15

To confirm my suspicion, I've just counted the average score over the last 100 questions tagged with . And I came up with -0.87

It seems the downvotes are given regardless of clarity of the question, whether the asker tried something and describes the efforts or not, even upvoted questions frequently have downvotes, and the answers (good, valid answers) are usually given in comments.

Sure, many of the questions are lazy and low quality, but not 59% of them! (that's the percentage of downvoted questions; only 18% is upvoted)

Does someone just hate this class of questions, downvoting them regardless of merit? Can the situation be improved somehow?

As for examples of reasonable questions that got downvoted:

Select entity by certian attributes - specifically, getting a player teleported upon throwing a specific type of item.

I'm trying to do a /give command where i get a crossbow with a custom firework - it's tricky because while the firework does indeed need to be in the player's inventory, the crossbow can be loaded with a particular projectile, and achieving having it loaded with that specific firework through commands seems either tricky or impossible through commands.

How can I set players to different teams? A simple question with an easy answer but doesn't seem to be anything wrong about it.

Minecraft: Keep player from falling - a very reasonable question about levitation or its alternatives; the asker tried quite a few options. The answer is quite unintuitive too: levitation effect with power 255.

Are there commands that can give players permissions like “can open containers” or "can build/mine - a very reasonable question with a reasonable, unintuitive answer.

Is there any way that i can test if a block is in a large area - a reasonable question, and yes, a duplicate, but of a question with terrible title that makes it quite hard to find if you don't know what to look for (the duplicate title actually suggests a wrong answer). Definitely not deserving -4.

  • 12
    Questions that are trivial or easily searched for usually earn downvotes, regardless of their quality, because they "show no research effort"... and that's a valid reason for downvotes. I often disagree with downvotes and already had a few discussions on Meta, but in this case the quality of questions really is pretty low compared to other tags. If that's not what you're looking for; do you have examples for good questions with bad score? – dly Feb 5 at 13:15
  • 1
    what's the average score over the last 100 questions with any tags? – arghtype Feb 5 at 17:59
  • 2
    @arghtype if my query on Data was correct, it's -0.02... deleted ones not included, as usual. – dly Feb 5 at 18:15
  • 2
    @dly wow, then it definitely looks like people are biased against Minecraft questions – arghtype Feb 5 at 18:29
  • 2
    @arghtype unfortunately it's not that simple.. yes, people vote a lot on Minecraft questions and mostly down. But since Minecraft players are usually a lot younger than the average gamer the questions are indeed worse than what people expect. Also the number of the last 100 questions doesn't help comparing the numbers. The age of those questions just isn't high enough to give people enough time to read & vote. I'd need the oldest Minecraft question taken into account and calculate the average of all questions posted after it to get something that we can actually compare. – dly Feb 5 at 18:39
  • 1
    @dly: Note, general Minecraft questions about gameplay are usually met with a much more positive response. Compare with the tag minecraft-redstone for example. It's specifically the commands (which is a domain definitely not targeted by the youngest players - it's a thing for map creators, not actual players) that get the hate. – SF. Feb 6 at 0:03
  • @dly also, let me disagree with "The age of those questions just isn't high enough to give people enough time to read & vote." The oldest of these 100 questions was asked in November. – SF. Feb 6 at 0:14
  • Hypothesis: downvoters feel these are more related to "programming" than gameplay. – Iiridayn Feb 6 at 0:52
  • 7
    I think it's more that minecraft-commands tend to be along the lines of "this is what I want, now write the command for me", which elicits a negative reaction (fairly, IMO). It's unusual for a minecraft-commands question to include details on what the asker tried, etc. – Schism Feb 6 at 5:09
  • @SF. November to February is a long time. The oldest question out of all tags is probably only a couple of days old. – dly Feb 6 at 7:09
  • And I know a lot of young people trying out redstone, commands, craftbook, etc. On my server the older guys just build and the younger ones experiment with craftbook and stuff. – dly Feb 6 at 7:22
  • One of the reasons could that there are multiple versions of Minecraft, and command in 1 version don't work in the other version, since there are syntax changes between the versions – Ferrybig Feb 6 at 14:54
  • 6
    I think Schism's hit the nail on the head. On StackOverflow, questions that ask for code but don't show what they've tried are not only heavily downvoted, but off-topic completely. Though I don't tend to look at any MC-command questions, the ones that I do look at rarely show any work and mostly just demand we do the work for them. Of the six examples you've put in your question, five of them don't show what they've attempted and the remaining one gives a general summary what they did but doesn't give specifics. It's no wonder then that the questions are downvoted; we aren't mind readers. – Mage Xy Feb 6 at 22:43
  • Here's the most important unanswered question: it costs 1 rep to downvote. who has a ton of rep to throw away? – Alexander Day Feb 7 at 15:48
  • 3
    @AlexanderDay No rep cost to downvote a question, only answers. – JMac Feb 7 at 18:12
17

My thought about it was that it could be the same old "new users never read how-to-ask" issue. To prove my point I've tried some SQL-fu on Data, here are my results.

  1. Last 1000 questions for :
Average_Score | Score_Sum | Post_Count | Reputation_Bucket
2.188         | 116       | 53         | over 400
2             | 4         | 2          | 301 - 400
1.266         | 19        | 15         | 201 - 300
1.089         | 73        | 67         | 101 - 200
0.061         | 6         | 99         | 051 - 100
-0.484        | -370      | 764        | 000 - 50
  1. Last 1000 questions over all tags:
Average_Score | Score_Sum | Post_Count | Reputation_Bucket
3.846         | 1027      | 267        | over 400
2             | 68        | 34         | 301 - 400
2             | 82        | 41         | 201 - 300
2.157         | 261       | 121        | 101 - 200
3.079         | 117       | 38         | 051 - 100
-0.467        | -233      | 499        | 000 - 50
  1. Last 1000 questions without :
Average_Score | Score_Sum | Post_Count | Reputation_Bucket
3.704         | 1089      | 294        | over 400
1.974         | 75        | 38         | 301 - 400
2.024         | 83        | 41         | 201 - 300
2.273         | 291       | 128        | 101 - 200
3.658         | 150       | 41         | 051 - 100
-0.319        | -146      | 458        | 000 - 50

General trend can be seen - new users' questions are not well received regardless of tags, and for we see ~40% more questions from new users.

Here is main query permalink to play with. One problem with it - I cannot figure out how to get user's reputation at the time of question creation.

  • 1
    I would mention that the second bucket (51-100) is also massively different from the average – Federico Feb 9 at 20:00
  • Are you assuming low reputation = new user? The OP's hypothesis could also explain this data; if a user asks lots of minecraft-commands questions and such questions tend to be downvoted, that user will remain in a low reputation bucket. – stewbasic yesterday
  • @stewbasic yes, that was my assumption – arghtype 10 hours ago
6

I think I should answer this, since I have downvoted 421 posts on this site, mostly questions, mostly (since I'm most active there) and 4 of your 6 example questions.

The answer mainly comes down to what Schism said in a comment: These questions usually have bad grammar, ramble a bit and eventually give a very unspecific description of what they want, then expect us to do it for them.
This is probably because Minecraft's target group is dominated by small children, who have no patience to read the rules, to put effort into their question or even to use a different wording than they would if they were talking.

Also, most of these questions have a score of -1, which is nothing special, since there's always a bit of noise in the vote behaviour. I know one user who I suspect to have downvoted almost every question that wasn't very good in the past. And they're allowed to do that. I don't think they do it anymore, but were for a long time.

And yes, I do indeed think that 59% of questions with this tag are "lazy and low quality", probably even more. I've already adjusted my behaviour for these: I used to treat all questions somewhat equally, tried to scrape the smallest bit of whatever I understood from them together and at least leave a wiki link or something in the comments (in the first months here I even often wrote a complete answer), but I noticed that it stresses me. So for my own health, I now just change tags if they're wrong*, downvote if it's really bad, vote to close if appropriate and then get out of there. I follow and for questions like t h e s e o n e s where I can apply my knowledge about the game to solve interesting problems and to build up a knowledge base for future reference on this site, not for "gimme the codez" questions or ones that can be solved in a minute using Google.

*Usually to or or similar. Or removing from questions. You might notice that there are only 33 left that have that combination of tags, there used to be about 200 before I cleaned that up.

Now to your 6 examples:

  • "Select entity by certian attributes Minecraft Bedrock Edition": The first thing I notice negatively is this "get it?", which shouldn't be part of the post. Wow, they made a pun, great. Not important for the question.
    Then they write down a command they tried, which is good, even with code formatting, but then it just ends with "doesn't seem to work right". In what way? Does it give out an error message? Does it do nothing? Does it make your computer explode? Also, that command had multiple pretty obvious syntax errors, which could have probably been easily solved if the asker looked at the error message.
    Then they ask how to differentiate the egg some other way, which is similar to the "XY problem": They ask how to fix their attempt of solving the problem instead of asking how to solve their real problem. In this case other alternatives would be dropping (which the asker excluded for reasons I do not know), shooting a bow, spinning in circles, whatever, there are endless possibilities to trigger something. But to know which one fits the person's need best, we would need more information.
  • "I'm trying to do a /give command where i get a crossbow with a custom firework": The only one of the four sentences that doesn't have a grammar error in it has only four words.
    Looking back, I probably shouldn't have downvoted the question, only closed as "unclear", since what I assume the asker probably means is changing what a crossbow shoots, which is indeed more complicated than it sounds. But I don't know for sure, because I don't understand the question text at all.
  • "How can I set players to different teams?": This is a very basic question, a quick Google search could have given that person the answer much quicker.
  • Minecraft: Keep player from falling: I didn't downvote this question and it's on a positive score, but two people did downvote it. Probably because they assume that everyone heard about levitation 255 already. But I find that it's a decent question: The asker lists different things he tried, he specifies his version, ultimate goal and the problems he encountered. Nothing wrong with that. Another possibility is also that the downvotes were made before the grammar of the post was corrected. But as long as it's understandable, bad grammar alone doesn't justify a downvote for me.
  • "Are there commands that can give players permissions like “can open containers” or "can build/mine": I didn't downvote this and the score is 0. The only downvote could have been random or because stuff like this was asked a few times before. But there are many wordings for what the asker wanted, so it's understandable if he didn't find a definitive answer. Alternatively, the downvote could have also been because the asker shows a lack of knowledge about Minecraft, but that's also no reason for a downvote in my opinion, since one of the best ways to gain knowledge is to ask questions.
  • "Is there any way that i can test if a block is in a large area": This questions sounds like it might be the XY problem again, but I don't know, since there's very little detail. The only mentioned attempt of a solution doesn't even include a full command and doesn't say what's wrong with it, it should work. They also list no other attempted solution or a reason to check for air (which might help against the XY problem).

In general, I think the better question is: What could be done against the fact that so many of these questions are in fact bad? Obviously we can't just ban all children from this site, they need help, too. But I would actually welcome even more downvotes for extremely bad questions, because then they would be more likely to be picked up by the automatic deletion process, for example at a score of -4 usually.

One thing that I still have planned is making some broad, really detailed FAQs that can then be used as duplicate targets, to save everyone here a bit of headache. For example one of them could be "my super long Minecraft command isn't working" and it would instruct how to reduce it to the minimum reproducible example, how to solve common issues with long commands (like unbalanced brackets, how to de-nest commands (like in one-command creations) and so on.

  • It seems to me you don't understand a lot of questions which I understand perfectly. Seems to me the problem is not with the questions but with your understanding, At the moment I listed these questions here, they were all downvoted. And in cases of the XY questions you see, I'm seeing even if solving problem X may not be the most optimal way to achieve Y, it's a way to solve a whole large family of problems (I can imagine a dozen reasons why you'd want to check if given area contains a block of given type) and so solution X is more valuable than most specialized Y. – SF. Feb 8 at 12:24
  • 2
    It's mostly not that I don't understand a question at all, usually I can guess. It's more that a question is supposed to be clearly understandable, without having to guess. And it's still possible to guess wrong if it's not clear, which then leads to different interpretations and potentially lots of invalid answers going in totally different directions. – Fabian Röling Feb 8 at 12:36
  • 1
    Downvoting doesn't help anyone in these cases. It's often difficult for the asker to spot what is unclear in the question - I personally found that out the hard way, details completely obvious to me would appear non-obvious to the readers, and unless someone tells me to clarify something I won't have a clue it needs to be clarified. A downvote in these cases is the least helpful and least useful course of action you can take. – SF. Feb 8 at 12:43
  • @SF. I vote to close unclear questions as unclear and usually also leave a comment. As I said in my response to your second example, that question probably didn't deserve the downvote, based on my guess what it means. I just didn't undo it just now because that would require editing it, which would use up that question's only chance to be automatically put into the reopen vote queue. – Fabian Röling Feb 8 at 12:46
  • 3
    @SF. downvoting helps other curators know that the question probably isnt worth their time looking at unless it has been edited. – Dragonrage Feb 8 at 21:04
  • 4
    @Dragonrage I disagree. Don't decide on what to do with a question based on vote score. Everyone should have their own opinion. A few people downvoting a question should be no reason to not read it. Ever. Maybe it's just a legit question with bad grammar/spelling.. we all know that questions like that get easily downvoted. – dly Feb 9 at 1:02
  • @dly Exactly as they should be. If you want to spend time fixing up those types of questions, feel free; nobody is stopping you. Those votes are to signal to everybody, curators, new users, and future readers, too, that the post has problems. There's a reason a question falls off the main page at -4 and you have to actively go looking for it, and that's because the system is being told that question is bad enough to require either major effort to salvage, or that it should likely be closed and deleted. – Frank Feb 10 at 14:23
  • 1
    @Frank that's bandwagoning and definitely not curation. Of course a question with a score of -4 is most likely a bad question and somewhat deserves to be removed from the front page, but it's a long way from 0 to -4. Feel free to ignore it, that's fine and your choice. But never downvote a question because it already has bad score. And choosing to ignore a question with bad score can't be curation either... how would that even be possible? Someone has to throw the flag to make it pop up in the reviews. Or someone takes it to Meta to discuss how wrong it was.. even that can (and did) happen. – dly Feb 10 at 17:51
  • @dly Actually, no. Nobody has to flag to make it appear. The system automatically puts it into the low quality review queue once a post meets certain criteria. Post length, current score, even something as an uncapitalized I, all play a part. Nor is it bandwagoning; I didn't even bring up anything related to how you should vote, so that's strawmanning. – Frank Feb 10 at 17:56
  • 2
    For the record, the vast majority of Minecraft questions are junk, and eminently downvote worthy in their own right. I don't care who asks them; that doesn't play a part in the decision making, and shouldn't. If we get a question that asks us how to do something, shows zero effort to solve it themselves, it gets a downvote. If, over time, it turns out others have that same problem, a good answer can make a crappy question useful. But first, it has to have an answer, and bad questions won't usually get one. – Frank Feb 10 at 18:00
  • 1
    @Frank sorry, if I misunderstood the voting part, but that's the point of this Meta so I probably misinterpreted this part. I still think that bashing on new user's questions is plain wrong. And looking at the numbers in the other answer by arghtype this is a big problem here. And for a question to appear on the VLQ queue should have serious grammar issues, but it won't detect content problems. – dly Feb 10 at 18:01
  • 1
    @Frank I already said that in the very first comment on this Meta: Questions that are trivial or easily searched for usually earn downvotes, regardless of their quality, because they "show no research effort"... and that's a valid reason for downvotes. I often disagree with downvotes and already had a few discussions on Meta, but in this case the quality of questions really is pretty low compared to other tags. If that's not what you're looking for; do you have examples for good questions with bad score? – dly Feb 5 at 13:15 yes, there's a lot of trash, but scaring off users won't help. – dly Feb 10 at 18:03
  • 1
    @dly If downvotes scare off users...that's misaligned expectations. If you're abstaining from downvoting bad questions because of who posted it, that's a misuse of the voting system. – Frank Feb 10 at 18:04
  • 1
    @Frank you're right that we need to vote, but also guide new users through the system. If we don't do that - who does? Downvoting as we currently do won't fix this issue. I've even stopped downvoting questions which already have a negative score (unless they're complete trash) and tried to comment on what the OP needs to fix in order to help. If that doesn't fix it we've at least tried. The statistics don't lie.. we have to do something to raise the quality of questions (without just deleting them). The guys on SO already try that. – dly Feb 10 at 18:08
  • 1
    Let us continue this discussion in chat. – Frank Feb 10 at 18:11

You must log in to answer this question.

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