Hacker News new | past | comments | ask | show | jobs | submit login
Comments Are Dead. We Need You to Help Reinvent Them (pbs.org)
99 points by miraj on May 14, 2011 | hide | past | favorite | 86 comments



Comment sections in exist to drive page views. Online publishers want flame wars to break out, because flame wars mean traffic.

There's no money in quality comments. Reddit is a vast money-pit for Conde Nast. Kevin Rose might have got paid, but Digg is going down the chute. HN serves as a recruiting ground for YC and I believe that Slashdot is treated as a loss-leader to support other Geeknet properties.

Personally, I think we need to take a lesson from the granddaddy of them all, WELL. Mandate real names and charge admission and you'll get quality interaction. Contrary to what internet culture might have us believe, exclusivity and accountability are generally good things.


Reddit is a vast money-pit for Conde Nast.

Really? How much does CN lose annually on Reddit?


Fair enough PG, that was bang out of order.

That said, Reddit does look to be in a bad way. They're experiencing chronic and severe downtime, mainly as a result of technical debt - AWS can only take so much blame when firms like Netflix got through the outage fine. Even when AWS isn't undergoing major outages, "emergency read-only mode" has become rather too common.

When two members of a three-man developer team quit at roughly the same time, that's a warning sign.

Reddit looks to me to be in a similar situation to 4chan - a huge, highly influential user base that is actively hostile to monetisation. It looks to me like the best case scenario is that Nast is deliberately smothering Reddit - all the other explanations for their current problems bode much worse.


Hey, they recently doubled the headcount of their sysadmin team!


Usenet, which had real names, was filled with non-quality interaction.

SA, which charges admission, is self-obsessed and psychotic.

If you have enough posters or your topic invites people who aren't motivated to debate rationally, your comments are going to be stupid.

EDIT: one thing that Usenet did have was the killfile. Now that's solving social problems with technology!


I agree. I think a sense of community is the main driver of quality comments: that's why news sites and YouTube have such poor comments, because most people followed a link to that specific article/video and have no long-term relationship with the site; whereas forums on a site devoted to a single topic do much less badly.

But I don't think real names are strongly causally connected to a sense of community.


For a good example, see advrider.com, which has free and (mostly) pseudonymous accounts but a great sense of community.


Reddit created Reddit Gold specifically because they were so cash starved.

I also question how much revenue they generate from ads considering the number of users who use Ad-Block.


Actually there were rumors that Reddit Gold was about internal Conde Nast politics -- CN takes paid subscriber count very seriously, so they get a lot more respect at CN now that they have lots of paying subscribers.


However, there's quite a few people that disable AdBlock specifically for Reddit.


It's true that publishers use comment sections to drive page views, but that doesn't mean they don't want quality discussions. Essentially, what they want are civilized flame wars. I know, I know ... crazy, right?


Comments aren't dead. The problem that the author of this article has identified lies more in moderation then the comment system.

Comments have evolved. Many high profile sites rely on passive, one way comments that are occasionally peppered with the odd reply, but for the most part act solely as the receptacle of one-way feedback. This isn't necessarily bad and at one time, provided adequate engagement.

Today, readers have become accustom to community and participation rather then simple contribution. We expect more, but we don't receive it.

The solution is a mix of additional functionality, increased participation and eventually clear identification of authority.

Additional Functionality: We've seen basic moderation systems evolve. From Digg style voting buttons to full on wiki style comment systems where users can add sources and refine messages, there have been technological advances that have been largely overlooked in commenting systems.

Increased participation: This is more then simply encouraging users to comment. Users need to be engaged and that happens by the author in the comments. After an article is composed, authors need to become conductors and individually guide conversations and continue to contribute their ideas and messages. Generally, commenters need to have their thoughts validated while the article is thought of as an introduction to a topic while the comments contain the larger message being presented.

Authority: Comment systems such as Disqus and Facebook Comments gather contributions across blogs. All of the comments a commenter has made are available through a common system that supersedes individual articles. This needs to be expanded and improved. Comment systems need to identify authority and encourage "cross seeding" through related people, their individual expertise and the interests they share.

We're getting there. However, we have a long way to go.


Yes, exactly. This morning, in response to a tweet from a developer at ASU's Cronkite School of Journalism's, I wrote a similar blog post trying to explain how journalism sites create one way comment systems and then ignore the comments and then wonder why the communities all go to hell.

At the risk of blog whoring, it's here: http://jerryasher.posterous.com/two-cents-on-why-journalism-...


At the risk of blog whoring ...

I appreciate links to other resources of interest, and often it's the author who knows best of the relevance, so whore away.


I'm a web editor at a regional news site, and moderating reader comments is a huge pain in the neck, largely because of the inherent anonymity of the web.

A few years back, I tried to come up with a solution. I created Truyoo (http://www.truyoo.com), which essentially requires a user to pay a one-time nominal fee (less than $2) to confirm their identity. They would then be issued a Truyoo ID, which could be used to comment on any site that used Truyoo.

For legitimate, constructive commenters, this is extremely inexpensive -- but it quickly becomes very expensive for spammy/abusive users.

Unfortunately, Truyoo never gained traction, largely because no publisher was willing to ask its users to pay even a tiny one-time fee to comment.


It has nothing to do with anonymity. It has everything to do with the content on which people comment. For example, political articles are bound to bring in opposing viewpoints. And if you allow anyone to comment, you're bound to bring in a lot of people.

I'm convinced if you want to keep the level of discourse up, you can have a (mostly) homogeneous group of commenters and a large scale site or a small group of heterogeneous commenters, but you cannot have both scale and heterogeneity.

By charging your users to comment, you're probably going to reduce scale.


It's a great point that there aren't any obvious examples in the US of large heterogenous groups with good comments. That said I also am not aware of any attempt to do even Slashdot-like crowdsourced voting (including explanation of reasons for voting, meta-moderation) at scale, or to take the next step and provide a richer user experience to make it easy for users to be able to filter out what they don't want do see.


it has a lot to do with anonymity. simply look at techcrunch and what happened overnight with real names- quite large and homogenous too.


From a publisher's point of view, it has a lot to do with anonymity. Ever try to block a determined troll?


Removing anonymity discourages normal users who don't want to leave a permanent record that may come back to haunt them later ("The Internet's not written in pencil, Mark, it's written in ink").


Just because anonymity might be useful in some online discussion forums does not mean it's necessarily appropriate or best for all online discussion forums.

Most newspapers require letters to the editor that appear in their print editions to be signed with the person's real name, and they suffer no shortage of letters as a result of that policy. I see no reason why they can't hold online users to the same standards.


I agree. Anonymity is essential for honest discussion. Anonymity is almost required to know what people "really" think. That's why we created Spottiness. In our site, every "spot" has to be anonymous or won't be published. Our challenge is to weed out the garbage while we make it very easy for everybody to say what they think.


So you think 4Chan is the epitome of honesty?


Never heard of that site. I gave it a look and realized that your comment is ironic, but still I'll reply: Our content is heavily moderated, so the obvious crap that 4Chan shows has no possibility of being published. Yet, if somebody has something interesting to say about pornography, we provide the anonymous channel, as long as the idea meets common sense standards of civility and good taste.


Why does your site have user accounts?

It seems like it's combining different audiences at once - the red ones are transmission of random teenage angst (better to put in any pseudonymous journal site or tumblr), the gold ones are positive reviews (seems like they're safe on Yelp), and no black ones have actually appeared in the time I've had this window open.

It's nice that the page live-scrolls, but then why does Read More reload the page?

> Yet, if somebody has something interesting to say about pornography

They might not care whether you think it meets your standard of good taste?


> Why does your site have user accounts?

We will use registered users to help us moderate our content. We also want to work, eventually, on creating a network of trust. By the way, user accounts are optional, not required to create spots or comment about them.

> It seems like it's combining different audiences at once...

The fundamental difference between our service and tumblr, yelp, etc, is that our main content (the spots) must be anonymous. We don't want to know who is the author of a spot. Anonymity is the core of what we do because what matters to us is only what people think.

So far, about 50% of the spots are goldspots, and the other 50% evenly divided between blackspots and lovespots. The fact that goldspots are dominating makes us happy.

> why does Read More reload the page?

Read More takes you the the page of the specific spot.

> They might not care whether you think it meets your standard of good taste?

Common sense is common (pun intended). If a spot doesn't look "publishable" to us at Spottiness, the probability that it will look the same to the majority is high.

Thanks for taking the time to look at our site!


It's probably up there. Would you mind spelling the name correctly?


Great concept - the more available approaches the better. It also works fairly well for Something Awful.

My problem with the site is that my intuitive malware/scam heuristics would get a false positive on the site, if you hadn't vouched for it here on Hacker News. If you changed the design, it might help a little, but you obviously still need to figure out the market for the service.

On the positive side, companies are starting to realize that they aren't handling comment systems very well.


Something Awful is also heavily, heavily moderated and it's a group of fairly like-minded people, especially in the subforums outside GBS. If I remember right, I think the $10 charge was more to recoup costs of running the site. Keeping out crazies is a side effect.


It's not very heavily moderated, but it does have moderators that can be summoned with the report buttons.

One of the first things you learn on SA is to "grow a thicker skin", because there aren't enough moderators with sufficient time to stop users from ganging up on some people and giving them a hard time. And as a result, moderators also allow a harsher tone.

The moderators are vital, but I wouldn't say that they are moderating heavily.

It does show the relevance of communities like SA versus random gatherings of visitors that most comment systems on websites constitute.


I think heavy, but not overly up-front moderation is the key to building a good discussion - this simply means clearing out comments which are superfluous - thereby creating richer, more readable content.


I tried the same with a social networking site (Like Reddit/HN) about a year ago. When I asked users to pay even $1 to try it, no one was interested in doing so.

http://e1ven.com/2010/09/15/lonava-com-retrospective/


Heh ... I tried to do almost exactly what you tried to do, with similar results:

http://coding.pressbin.com/50/10-web-start-ups-all-mine--and...

Check out the blurbs on Truyoo, Utopic, and ioFeed.


Wow, reading through that, and looking at Utropic, it's amazing how similar our projects were. We had many of the same aims, ambitions, problems, and results.

I think your analysis of the situation (No Co-Founder, Not fully invested) was true for both of us, but I also, particularly after reading your ioFeed post.. We're creating technical solutions to problems.

That's not enough. Saying "I can improve on Facebook" doesn't take into account the huge network effect, and it doesn't solve the problem of getting people onto your site.

It's a good reminder that the tech isn't enough by itself.


ended up working extremely well for metafilter, though


What was the incentive for them to pay?


"I'm a web editor at a regional news site, and moderating reader comments is a huge pain in the neck"

Right. You're doing it wrong.

Stop moderating reader comments. Let readers moderate reader comments.

Stop creating one way comment systems. Create ways for readers to converse with reporters.

At hacker news for example, all sorts of entrepreneurs come by and all sorts of startup wannabe employees come through as well.

When was the last time you saw a reporter have any serious interaction with his readers in the comments?


Bear in mind that mainstream, general-interest news sites can't get away with (or don't perceive themselves as being able to get away with) allowing incredibly offensive content to just sit there, waiting for other readers to flag it.

Works for some brands -- not for others.


> For legitimate, constructive commenters, this is extremely inexpensive

actually compared to the status quo, free, it's extremely expensive.


Only if your time has no value.


Require sign-up. Implement karma, with voting. Restrict user commenting to 1/day until karma > X. No downvoting possible until karma > 2X. Put temporary blocks on users who derail threads or attempt to people-please with one-liners, etc. Ban users who attack other users.

Most importantly, make downvoting cost one karma point per downvote.


signups are fine, karma is fine, but killing anonymity is not necessary. i read plenty of sites that allow anonymous commenting and the lynchpin for success here is the rarely-considered feature of user flagging. metafilter does great with this, though they don't allow pure anonymity.

if you have a way for people to flag what they think is bad stuff, you don't have to moderate every comment (a la boingboing and others) and you don't give people a free-for-all. additionally, it allows a user culture for the site to emerge as a collective tone is carved through the kinds of posts that get removed or downvoted.

this is a big thing. i think a lot of sites are jumping on the no-anonymity bandwagon because that's what third-party commenting sites push in order to pump up (i'm speculating) registration numbers. disqus allows (or allowed at the beginning) anonymous commenting, but i haven't seen any site that uses disqus to allow it.

i think this is a market failure. i saw third-party commenting as a way for sites to no longer have to implement their own user registration facilities regardless of whether they allowed anonymous commenting. they just figure hey, we can have our bloggers be able to log in, keep that part simple, and then throw the rabble to disqus for them to deal with. typically i've found this causes the commenting culture to shrink a great deal.

i think there's space for a third-party comment provider to allow arbitrary name/email(will not be shown)-type anonymity, while still implementing flagging, voting/karma and, perhaps most-importantly, IP-banning. it wouldn't be perfect, but i do think it would allow more freedom on both sides: to comment, and to be (more) free of moderation overhead.


Why do you feel that paying for downvotes is important? It's possible we're looking at the same problem differently, but I'd be just as excited about limiting the number of upvotes. I feel it's the ratio that matters, not the sheer numbers.


"Restrict user commenting to 1/day until karma > X. No downvoting possible until karma > 2X"

Which would mean that your comment wouldn't be allowed as your account was created two hours ago ;)


on a per-site basis? coming from google & aggregators, who would bother to sign up?

similar reason I don't contribute on SO... I don't have anything to prove, I might have wanted to share some help, but I don't need to waste my time jumping through their hoops and silly badge games to contribute.


It sounds like their problem is solved by comment voting a la HN, StackOverflow, reddit, etc... But the article rejects these as not good enough or not innovative enough. I must be missing something they're shooting for.


Yeah, Slashdot kind of solved this a dozen years ago. It's an imperfect solution, but it's a 90% solution and I think that's good enough.

Part of it depends on content. If you're a real community, then having people sign up for commenting and rating comments is a good solution. It helps promote good content and the requirement to sign up keeps the community free of a lot of hit and run commenter who might often degrade the experience.

If you're a news site, then most of your content is from hit and runs (hits and runs?). You don't have people hanging around your site all day, waiting to post on the next story. You have Drudgereport linking to you and then a flood of mouth-breathers leaving a quick comment about "libtards" or "republicraps" or a quote from some religious text and so on. In these communities, the discussion will never be lifted up and no voting system will work, because nobody is invested in it. Further, if you require a sign-up, nobody is going to bother.

Then, there are blogs. Sometimes you may want people to have to sign up but in other cases, you might need or want to just let anyone freely comment without any hurdles. Nobody wants to go through a signup process if they're only going to see your blog once or twice in their life (when linked to from HN or elsewhere, for example). It's hard to choose the right path, here.

I think much of the answer comes down to two things;

1) Human moderation - both by the community itself and designated moderators - is very valuable. 2) Not every page on the planet needs to allow comments. Frankly, the local CBS news website doesn't need to accept comments on stories. Neither does the BBC's articles. Neither do most.

Of course, they have dismissed everything in this article, so the only thing left is for the government to require every internet user to use their real identity online and then have a single discussion/forum/conversation system that all runs on the same API (or have a standard API that ties together very different discussion/forum/conversation systems). Then, you will have no choice but to post as your real world identity and every comment you ever leave anywhere on the internet will be part of that identity and you will be rated on that identity, always. A truly unpleasant idea and I would rather put up with having to ignore the dregs of the internet versus losing the option of anonymity. Of course, people who live their lives in public often have a different view than this, since their personal desire is to get their own identity out there. Not everyone has that same goal or desire.

(I won't even bring up my distinct hatred of things like Disqus and Facebook whatchamacallit that sites implement instead of drawing up their own conversation tools).


pstack, I agree with you, but then I wonder how news sites like the Economist manage to make a voting system work even for extremely contentious articles. Any article on the Economist that mentions China or discusses bailing out a small European nation tends to explode into a frenzy of commenting, but the voting system over there holds up extremely well under pressure.

Regardless of the contention, the community at Economist.com maintains a relatively civil atmosphere. I have always been perplexed by this - especially because you can only upvote, no downvoting is in (yet).

The funny thing is that back in the day (when I first started commenting over there) they did not have any type of voting and it could get bad sometimes. Users begged for voting and we got it, now things are pretty nifty for all commenters.

Maybe it is because a different group of users frequents the Economist. In which case, the answer to improving your comments is to try to improve your sites culture?


My guess is that this comes down to what the quality of the majority of people hanging around in any forum is.

As long as there are more "rational" and civil people in any forum it can survive on this alone.


I think the Economist might be one of the exceptions, where they are prone to less hit and run activity and have a more participatory readership of regulars. Compare that with any news or magazine site that you might click on right this very moment that is linked via Drudgereport, which I think is far more common.


> if you require a sign-up, nobody is going to bother.

Nobody is going to bother to hit-and-run, but people who read the same site over and over again and find themselves wanting to respond regularly will bother. I think that is a very good thing. If people are using persistent usernames, you also increase user-to-user connections.

Combine that with even rudimentary voting and downvote-based visibility and you've got, if not a 90% solution, at least a 75% solution.


People will absolutely bother to hit and run. Requiring registration makes moderation easier and keeps spam out a bit better.

Also, with voting systems, you run the risk of people using them as a bludgeon against one another. Sites like HN and others mostly work because the group of people commenting are very similar to one another. There's not a whole lot of reason to troll someone you mostly agree with.


One of the reasons why I quit Reddit was because so many of the most upvoted comments were ones trying hard to be witty OR memes. Same goes with some of the things that are upvoted on the subreddits.



There is an incredible amount of hand waving on that page but nothing really solid for "evolving" comments.

It seems like a simple comment-voting system will answer the majority of the concerns.


Want quality comments? Then you've got to frame your content accordingly. Make it open-ended, and make it pose an interesting question (literally or not literally). Controversy helps, but if you're too mercenary in courting controversy, people will quickly tire of your shtick.

Think of content-comment flow like a conversation. If I'm interviewing someone, and all I'm doing is making declarative statements to him, how likely is he to respond deeply and meaningfully? On the other hand, if I ask him interesting questions, I'll get interesting replies. Also, much like in an actual interview, I'll need to ask follow-up questions. When I look at most content sites out there these days, I see little in the way of back-and-forth exchange between the author and the commenters -- except in the form of moderation. Sooner or later, smart content authors are going to realize that their involvement does not end at the posting of the original piece.

Of course, you need to inform just as much as you open up the floor. And therein lies the tricky balancing act.


This doesn't account for the inverse case, where content can not be framed in a way that doesn't attract lamers. There are people who will take a crap in the middle of any thread having to do with Obama, for instance.


I'd argue that, to some extent, "lamer" is a relativistic term. What is a lamer? Someone who posts content that a preponderance of users in a community think is stupid, non-useful, or distracting. But said content is stupid, non-useful, or distracting with respect to the preferences of the majority of users in the community.

Using this logic, you could argue that a community with a lot of lamers is a community that hasn't properly attracted a target userbase, or has failed to maintain one. In this model, lamers are symptoms and not root causes. The cause is, as you suggest, a flawed content strategy.


Of course it's relative, that's what separates communities from "everybody." User flagging is how the community can evolve.

you could argue that a community with a lot of lamers is a community that hasn't properly attracted a target userbase

A person who sees most of a community as lamers is not a member of the "target userbase," and has no business speculating what the "target userbase" of that community should be or even whether they're doing it "properly" (h/t C&C). "If you look around and all you see are assholes, the asshole is you." Nothing wrong with this, that's why people form communities and subcultures. Not everything has to be for everyone.

It seems you're trying to work with a universal definition of "lamer," which doesn't exist. It's always in context. I'm a lamer when it comes to baseball and I'm not going to go into a bar and force people to respond to my endless questions about why bats aren't all the same size and "what is a home run, anyway?"


It seems you're trying to work with a universal definition of "lamer,"

Just the opposite, actually. Your point about being a lamer in context of something you don't understand or like was precisely my point. My point was that a community full of "lamers" is a community that has somehow attracted posters or commenters whose tastes, opinions, knowledge level, interest level, and/or posting intent do not mesh with those of the majority of the userbase. To the extent that these lamers -- as defined by their context -- exist, that's usually a function of a community's having lost focus (or never had proper focus to begin with). Because lamers are defined by context, the only way to eliminate lamer presence is to sharpen content or context focus around the set of users you want to keep.

I think we're agreeing more than we're disagreeing here. And if it doesn't sound that way, I'll concede that my post may not have been worded clearly enough.


Comments dead? What a load of crap.

Comments don't scale very well for popular sites, but a strong moderator presence is vital to creating a good commenting community.

The problem with this for sites like PBS is that the moderation alienates some users and feeds the confirmation bias of some people.

You have to decide for yourself what level of discussion you want and invest a proportional effort. The problem is that most people want good comment discussions without any effort, and that's not going to happen.


Two months ago I got a form letter rejection from the semi-final round of the Knight News Challenge on my proposal to do exactly this. It was based on what I'd learned about news forums from working at the Washington Post, machine learning, and a bit of secret sauce. It seemed like an acute problem worth solving and so far the responses on their site are about 80% ideas I included. Validation, I guess, but I'm sort of a mixed bag of emotions about seeing this.


I'd like to hear more about your proposal. I, too, submitted a proposal dealing with these issues, and my submission was passed over.


How would both proposals allow for infinite scale though?

It seems like the machines can eventually learn to emulate human moderation, but a website like PBS may get 500+ comments. The user would immediately get overwhelmed at 50+ comments or more (guess). For example, on Reddit, people are learning to avoid posting if a thread has more than 200 posts because their highest-level post would be useless.


They don't have to scale infinitely, only a handful of news sites serve more than 20m pages/day.

As for not overwhelming users or moderators, that was part of my secret sauce. :)


I'm actually really against comments, which is ironic, given what I'm doing right now. Most comments either reaffirm the reader's view or make the reader a tiny bit angry. I don't think they're that effective at actually causing people to think.


Think about how many great discussions there are here, or in the best threads on Slashdot, Reddit, and plenty of other places. So it seems to me there are plenty of great comments that help people think, learn, grow, etc. True, there are plenty of comments that aren't so great ... so one way to look at "reinventing comments" is to how to magnify what's good and minimize the "get you angry" factor and other downsides.


I like the idea of having a single ID for Internet. It reduces the burden of creating accounts everywhere. The ID could be used in any site that would like a non anonymous comment or activity. A request for ID info could pop-up for an accept button. There are solutions in the market, of course, but they have not yet got critical mass, I guess. The Id together with karma and voting would help a lot to select interesting comments.

A lot of our browsing activity is not anonymous anymore, I can see the ads are tracking me! So why not use this 'system' for something else?


OpenID, Facebook Connect and Twitter Login. There's some more if you log out and look at the log in options on this website.


Personally, I'd to see better editorial tools in comments. Stack-overflow / exchange achieves this with tagging and a basic rep. system. I'd like to extend this to comment forums.


It would be nice if comments from e.g. Hacker News were overlaid on the page you were visiting. A browser plugin could show you comments from multiple communities. That way you discover the comments even if you find the link another way, and it would be easier to open a bunch of tabs without remembering to open the comments too, etc. People wouldn't need to add comments to pages themselves.


I was thinking about a system like this when reading the article. Does something like this already exist?


Comments on news articles are dead, perhaps, but were they ever really alive? Comments on smaller blogs seem to be doing just fine.


Personally, I prefer to discuss content on another site. Not because their commenting systems are better, but because I have a wide range of interests, and it is more convenient for me to discuss information where any content can come in.

Also, on a site dedicated to discussion, I don't have to worry about moderation or permission of the content creator.


Exactly. There is no point whatsoever in commenting on a multitude of blogs and website across the internet - then your opinions, views and contributions are fragmented in different places across the internet too; with no way to keep track of them, and no way to build rapport.


I like Facebook's approach to comments (https://developers.facebook.com/docs/reference/plugins/comme...).

Pros:

- Most people are already signed-on in Fb, no friction to start commenting.

- There is a hierarchy of posts, similar to HN, reddit, etc

- Not anonymous.

Cons:

- Who owns the data? Probably Facebook.


Why is "not anonymous" a pro? Usually, you can't say what you truly think if you know that your identity is revealed, particularly if what you think is negative in nature. In Facebook people wear the same masks that they wear in real life.


Here's an interesting alternative perspective on Facebook comments, coming from one of its most prominent users, TechCrunch: http://techcrunch.com/2011/04/03/facebook-comments-epitomize...

The flipside to this is that there have been a number of articles from TC praising the new comment system, so it's not entirely negative; but I found this article to be really interesting nonetheless.


They Do let you suck down the comments it seems.


Is upvoting generally a good idea in a community? What kind of commenting behaviour does it lead to?


The way PG controls comments here is with the up-vote down-vote ranking. It works in separating the article discussions from the opinion oriented posts. Without the ranking the opinions mix with the article oriented posts, creates a sloppy discussion board. Another approach is not to bump opinion posts downward but horizontal into a new list of posts where you expect the posts to be bumpy at best and offtrack of the original topic. On the opinion list, confrontational posts are upvoted, 5 strikes and your out for a month.


Peer weighting.


Twitter seems like a fairly good alternative?


Twitter appears to be the problem - further fragmenting discussion across the internet, not bringing it together.


What about twitter search? They have a widget made just for that: http://twitter.com/about/resources/widgets/widget_search


one can take a multiparametric approach to traditional voting, like here: http://textchannels.com/page/about . Not much of progress compared to traditional voting, but it might help tailoring the comments to the subject matter.




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: