I also have to come up with alternative ways for people to find the comments they're most interested in on big threads
I think this is the key. People come to HN because the quality of the discussion is high. With increasing community size comes dilution, but only of the average comment quality.
In fact, as the community becomes larger there are more, perhaps many more extraordinarily good comments, so long as the original contributors are all encouraged to continue participating. These great comments are really the key.
Consider the Netflix Prize. Entries were judged based on RMSE. This is not the right metric, though, because out of all of the thousands of movies in the dataset, only a few dozen would ever be recommended to a given user based on any conceivable algorithm. And the user only really cares whether the particular movie that they use the algorithm to select is good or disappointing.
Similarly, users come here to read the exceptional comments, and a community in which every comment deserved to get 3 karma points would soon find itself without users.
In this way, removing comment scores may be part of an effective solution, but it may not be the crucial component because it only indirectly changes the accessibility of excellent comments.
Instead, actively trying to make it easier for people to read what they consider exceptional is a goal worth pursuing.
I think this is the key. People come to HN because the quality of the discussion is high. With increasing community size comes dilution, but only of the average comment quality.
In fact, as the community becomes larger there are more, perhaps many more extraordinarily good comments, so long as the original contributors are all encouraged to continue participating. These great comments are really the key.
Consider the Netflix Prize. Entries were judged based on RMSE. This is not the right metric, though, because out of all of the thousands of movies in the dataset, only a few dozen would ever be recommended to a given user based on any conceivable algorithm. And the user only really cares whether the particular movie that they use the algorithm to select is good or disappointing.
Similarly, users come here to read the exceptional comments, and a community in which every comment deserved to get 3 karma points would soon find itself without users.
In this way, removing comment scores may be part of an effective solution, but it may not be the crucial component because it only indirectly changes the accessibility of excellent comments.
Instead, actively trying to make it easier for people to read what they consider exceptional is a goal worth pursuing.