Reddit, Hacker News, and similar sites work on the “title goes to source, comments go to comments” model. I suspect it will be more expectation-violating overall to have different behavior here. (I agree that there is an expectation shift in going from LW as a place with only self-posts to a place with linkposts.)
For the record, Hacker News and Reddit also annoy me every time this happens.
I do think it was a fine design choice given that it does seem to work for a lot of people.
But I’d personally rather have a convention where you click the link, and then see a typical discussion page with short summary / conversation prompt, followed by comments all in one place.
Maybe we can build a user setting for this (excluding the summary)? Or, actually, if we’re already building a system to allow users to edit tags (a la Stack Overflow), maybe it wouldn’t be terrible to let users edit the summary (a la wiki).
It’s awesome that you guys are really considering ways to incorporate changes people want.
I wonder, since you’re going to have to put a lot of work into the refurbishing project and resources are finite, would it be worth generating some kind of survey for members to take about what kind of features/alterations/options they’d most like to see? I ask because it occurs to me that soliciting ideas in open threads, while absolutely useful as far as encouraging discussion and exchange of ideas goes, might present a patchy or unduly-slanted picture of what the majority of members want. Prolific commenters (like me!) might dominate the discussion, or certain ideas might look more important because they generate a lot of discussion. A survey of some sort might give you clearer data.
That’s not to say you should necessarily do things because the majority want them, this isn’t a democracy as far as I’m aware and some popular requests might be unworkable. It just could be useful to know. Of course you’re better placed to determine if it’s worth the effort.
(Also, this isn’t in any way prompted by Raemon’s point about the link posts—it was your reply about possible implementation options that put it my head).
Reddit, Hacker News, and similar sites work on the “title goes to source, comments go to comments” model. I suspect it will be more expectation-violating overall to have different behavior here. (I agree that there is an expectation shift in going from LW as a place with only self-posts to a place with linkposts.)
For the record, Hacker News and Reddit also annoy me every time this happens.
I do think it was a fine design choice given that it does seem to work for a lot of people.
But I’d personally rather have a convention where you click the link, and then see a typical discussion page with short summary / conversation prompt, followed by comments all in one place.
Maybe we can build a user setting for this (excluding the summary)? Or, actually, if we’re already building a system to allow users to edit tags (a la Stack Overflow), maybe it wouldn’t be terrible to let users edit the summary (a la wiki).
It’s awesome that you guys are really considering ways to incorporate changes people want.
I wonder, since you’re going to have to put a lot of work into the refurbishing project and resources are finite, would it be worth generating some kind of survey for members to take about what kind of features/alterations/options they’d most like to see? I ask because it occurs to me that soliciting ideas in open threads, while absolutely useful as far as encouraging discussion and exchange of ideas goes, might present a patchy or unduly-slanted picture of what the majority of members want. Prolific commenters (like me!) might dominate the discussion, or certain ideas might look more important because they generate a lot of discussion. A survey of some sort might give you clearer data.
That’s not to say you should necessarily do things because the majority want them, this isn’t a democracy as far as I’m aware and some popular requests might be unworkable. It just could be useful to know. Of course you’re better placed to determine if it’s worth the effort.
(Also, this isn’t in any way prompted by Raemon’s point about the link posts—it was your reply about possible implementation options that put it my head).