4 ms·

This is a waste. Few people upvote questions because they are well-written; by and large it is that they had the same question and you asked it before they had

This is a waste. Few people upvote questions because they are well-written; by and large it is that they had the same question and you asked it before they had to do so themselves. This heavily biases upvotes on extremely novice questions from 8 years ago (before they were all asked), and gives new question askers a bad example of what a good question is ("I just did the same thing this person did, and they got 200 upvotes, and my question was closed in 2 minutes!"). Question askers get the reward of having their question answered. I think votes on questions should be done away with entirely. Questions that should be exemplified can have some other metric (stars for "this is well written and generalizes easily for other users"), and low-quality questions should be flagged or closed, which at least gives some sort of reason. Downvotes aren't useful to askers because they have no idea what they're doing wrong when they receive them.

tenebrisalietum 7y ago
Keeping the participation friction as low as possible ensures the maximum number participate, which ensures questions have maximum visibility. Changing the simple vote up/down process with something requiring to specify a reason can reduce the number views a question gets. Also nothing is stopping a downvoter from leaving a comment explaining why the question needs improvement, other than lack of reputation (you need a minimum reputation for your downvotes to be visible anyway). People who care to explain their downvotes can already use the existing process. Since comments can be upvoted (but not downvoted), there is already an incentive in place for that as well.
- yabadabadoes 7y ago
  Many people mention they avoid commenting on downvoted questions for fear of the wrath of the asker. People seem to feel differently about close voting, perhaps since they are collected and presented as a group. I don't really get the point of the vigilance toward bad but not intentionally spam questions. The question # has been used, derank it and call it a day until someone sees a reason to revive it.or the asker ASKS for help improving their question.
  - jrumbut 7y ago
    The specific change I would like to see is to really raise the bar on what is considered a duplicate. I think a rule could be "if someone with a decent track record (for some definition of decent, maybe 100 points? 1000?) felt the need to ask it, it's not a duplicate." The bar could be raised in other ways as well, but this one I think would reward people who have done the work to be a positive part of the community. We all have varying degrees of ability to generalize the principle of one answer to another. The other part of this is that you get better at asking questions through practice, which is hard to get early on. At this point, most problems I can't answer are good questions or too specific, but it took a while and I don't want SO to be the place where only those with 5-10 years in can get help.
    - yabadabadoes 7y ago
      It seems to me like someone with few points is least likely to understand a solution to a duplicate of their problem.. I don't really see the harm in linking it and still letting people answer or comment on the proxied solution. SE spends half the time saying an overlapping diversity makes it easy to find a question and answer that fits and the other half eliminating that diversity (and bringing non-adept people to dead ends in their searches that land on SO.)
      - jrumbut 7y ago
        My proposal is intentionally moderate. If I had my way, I'd go further, but this is something I've thought about and think it may be palatable to those at SO who want to preserve the strictness and maintain respect for those who contribute. You still have to put in a lot of work to figure out how things work, but if you do the work you get credibility. I see it as a way to meet in the middle.
    - Bnshsysjab 7y ago
      I also think being able to override votes to nuke an incorrect answer would be handy, I recently begun getting involved in HTML/CSS for a somewhat obscure project, some of the older answers around front end topics are extremely dated and incorrect by today’s standards (flexbox, for example, deprecates a bunch of hacks). Sure there might be the correct response 3-4 answers down, but I shouldn’t need to dig for that.
papln 7y ago
Votes (both up and down) on questions are a way to prioritize questions that need answers. Upvotes should be welcomed from all (fellow users who want an answe), but downvotes should be welcomed only from reputable answerers who took time to read the question and decided it's poorly formed. Answered questions would be fine having their vote button disabled.
melling 7y ago
I’m glad they finally came around. I used to ask lots of questions. e.g. https://stackoverflow.com/questions/2170528/writing-hello-world-in-emacs https://stackoverflow.com/questions/2170528/writing-hello-wo... Unfortunately, the moderators and I had a disagreement about the importance of asking questions without attempting to first solve the problem. Well written (and well-defined) questions are the key StackOverflow. From the questions, you could extract a Rosetta Stone of programming. An experienced programmer could learn most languages on the job.
sytelus 7y ago
I sort of agree with you. The demand and supply for askers and answerers is imbalanced: questioners: 24035 answerers: 10888 It's not that we are running out of people who have questions. The reward system needs to reflect this imbalance. Also, I'm afraid there will be increase in questions and corresponding increase in lack of answers demotivating people who genuinly wants to ask without regards to reputation. To curb race to first asker, SO should implement exponentially discounting reward. This makes final commulative reward finite.

tenebrisalietum 7y ago

Keeping the participation friction as low as possible ensures the maximum number participate, which ensures questions have maximum visibility. Changing the simple vote up/down process with something requiring to specify a reason can reduce the number views a question gets. Also nothing is stopping a downvoter from leaving a comment explaining why the question needs improvement, other than lack of reputation (you need a minimum reputation for your downvotes to be visible anyway). People who care to explain their downvotes can already use the existing process. Since comments can be upvoted (but not downvoted), there is already an incentive in place for that as well.

yabadabadoes 7y ago
Many people mention they avoid commenting on downvoted questions for fear of the wrath of the asker. People seem to feel differently about close voting, perhaps since they are collected and presented as a group. I don't really get the point of the vigilance toward bad but not intentionally spam questions. The question # has been used, derank it and call it a day until someone sees a reason to revive it.or the asker ASKS for help improving their question.
- jrumbut 7y ago
  The specific change I would like to see is to really raise the bar on what is considered a duplicate. I think a rule could be "if someone with a decent track record (for some definition of decent, maybe 100 points? 1000?) felt the need to ask it, it's not a duplicate." The bar could be raised in other ways as well, but this one I think would reward people who have done the work to be a positive part of the community. We all have varying degrees of ability to generalize the principle of one answer to another. The other part of this is that you get better at asking questions through practice, which is hard to get early on. At this point, most problems I can't answer are good questions or too specific, but it took a while and I don't want SO to be the place where only those with 5-10 years in can get help.
  - yabadabadoes 7y ago
    It seems to me like someone with few points is least likely to understand a solution to a duplicate of their problem.. I don't really see the harm in linking it and still letting people answer or comment on the proxied solution. SE spends half the time saying an overlapping diversity makes it easy to find a question and answer that fits and the other half eliminating that diversity (and bringing non-adept people to dead ends in their searches that land on SO.)
    - jrumbut 7y ago
      My proposal is intentionally moderate. If I had my way, I'd go further, but this is something I've thought about and think it may be palatable to those at SO who want to preserve the strictness and maintain respect for those who contribute. You still have to put in a lot of work to figure out how things work, but if you do the work you get credibility. I see it as a way to meet in the middle.
  - Bnshsysjab 7y ago
    I also think being able to override votes to nuke an incorrect answer would be handy, I recently begun getting involved in HTML/CSS for a somewhat obscure project, some of the older answers around front end topics are extremely dated and incorrect by today’s standards (flexbox, for example, deprecates a bunch of hacks). Sure there might be the correct response 3-4 answers down, but I shouldn’t need to dig for that.

papln 7y ago

Votes (both up and down) on questions are a way to prioritize questions that need answers. Upvotes should be welcomed from all (fellow users who want an answe), but downvotes should be welcomed only from reputable answerers who took time to read the question and decided it's poorly formed. Answered questions would be fine having their vote button disabled.

melling 7y ago

I’m glad they finally came around. I used to ask lots of questions. e.g. https://stackoverflow.com/questions/2170528/writing-hello-world-in-emacs https://stackoverflow.com/questions/2170528/writing-hello-wo... Unfortunately, the moderators and I had a disagreement about the importance of asking questions without attempting to first solve the problem. Well written (and well-defined) questions are the key StackOverflow. From the questions, you could extract a Rosetta Stone of programming. An experienced programmer could learn most languages on the job.

sytelus 7y ago

I sort of agree with you. The demand and supply for askers and answerers is imbalanced: questioners: 24035 answerers: 10888 It's not that we are running out of people who have questions. The reward system needs to reflect this imbalance. Also, I'm afraid there will be increase in questions and corresponding increase in lack of answers demotivating people who genuinly wants to ask without regards to reputation. To curb race to first asker, SO should implement exponentially discounting reward. This makes final commulative reward finite.