News for hackers. Stuff that matters.
Re:Why Online Communities Decay Over Time by jere on Friday May 23, 2014 @04:52PM (#7791270)
Daydreaming here... so how about a community that funnels users into buckets automatically? Think of an HN where you only discuss an article with a pool of ~500 people instead of 80,000. When new users join, they're distributed to the smallest groups. Nobody is overwhelmed with an influx of noobs. And you might actually be able to start remembering the names of the active people in your group. Perhaps you could see top comments from other groups, but not interact with them directly.
Re:Why Online Communities Decay Over Time by jacquesm on Friday May 23, 2014 @04:54PM (#7791278)
I've had some pretty extensive conversations around that theme with a bunch of other HN'ers (some of who have moved on in the meantime), and that's pretty much the solution we hit on. Keep only the +x and -x active users in view for you at any one time.

One problem we foresaw with that approach is that it becomes hard not to have duplicate discussions or missing chunks of information. In the end we did not do anything with the idea but it was certainly interesting.

Re:Why Online Communities Decay Over Time by RKoutnik on Friday May 23, 2014 @05:00PM (#7791309)
That's a rather interesting concept. My big issue is that I'd be missing the best content from other buckets. Perhaps content can 'emerge' from a bucket after garnering a certain % of support from the bucket?
Re:Why Online Communities Decay Over Time by NaNaN on Friday May 23, 2014 @05:09PM (#7791357)
To avoid the Matthew effect, the best users should be re-distributed, too. -- still nice?
Re:Why Online Communities Decay Over Time by RKoutnik on Friday May 23, 2014 @06:00PM (#7791574)
If that was the case, I'd worry about the buckets becoming like German universities - the best minds can't get together, so we all lose out.
Re:Why Online Communities Decay Over Time by NaNaN on Saturday May 24, 2014 @08:48AM (#7793217)
Maybe, if there are not enough geniuses. Now I'm thinking about the relationship between birth of lives and birth of communities, and why we must rush into a crowded place.

In the past, people like to read newspaper in the streets. Now, we read news from HN. If there is no comment box any more, no one can try to be a superstar in such a prominent place. Then we can just upvote and downvote a news, then talk about it somewhere else. And HN can let the submitter recevie links about discussion from others and show them under the post. Or just receive any links, then the submitter can place it at another positon.

Re:Why Online Communities Decay Over Time by gwern on Friday May 23, 2014 @06:20PM (#7791671)
I don't think most HN users know anything at all about German universities, much less any bucketing they might do, so that's probably not the most helpful analogy.
Re:Why Online Communities Decay Over Time by RKoutnik on Friday May 23, 2014 @06:33PM (#7791737)
I learned about that particular situation from one of pg's essays (something I figured I could assume was common knowledge around here).
Re:Why Online Communities Decay Over Time by gwern on Friday May 23, 2014 @10:17PM (#7792380)
I've read most/all of pg's essays as well, but I have no recollection of anything like that, so...
Re:Why Online Communities Decay Over Time by RKoutnik on Saturday May 24, 2014 @03:07PM (#7794073)
> German and Dutch governments, perhaps from fear of elitism, try to ensure that all universities are roughly equal in quality. The downside is that none are especially good.

http://paulgraham.com/america.html

Re:Why Online Communities Decay Over Time by gwern on Saturday May 24, 2014 @03:30PM (#7794138)
Only 2 sentences? No wonder I don't remember it at all.
Re:Why Online Communities Decay Over Time by lifeformed on Saturday May 24, 2014 @04:01AM (#7792889)
Can someone explain the German university thing to me? It sounds interesting.
Re:Why Online Communities Decay Over Time by NaNaN on Saturday May 24, 2014 @08:36AM (#7793199)
Well, I searched on HN and googled it, but nothing found easily. So this is why I don't like slang sometimes.. :-$
Re:Why Online Communities Decay Over Time by RKoutnik on Saturday May 24, 2014 @03:07PM (#7794074)
See my reply to gwern above.
Re:Why Online Communities Decay Over Time by api on Friday May 23, 2014 @07:39PM (#7791968)
That's a very interesting concept!

What if it were geo-local? When you post an article or text post everyone sees it but you only see comments from people within 100km of your location.

Re:Why Online Communities Decay Over Time by jere on Friday May 23, 2014 @08:25PM (#7792129)
I think there are already several apps in that space, but maybe I'm thinking of things that are more localized.

You definitely can't just do everyone within a given radius from you because a person on one side of you won't be able to communicate with someone on the other and that's incredibly confusing. You could split the buckets up by location, but some people would have to put up with living on the border (well, just like a country or state).

Re:Why Online Communities Decay Over Time by vostrocity on Saturday May 24, 2014 @04:55AM (#7792964)
Slightly tangential, but this is a problem that is being worked on in MOOCs. It is ineffective to have a forum for all 10,000 or 100,000 participants in a class. Issues may be surfaced there but it is a terrible venue for authentic discussions and building relationships with classmates.

There is one startup spun out of Stanford that is doing MOOCs by emphasizing the small group. At the beginning of the course, you can either create your own group and recruit others or join a group (generally 4-8 people). Those who do neither after the first week or so of class are randomly assigned into groups with others who were also not proactive about joining a group. It is easier to do discussions and group projects when there is a permanent group of 4-6 other students, and some of the courses have components where the group will submit work together and the rest of the class can see it.

Re:Why Online Communities Decay Over Time by kevb on Saturday May 24, 2014 @03:39PM (#7794162)
Perhaps something that groups (or even just a weighted score) people and comments by voting habits. If I consistently vote up memes and receive votes from other meme voters, then show me more of that kind of content and give my posts more visibility to that crowd. But if I regularly downvote memes then sink those posts from me, and sink my posts from them.

I imagine tuning such a system would be very difficult though, and you don't want users stuck in a bucket that they don't feel fits them.