The main problem I have with voting to reach the proper score is that to do it well means either that
I’d have to constantly check all the posts I have voted on to see whether the score hasn’t overshot and my vote isn’t already moving the karma away from my desired value, or
I’d have to guess the future quasi-stable rating of the post and vote according to the difference of this guess from my desired score (assuming such a rating exists, i.e. that karma of posts or comments stops wildly fluctuating some time after they are published).
(1) is obviously waste of time and (2) is too complex and error-prone. The default position of many LWers, i.e. voting according to the present standing and not checking afterwards, I dislike because the resulting score depends on the order in which the readers vote. Consider e.g. 10 readers finding the post unremarkable and deserving 0 and another 10 readers finding it worth +5. Then if the former group votes first and the latter votes last, the post ends up at +5, while in the reversed order it would stand at 0.
(I have observed that many bad posts sink rapidly soon after being posted and then slowly climb back to around zero level. I hypothesise that may be because the regular readers who may be more demanding in quality are more likely to read a post soon while the more forgiving occasional readers who check new content less often read posts later; even if there were many more regular readers than occasional readers, since most of the former group would forgo voting because the post had already got what it deserved, the latter group would dominate in the long-term voting results.)
I should note this is well-travelled ground; we have this conversation about twice a year.
You are, of course, correct about the failure mode of the adjust-towards-preferred-score strategy.
The vote-without-reference-to-preferred-score strategy has the problem where, if comment C1 has a score 5x higher than C2, someone reading them can’t tell whether that reflects 5x as many people reading C1 as C2 and endorsing it about as much, or a 5x higher proportion of readers endorsed it. Which is kind of a big deal, if we want karma scores to serve as feedback about comment quality/desirability rather than thread popularity, which some of us do.
People vary in terms of which problem they’re more motivated to avoid, so you end up with both behaviors in the community. When the subject comes up, proponents of adjust-toward-preferred-score typically dismiss the concern you describe by positing that there’s no systematic difference between early voters and late voters, and proponents of ignore-preferred-score dismiss the above concern by positing that there’s not that much variation in how many people read any given pair of comments. (Personally, I suspect both of those statements are false.)
The main problem I have with voting to reach the proper score is that to do it well means either that
I’d have to constantly check all the posts I have voted on to see whether the score hasn’t overshot and my vote isn’t already moving the karma away from my desired value, or
I’d have to guess the future quasi-stable rating of the post and vote according to the difference of this guess from my desired score (assuming such a rating exists, i.e. that karma of posts or comments stops wildly fluctuating some time after they are published).
(1) is obviously waste of time and (2) is too complex and error-prone. The default position of many LWers, i.e. voting according to the present standing and not checking afterwards, I dislike because the resulting score depends on the order in which the readers vote. Consider e.g. 10 readers finding the post unremarkable and deserving 0 and another 10 readers finding it worth +5. Then if the former group votes first and the latter votes last, the post ends up at +5, while in the reversed order it would stand at 0.
(I have observed that many bad posts sink rapidly soon after being posted and then slowly climb back to around zero level. I hypothesise that may be because the regular readers who may be more demanding in quality are more likely to read a post soon while the more forgiving occasional readers who check new content less often read posts later; even if there were many more regular readers than occasional readers, since most of the former group would forgo voting because the post had already got what it deserved, the latter group would dominate in the long-term voting results.)
I should note this is well-travelled ground; we have this conversation about twice a year.
You are, of course, correct about the failure mode of the adjust-towards-preferred-score strategy.
The vote-without-reference-to-preferred-score strategy has the problem where, if comment C1 has a score 5x higher than C2, someone reading them can’t tell whether that reflects 5x as many people reading C1 as C2 and endorsing it about as much, or a 5x higher proportion of readers endorsed it. Which is kind of a big deal, if we want karma scores to serve as feedback about comment quality/desirability rather than thread popularity, which some of us do.
People vary in terms of which problem they’re more motivated to avoid, so you end up with both behaviors in the community. When the subject comes up, proponents of adjust-toward-preferred-score typically dismiss the concern you describe by positing that there’s no systematic difference between early voters and late voters, and proponents of ignore-preferred-score dismiss the above concern by positing that there’s not that much variation in how many people read any given pair of comments. (Personally, I suspect both of those statements are false.)