A comment feed that starts from the top-rated comments since my last visit would be much more useful for me than the chronological feed, and would probably increase the number of discussions I become involved in.
A good system for finding old content. Often old posts or comment threads represent the best content on a given topic, but if there isn't a way to browse by topic they're basically lost. This makes me less excited about engaging in the first place -- vice versa a sense that something will persist is a draw in a venue for circulating ideas.
This is one thing where I think the forums/subforums structure of old UBB boards is actually superior to most more modern replacements (although such boards have other significant problems, most notably lack of threading).
Interesting. I agree with this, but am not super sure about the best solution. I was hoping we could fix this with something like the sequences feature, which would be optimized for creating persistent content, and a sense of canon in the community.
I.e. there would be a set of sequences that would be considered core, and that would be prominently placed on the page. Users would have the ability to add sequences, and if they are good enough and popular enough, they get added to the core sequences
This seems to make past content more accessible, though it is better suited for making top-level posts findable, and worse suited for making very good discussions findable. Unsure how to best make the discussions more findable.
There seem to be two related problems here: content discovery for old posts without a specific topic; and finding posts on specific topics.
For the former, Less Wrong currently has two mechanisms: the Sequences wiki articles, and http://lesswrong.com/top/ . Neither of these is entirely adequate. The Sequences wiki articles fail to mention some of the best content. /top/ is close, but it's a feature that you can only use once; if you go back a second time, it works badly because you have to click Next a bunch of times before you get to something you haven't already read. This seems fixable! It might be good if LW tracked which articles you've read, in a way that's more reliable/persistent than browser history, and provided a feed of unread posts with a mix of new posts and highly-upvoted old posts.
It seems to me like tags are potentially a good solution here, so long as citizen-editors have the power to create tags and keep them orderly.
I don't think that using up prominent real estate on core sequences is a good idea. I think that /top/ on lesswrong is a better idea, and agree with jim that mixing in top/unread with recent stuff would be pretty reasonable (e.g. if 1/3 of stuff I saw was just the highest rated article I hadn't read or skipped too many times).
Putting everything into a giant book is not a reasonable form of organizing institutional knowledge. (At least not beyond a certain point.)
In general, I feel like the sequences approach strangled the intellectual growth of LW 1.0 by sending people the message that unless their content is in the top 0.1% of content they should just go home. So that's what people did.
If you want a thriving discussion you're going to need to have a space for discussing things with reasonable but not incredible quality standards. You do not want to give users the message that for their discussion to be worthwhile it must ultimately be worthy of inclusion into a canon. Trying to build a canon too early just strangles everything. You're overemphasizing it because it's what EY did when you really should if anything underemphasize it.
p.s. Overemphasizing the construction of a canon also has negative PR effects, because it gives the impression of a fanatic group.
I don't foresee posting to LW in place of posting on my blog without some unanticipated value add. I would be happy to use a shared comment system or something like that.
If interesting discussions are occurring I would be reasonably likely to chime in from time to time, as I currently do on LessWrong but with a frequency determined by the frequency of interesting conversations.
What do you feel is the biggest benefit of using your blog instead of LessWrong? I definitely have a sense of why many authors in general prefer writing on their own blogs, but I would be interested in a short summary of what causes that preference in your specific case.
People browsing my blog seems much more likely than people browsing my user page. More generally, stuff I post on my blog seems to reflect on me in a way that stuff I post on LessWrong does not (instead reflecting on LessWrong).
A big motivator for me, I think, is the quality (and quantity) of comments on my own stuff. No feedback feels like negative feedback when I post to LW, whereas no comments is what I expect when I post to my personal blog and feels fine. And obviously, low-quality feedback can be demotivating.
Similar here. Two things that discourage me from posting on LW:
- knowing that if the debate gets interesting and will have even smallest political connotations, Eugine will likely harass the participants (well, before the downvotes were disabled, but disabling them brought new problems); furthermore, the potential participants are already aware of this, which makes them less likely to participate;
- there is a ton of stuff each week in Discussion, some of that low quality or just a link; regardless how much effort I put into my article, in two or three days it will be scrolled down into oblivion anyway; this was traditionally solved by Main vs Discussion, but the articles in Main paradoxically had less visibility, and the threshold for both of them was constantly lowering anyway, Main becoming the old Discussion, and Discussion becoming the old Open Thread comments.
So I'd like to see the discussion kept nice; and I'd like to see the good articles have a longer timespan.
(Not insisting on any specific technical solution, but for example the problem with "actually less visible Main" could be solved by simply posting links to 5 latest Main posts at the top of the Discussion page. There, everyone would see them, even if they only read Discussion.)
(Similarly, the "Discussion becoming new Open Thread" could be solved by having three levels, corresponding to previous Main, Discussion, Open Thread, but perhaps using different names, e.g. "Promoted", "Articles", "Forum". To avoid having to estimate the category of your own article, all new articles would go to "Forum", and then moderators would move them into higher levels using their own judgement, with karma as a guideline.)
So a few things I've been thinking about to solve these problems:
I hope we can deal with Eugene after we have a better Karma system and better moderation tools. Though we will see how hard this problem turns out to be. But my guess is that even if it turns out to be harder, it is a problem that can be solved with sufficient, but not prohibitive, engineering effort.
This current page already has a more aggressive scoring system that keeps highly upvoted posts at the top for longer, and which applies an exponential decay function over time, which results in you getting a mixture of recent posts and top posts as you scroll down the frontpage, instead of mostly a historical list of posts. I hope this increases visibility. I was also thinking of maybe adding a "promoted" section to the top of the posts list that always shows the top 5 promoted articles (by time-decayed score), so that you have something similar to main, without it being a whole click away.
I am currently hesitant to create more top-level distinctions like main/discussion/forum, both because I think it decentralizes discussion, adds mental overhead and encourages styles in different parts of the page, which makes it harder to promote the best content from anywhere on the page (i.e. if the forum is mostly written in a style appropriate to forum, then it's harder to promote that content to the frontpage with all the long-form content)