Technically, I think this is more Trike’s problem, so it would be more like ‘how much more of Trike’s time is this supposed to take up or be diverted from other LW maintenance/improvements?’
Several months ago, another user offered to set up a fork of the reddit enhancement suite that could achieve this and other features for users interested in them, but the project never took off. Arguably, this is a poor way of solving the problem, because it requires opting in, and most users would continue to see the old look instead. But it would be better, perhaps, than doing nothing.
Seriously, how long would it take? Unless the number of upvotes and the number of downvotes are not stored separately in the first place (which I don’t think it’s the case—aren’t the “Best”, “Popular” and “Controversial” sorting criteria based on them?), I wouldn’t expect that to take more than 0.1 programmer-hours.
Several real modifications to the code base I work on have taken less than 0.05 programmer-hours. About half of which was booting up and connecting to the server.
In this case, it would be a matter of changing something along the lines of this:
I don’t know the exact implementation details server-side on this, as I don’t have access to the LW source code, but it really would be a minor tweak (unless upvotes and downvotes are counted together, in which case it would be nontrivial).
You are talking about code on which you work. That means that you already invested a lot of time into knowing the code base. If SI search someone to implement this they probably need to spend some time to understand the code base before they can make changes.
I don’t know the exact implementation details server-side on this, as I don’t have access to the LW source code
They HAVE implemented a feature that pops up a question “Replies to downvoted comments are discouraged. Pay 5 Karma points to proceed anyway?” and then you choose whether to spend 5 karma points or not to reply to the downvoted comment. It doesn’t seem to me the feature to report positives and negatives separately would be less coding than this one they have put in place. See http://lesswrong.com/lw/fnk/meta_retributive_downvoting_why/7wls?context=3 for a comment you can see the pop-up question at when you try to reply.
It sounds like you’re saying something like: “A friend loaned me five bucks, so surely he can loan me five more.” By induction, the friend has infinite amounts of money.
No, it is something like “A friend went to a movie 12 times in the last 15 weeks for $5 each. So if I find a movie the friend will like and it is $5, he’ll probably want to go.”
This has only been a feature request since the forum started.
This suggestion usually gets a lot of support, is there a reason why it hasn’t been implemented?
Time, resources, and programmer-hours: how much more is SI supposed to spend on incidental forum problems?
Technically, I think this is more Trike’s problem, so it would be more like ‘how much more of Trike’s time is this supposed to take up or be diverted from other LW maintenance/improvements?’
Several months ago, another user offered to set up a fork of the reddit enhancement suite that could achieve this and other features for users interested in them, but the project never took off. Arguably, this is a poor way of solving the problem, because it requires opting in, and most users would continue to see the old look instead. But it would be better, perhaps, than doing nothing.
Seriously, how long would it take? Unless the number of upvotes and the number of downvotes are not stored separately in the first place (which I don’t think it’s the case—aren’t the “Best”, “Popular” and “Controversial” sorting criteria based on them?), I wouldn’t expect that to take more than 0.1 programmer-hours.
I was upvoting for the idea, but then retracted at the suggestion that ANYTHING real takes less than 0.2 programmer-hours.
Several real modifications to the code base I work on have taken less than 0.05 programmer-hours. About half of which was booting up and connecting to the server.
In this case, it would be a matter of changing something along the lines of this:
and tacking something along the lines of this to the end:
I don’t know the exact implementation details server-side on this, as I don’t have access to the LW source code, but it really would be a minor tweak (unless upvotes and downvotes are counted together, in which case it would be nontrivial).
You are talking about code on which you work. That means that you already invested a lot of time into knowing the code base. If SI search someone to implement this they probably need to spend some time to understand the code base before they can make changes.
The source code is open source.
They HAVE implemented a feature that pops up a question “Replies to downvoted comments are discouraged. Pay 5 Karma points to proceed anyway?” and then you choose whether to spend 5 karma points or not to reply to the downvoted comment. It doesn’t seem to me the feature to report positives and negatives separately would be less coding than this one they have put in place. See http://lesswrong.com/lw/fnk/meta_retributive_downvoting_why/7wls?context=3 for a comment you can see the pop-up question at when you try to reply.
I don’t understand the logic here.
It sounds like you’re saying something like: “A friend loaned me five bucks, so surely he can loan me five more.” By induction, the friend has infinite amounts of money.
No, it is something like “A friend went to a movie 12 times in the last 15 weeks for $5 each. So if I find a movie the friend will like and it is $5, he’ll probably want to go.”
There’s even an issue for it.