent answers provided by Tzenes users have been reluctant to close them ever since. This is all fine and dandy, except that they are nothing more than a broken window. We can't exactly go around closing equally vague and broad questions while these three still exist. It just doesn't make sense.

So, we should close them then, right? Well, no. The problem is that closing is a temporary state, and any question closed as anything other than a duplicate (or a historical lock, see below) should eventually either be re-opened or deleted. They are not meant to sit in closed limbo forever, that's just not what closing is intended for.

Okay, so just delete them then, right? Well, no, that also has problems. Because although the questions may no longer be acceptable (in their current form) by the site, they contain some of the best and most useful answers currently hosted by our site. Yes, they no longer meet the current standard of the site, but the reason they exist is because we messed up. These questions should never have remained open long enough for Tzenes to post his answers in the first place. They should have been closed, and then either fixed and re-opened, or deleted. Deleting such excellent information from our site because we screwed up is not a good solution in my eyes. After all, one of the reasons Stack Exchange exists is to be a resource for information like this.

The way I see it, we have four possible solutions:

  • Attempt to Reword the Question

    The ideal solution would be to find some way to reword the question to make it acceptable on the site, while still having the current answers valid.

    The benefit of this solution is that the question would no longer be a problem, the information would still be on the site, and could the question can still be modified and answered by users in the future.

    The problem with this solution is that this could prove a very difficult task, if not impossible.

  • Close with a Historical Lock

    This the type of situation a historical lock exists for. A question which no longer meets current site standards, but is an excellent source of information.

    The benefit of this solution is that it keeps the information on the site, and gives us a justifiable reason why similiar questions should be closed despite the existence of these questions.

    The downside to this solution is that we've managed to avoid using historical locks up until now, which is a good thing. They are really meant to be something of a last resort, when we have no other viable solutions.

  • Delete the Question

    Deleting the question is another solution.

    The benefit to this solution is that it would remove the broken window, and allow us to close future questions without any concerns.

    The downside to this solution is that all the valuable information from the question would be gone.

  • Do Nothing

    The final option is to simply do nothing, but come up with a reason why this question is an exception.

    The benefit to this solution is that the question and it's information remain, and more answers can be added in the future. It would also provide us with justification why this question remains open when similiar questions are closed.

    The downside to this solution is that I feel like any justification for keeping this open would be flimsy, and would just pave the way for further exceptions, which will lead no where good.

So, Arqade community, what shall we do?

  • 13
    I dispute the notion that these questions are inherently problematic. – LessPop_MoreFizz Jun 16 '12 at 2:54
  • @LessPop_MoreFizz The questions are just asking for build orders for each of the three races, which is a fairly solid example of "bad subjective", and a bad build question. They also ask for currently popular builds, but in a game like SC2 that is a thing which is always changing. – Wipqozn Jun 16 '12 at 20:31
  • Here are two examples of identical questions which were closed. If we think the SC2 ones are fine, then obviously we need to re-open those, along with any other questions which are of a similiar nature (of which I'm sure there are many). – Wipqozn Jun 17 '12 at 14:10
  • 2
    First off, those are two links to the same question. Secondly, I don't see how they're all that different from the various D3 Build questions we have. The question of popularity is off topic and should be edited out of them, to be sure, but asking for an effective build seems like the epitome of good subjective - remember, just because a question can have many answers doesn't mean it's bad, so long as any one of those answers could stand on it's own, and the answerer can back up that answer with real facts and experience. – LessPop_MoreFizz Jun 17 '12 at 14:14
  • First, here's the other question I meant to link to. Now, my problem with these questions is not that they are subjective, but that they are overly broad, which fits our "Not a Real Question" close reason quite well. now I'm willing to grant there may be something special about build questions that gives them more leeway on this topic, but we'll need to come up with a concrete reason of why this is the. Furthermore, we at least need to be consistent in our closing/deleting of said questions... – Wipqozn Jun 17 '12 at 19:18
  • ...since I don't see any difference between the closed/deleted questions I linked, and the the three Starcraft 2 answers I included in my question. There has been talk on this topic before, but it never really went anywhere. Perhaps it's a topic that needs revisiting. – Wipqozn Jun 17 '12 at 19:20
  • Here are some more examples of closed build questions, showing that we appear to be inconsistent with how we deal with these questions. – Wipqozn Jun 17 '12 at 19:26
  • 8
    One key difference is that despite the use of the homonym, a build order in an RTS is emphatically not the same as a build in an RPG - it has much more in common with an opening in chess. – LessPop_MoreFizz Jun 17 '12 at 19:53
  • The only problem I see with these questions is a potential one -- people seeing them and creating bad recommendation questions, citing these as examples. Minor edits could tweak them away from that, I don't see how it's "very difficult or impossible". I agree with LessPop_MoreFizz, basically. There is nothing so problematic with these questions that it warrants deletion, certainly, and I can promise I would be a colossal pain in the ass about it if tzene's epic answers were made unavailable :P. I know you don't support it, but you did mention it! – Matthew Read Jun 18 '12 at 3:16
  • 1
    @MatthewRead: Users citing these as examples are exactly my problem. I personally have trouble seeing how they could be edited, but if you or LEssPop would like to take a stab at it that would be great, because I don't want to see these questions deleted anymore than you do. – Wipqozn Jun 18 '12 at 13:44
12

There's nothing wrong with these questions to begin with.

You say that the problem with these questions is that people point to them when you try to close their bad questions.

The problem isn't these questions, the problem is that you're bad at communicating why the bad questions you're trying to close suck. Blindly hiding behind a citation of a site policy doesn't work. If you're going to call a question bad, you need to explain why it's bad, and you need to do so in terms that a new user can understand - don't use jargon and call it a 'list question' or a 'recommendation question'. Just tell them that their question is too broad to be able to cleanly and definitively answered. Yes, that's a mushy and kind of subjective standard. But that's why closing and deleting questions isn't done unilaterally. The whole reason it requires multiple votes from trusted users to do those things, is because it requires judgement and understanding of the rules - not simply the ability to recite them.

So what should we do about them?

Nothing

Vote on them, if you so please, but they should not be deleted or locked or closed or any other such janitorial thing.

Addendum:

Wipqozn has asked me to address the broader question here, beyond these three questions, regarding build order and build questions in general.

So, first off, let me say this: Build order questions are not analogous to build questions in RPGs. A build order in an RTS is more akin to an opening in chess - and while they can be many and varied, they generally don't require any more specificity from the asker in terms of seeking one than "I want a build order for [army]." Optionally, "That is good at countering [thing]" can be helpful in getting a more detailed or useful answer, but the general case question is still useful, just as it's still useful to know "How are verbs ending in -er conjugated in spanish?" Even though there are a variety of irregular verbs to which the answer won't apply.

Now, Build questions in general are a bit thornier, but again, a blanket ban is not the appropriate solution. Yes, a lot of character build questions are really bad questions; cf. . This does not make questions about character builds inherently bad. The main thing is that the question needs to reflect some actual experience - a concrete problem faced by the player. "I want a build that is totally cool and sneaky and uses big swords and magic lasers and stuff" is not a good question. It's unfocused, vague, and doesn't really provide any useful information for answerers to provide a good answer. The thing that gets tricky here, is that the criteria that separates a good build question from a bad one is very often specific to the game in question. A good build question for is generally going to specify a class, any skills that are 'mandatory' to the asker and must be built around, and a desired play style (tanky, hardcore, low-gear dependency magic-find, whatever). A good build question for by contrast, can generally be a bit more vague, and will generally expect it's answers to be as much or more focused on play style as it is on skill selection. Good build questions for tend to be either extremely specific (and focus on whether certain specific skills or spells are useful or synergize as they appear to), or very general and focused more on the meta-strategy of how to pick skills, rather than on particular skill choices. Good build questions about an MMO like will tend to be fairly mathy and depend on optimal DPS/HPS/Threat/Effective Health numbers - however, even with these, there may be multiple answers as different builds will make different tradeoffs and a multiplicity of answers arguing for these many variations is a good thing.

Remember, just because it's possible for a question to have more than one right answer, shouldn't mean it's not welcome. The problem isn't questions with more than one right answer. The problem is questions where any individual right answer cannot stand on it's own. When you see a question and are thinking about closing it for being too subjective, ask yourself: If this question only ever got one answer, would that answer be able to be complete and useful to the asker? Can that answer be backed up by facts, references, or experience? Does the question appear to have a grounding in a real understanding of the mechanics of the game, or does it appear to be a thought experiment, idle speculation, or poorly concealed wishlisting?

So I'll reiterate: What should we do about build questions, as a general class?

Nothing

And I'll add to that that we should continue to be vigilant in closing and downvoting bad questions, whatever they happen to be about. Furthermore, I'd suggest that if you see a question of this class that you aren't sure about, and you don't have any experience with the game in question - maybe hold off on that close vote. I'm not saying don't cast it, but perhaps try asking in chat, or with a comment, for some clarification as to whether a question really is too vague to be answered. We're supposed to be a Q&A site for experts, but not everyone is an expert in everything. And sometimes it takes an expert to be able to tell quality content from junk. Sometimes a question that looks perfectly useful to someone that has never played a particular title is obviously and completely a stupid 'what if' to anyone with a bit of experience. If you don't feel qualified to make that judgement call, it's perfectly acceptable - I'd dare say appropriate - to step aside, and wait for someone who is to do so.

Finally, I really do want to reiterate that I think above all, this is a problem of people being bad at communicating why questions are bad. Pro Forma comments and jumping to conclusions based on trigger words in a question is a bad habit that is rampant on this site, and it really is the single biggest problem here. If you really feel it's necessary to leave a comment on a question that you want to close or down vote, make damn well sure it's a good one, and not some canned copypasta. Engage with the asker, and explain exactly why you think their specific question is a poor fit. And that doesn't mean telling them which one of our special magic boxes of not-allowed it happens to fall into. That means explaining exactly why their question is in that box. Don't tell them "Discussion questions are prohibited by the FAQ". That doesn't mean anything to them. Tell them "Your question solicits opinions in a way that doesn't really fit well with our rep-based, vote-based, Q&A format. Since Arqade isn't a forum, we don't have the tools to really engage in the back-and-forth of a discussion. Instead, we expect questions here to be able to have clear, definitive answers. There can be more than one of them, but they should be able to stand on their own, and they should be based on facts and experience - not opinions."

And I swear to god if one of you people starts copy-pasting that or something similar onto questions you're voting to close YOU WILL PAY FOR IT.

  • 1
    +1. Most of these, at their heart, really boil down to "I'd like help understanding how to effectively play X.". These are the exact sort of questions where expert knowledge is useful. Sometimes they're badly asked (i.e., asking what's "popular"), but really, the asker directly doesn't care about what's the most popular. What he's really asking is what is effective. We can reword such questions if they really bother us. Some of my highest voted answers have been to these sorts of questions. – Sterno Jul 15 '12 at 9:47
  • And while I understand that "high votes" doesn't mean everything, it's clear from my own experience plus the votes on the questions linked in the OP that people care about these things. It seems ridiculous to me that we'd close them because we think they're badly worded when clearly they will typically generate useful, quality content. – Sterno Jul 15 '12 at 9:49
  • +1 I've been going back and forth between these questions on awhile, and you've managed to grab on the one point that kept giving me trouble: What's a good/bad build order questions varies from game to game, and in StarCraft 2 build orders are closer to chess. Although there are a large variety of opening, there are normally only a handful of openings considered strong, and most openings can usually be reduced to one core type of build (i.e. hatchery first, zergling rush, et cetear) – Wipqozn Jul 18 '12 at 18:50
  • 1
    I agree with Sterno. These sort of questions call for experience based answers rather than 'recite the wiki' answers, which IMO is much more valuable type of answer. Anyone can read a wiki, but not everyone has expert experience with the game. That's a niche we fill. – Robotnik Feb 16 '14 at 1:26
  • Interesting parallels with char-op (character optimization/build) questions on RPG. – C. Ross Feb 22 '14 at 17:14
  • @Cross I'd love to hear you elaborate on that... – LessPop_MoreFizz Feb 22 '14 at 17:16
12

I am okay with any of the solutions except deleting. There are plenty of "legacy" posts on StackOverflow and in the SE system; there must be a way to similarly preserve these posts.

These questions and many of their answers are excellent in their own right. And no doubt they bring in plenty of site traffic too. Even if they have a negative "broken window" effect, I believe they have a greater positive effect in terms of showing excellent answers, lots of insight, and the community rallying to centralize knowledge and build archives of information.

  • 4
    A Historical Lock seems like the best way to go here. – fredley Jun 21 '12 at 15:50
  • Wasn't there a podcast or something talking about historical questions on SO that would now be considered off topic? I'll have to look for that in the morning. – MBraedley Jul 15 '12 at 2:34
  • 1
    Found it! Well, one of them, I think there's another podcast earlier, but couldn't find the earlier one. – MBraedley Jul 15 '12 at 13:13
3

My preference is to first attempt to salvage the question to make it acceptable by current site policy. If that's impossible, then I'm in favour of using a historical lock. As far as I'm concerned doing nothing or deleting the questions are terrible solutions, for reasons I touched on in my original post.

  • 1
    Doing nothing is worse than deleting, but they're still both worse than improving or locking. – Mr Smooth Jun 16 '12 at 2:24
2

I agree completely with rewording the question to allow for a discussion that can be used a resource for new SC2 players. I think the question wording isn't too problematic to fix.

After rewording the questions, how about changing the questions and answers to community wikis so that the content can be periodically updated to reflect changes in gameplay?

2

Just use a historical lock, you did say that this is exactly what they are used for.

I dont exactly see any downsides to this except spoiling your no historical locks streak.

The last resort argument makes it sound like you are nuking the question from orbit.

Just do it.

You must log in to answer this question.

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