Ask HN: We are looking to rethink discussion platforms. Your suggestions?

20 points by rrpadhy ↗ HN
We are working to re-think discussion platform.<p>What are the improvements you would like in Hacker News or Reddit, etc? What is your fav discussion platform? Why?

46 comments

[ 3.2 ms ] story [ 95.4 ms ] thread
You should take a look at discourse: https://github.com/discourse/discourse.

They have rethought it and implemented it quite well.

Yes, we like discourse to a great extent ...

however, one problem which we always face is following thru the comments on HN or reddit .. once the comments cross 100 ... or even 50

Discourse does not solve that either ... Is that a problem for others?

Personally, i Hate reddit's UI. I am not sure why, but i just do. Comparatively, 4chan is a breeze. Sure, it doesn't have the tree structure, but it doesn't look DAUNTING. It's relatively simple. And it does allow you to follow a chain of posts (forward and backward mentions). In fact, if you think about it, 4chan actually supports GRAPH structure for comments instead of the traditional tree, as the same comment maybe replying to multiple posts.

I may be contradicting myself here, but somehow i do like HN's UI too. It has a tree, but pretty neat. Though the levels can be a little ambiguous at times.

So what solutions are you guys thinking of? Or is this just one of those "Hey! let's make a discussion platform! I wonder what's wrong with the current ones... let's ask!"

Have not really liked 4chan, but I will give it a re-look now.

We are having a few things in mind. One of them being applying categories or tags to comments, along with the nested comments.

We believe, each discussion essentially has multiple parallel threads running, and each thread should have its own structure.

The other thought is about users able to post as anonymous, but with certain restrictions.

For the anonymity thing, look at my other comment[1]

Tags/categories may seem like a fair idea, but what i don't like in UI's is information overload. So keeping info on a discussion would be pretty cool, but make sure i don't see it if i don't WANT to see it. it should be available, but not "in-my-face".

Also, what do you think people will react to having to type tags or subjects for their posts?

You know what would be super amazing? A discussion visualization/organizing tool. That could perform an analysis on the comments and perhaps group them together by relevance/chaining/sentiment. Would be cool if i open a 200+ comment page on HN or anywhere an immediately see a visual representation of what's going on. What aspects are being discussed and in what sentiment. Like someone posts a Product Landing page. I open the comments and see a lot of discussion about "UI" or "security" or "usability" or "heath risk" or "privacy" , each marked with +ve or -ve sentiments. Now THAT is something worth working on.

EDIT: the tool could do something as simple as marking keywords and giving posts a red or green BG with varying shades based on the sentiment...

[1] https://news.ycombinator.com/item?id=6933982

Point taken.

We are already having thoughts on the automation of tagging, grouping or categorization of comments. However, before we start working on it, we are validating on how important or necessary this feature would be.

What if, we spend a few days making an awesome automation to do what you mentioned, but no one really gives a st about it. :)

An idea:

A dialectic platform - when two users disagree someone should write a third post that both agree with.

That is an interesting point.

It remains to be seen if two users are actually agreeing to the third person, or still continue to argue with their points

This seems like a good usecase for that "replies to multiple posts" feature i was telling you about in my other comment (would't it be nice if a mention of my other post was a click away? Unfortunately, the process for that is comparatively cumbersome in HN and i am not gonna post the link just to annoy you and make you realize the feature's importance)

4chan maybe a ton of bad things, but what it truly nails, is a simple-to-use and effective UI.

(Legend has it that whenever you speak of 4chan, moot makes an appearance. Wonder if it'll happen this time)

Helpful points here.. - Replies to multiple posts - Easy linking to last comment/post

As about moot, lets see if that happens?

What do you feel about anonymous user handle? Is it a must? Or discussion platform should not have anonymous users, as it leads to trolling.
I like how HN handles that problem. Users are associated with an "account" so their posts etc on THIS site are linked together. And it helps identify trolls and throwaway accounts easy.

But it also gives you a sense of optional anonymity as your HN account doesn't have to necessarily link with the rest of your web presence. No one can find you in the real world just from you HN account if you don't want them too. And still, the option to provide further information gives you the OPTION to present your credentials and connect "who you are" with the HN account.

I find it really interesting how sometimes VERY simple features are solving relatively complex problems

Yes. HN or forums do have this advantage and it has solved the issue of troll to a large extent.

Quora on the other hand, has this feature of being able to ask or answer questions as anonymous, and this has resulted in an avalanche of stupid questions and unrelated answers.

We are somewhere in between, while we like to give a real identity like Quora has, we also believe, anonymity is an important feature to have, for many different reasons.

I think Quora handles this better than HN. It gives you the option to be completely anonymous on a case by case basis, reap the benefits/upvotes even when you are anonymous. Most people on HN/Reddit resort to throwaway accounts when they want to be completely anonymous and not tarnish their username, even though that username has no connection to a real world ID.

People/usernames are brands in online discussion, so the way we generally weigh people's opinion is affected by those brands, which is kind of a function of their comment history and any IRL connections. Say, if the username is 'pg', even if it's a hit-and-run comment with not much thought put into it, you view it differently.

As said earlier, identities is quite important to many users. Quora has done it quite well.

However, with the anonymous feature to ask question, I have seen a rapid degradation of the quality of questions coming on Quora. I am not the only one who has felt this way. We can find a lot of people complaining about the quality of questions of late.

I for one don't take part in discussions where real names are required, as the requirement discourages people with (non-trolling) strongly opposing views from taking part in the discussion, so it's pretty much guaranteed to be an echo chamber (or at least biased towards reaffirming the consensus) - that is, it is worthless for me to learn from, and it carries the risk of making me think I have validated my own opinions against opposing points of view.
Completely agree. I had read this someplace, and I go by this statement.

"A good design is something which takes care of both extreme users, and not just a center large portion."

The user feature of a website should be such that, they should feel to interact either as real identity (gets them social karma) or as an anonymous user (without any concerns)

LinkedIn and Facebook Groups can re-think about having a better way of engaging people!
Absolutely.

The original idea for a better discussion platform came after using fb group for 2.5 years and feeling the pain. We had started the Bangalore Startups group (https://www.facebook.com/groups/blrstartups/)...

Although we keep getting new users fairly well, control and structuring discussions is completely missing..

Don't use the Web, don't build a "platform", build a peer to peer system, support threading, use local storage of both postings and read status information, use a native interface for fast and efficient navigation.

Or in other words: Re-invent NNTP.

Yes, NNTP has its problems, too, and could possibly use some additional features, but it's silly when you look at how easily you could/can follow huge discussions using slrn, and how badly current online discussion technology stacks up against that. It's all shinier now, but the usability is absolute crap for the most part.

I guess I should spell this out explicitly: Don't integrate data model and UI. That's the most silly thing about the web nowadays: Every instance of the same kind of content comes with a different UI, but without an API, even though it is all computer-readable data in principle. That is completely braindead, as far as the possibilities of information technology go. Wherever there is a discussion offered, I should be able to add it to the discussion software of my choice, so that (a) I have all the discussions in one place and with a consistent UI, which then can reasonably be powerful, as it's reasonable to, say, learn key bindings, for your one discussion client, as opposed to learning keybindings for every web forum out there, and (b) if you don't like the UI I am using, you can use a different UI that you like, and still participate in the same discussion. And in any case, native software is just so much more responsive than any of this web stuff.
I have the exact same views.

In one of the discussion on building the structure, I was amazed to find that people do not think in terms of API.

As for me, I have always thought of having 3 different layers/servers independent of each other. Front-end, Back-end and Database

It may be that we mean slightly different things here. I am not talking so much about the software architecture, but about the fact that the API between "front-end" and "back-end" should be the public interface. That does require a certain software architecture, though, of course ...
Just build your private API as a public API. That's what we've been doing and so far its turning out great
Yes ... But if the software arch does not decouple the front-end and back-end, then there are essentially no APIs.

An API approach is the best approach. Later it can always be opened to public.

This.

Every time I see attempts to use Twitter et al. as a "platform", I end up thinking how much better something NNTP-like would fill the role, if only it still had significant penetration.

NNTP had an adoption problem. Nobody built a mobile app for it :). Maybe someone can :D But twitter's more than just news.
I am not a big fan of peer to peer system, but will give it a look.
End to the passive-aggressive hellbans and service degradations (e.g. slowban, rankban) for obviously non-spamming users.

Re-show comment karma in order to restore trust, in the wake of awareness of rankbanning, since that will provide transparency in comment placement.

Hacker News still has a community advantage. (I'll take HN's irritating-- but at least book-smart-- libertarian naifs over the actual idiots of most forums any day). The moderation could use some work, though.

Effective moderation is always a pain. The biggest curse of moderation is the inability to keep everyone happy. However, it is a necessary evil, which can not be avoided as well.

This point does give me an idea on how to handle the moderation a bit better though. Let me polish the idea, before telling more about it.

My favorite discussion platform must involve food. Something special happens when people gather around food. And some type of structured moderation that becomes culture.
Food is an absolute must. Unfortunately, HN does not have that. :(

BTW, what do you generally like in food discussions? Is it good pics of food dishes, or recipes or something else?

He meant ACTUAL FOOD :P as in physically gather around food and discuss while eating.
lol .... yeah

I guess I got too immersed in thinking about discussions only. :D

Indeed. Trying to think outside of the box. Why not start a discussion startup with real-life discussions?
My favorite discussion forum is WordPress, Tumbr. Effective UI, easy to navigate buttons, Accessibility options, ease of app use on mobiles are the features I look at.
While going through this discussion, I realized HN does not have any form of notification system.

How important is notification for an effective discussion platform?

HN notify is kind of an external plugin... But considering that i did apply to means that yes, notification is important. But over-bombardment of notifications can be annoying
Yes. Another feature which is tough to design, such that it is not over annoying and yet solves the purpose. :)
The need for notifications is just a symptom of an idiotic system overall.

When was the last time that a company offered you to notify you by (snail) mail when there was a letter available for you to pick up at their headquarters?

It's a workaround for having to poll multiple event (posting) sources. Which only arises because we have multiple event sources. Which is completely unnecessary, technically. (I am talking about the user interface here, of course - it's perfectly fine to have the software poll multiple sources, or, even better, to have some protocol-level notification/push mechanism, of course, but the user should not ever have to deal with it.)

That also does not mean that integrating with email, say, is necessarily a bad idea - but not in order to tell a human that they should poll some event source, but as an interface medium for the actual discussion (that is, essentially as a mailing list). e-mail supports much of the necessary basic functionality, and it's also an existing infrastructure that give the user one UI of their choice for all communication partners.

How important is meeting new people on a discussion platform?

How often do you like to follow/friend/private message interesting people you may find on a particular discussion?

What's wrong with http://retroshare.sourceforge.net/ ?

It's encrypted, distributed, open-source. It has chat, messaging, file sharing, forums, etc.

What more do you want? If anything, why not just write a plug-in for it or otherwise contribute?

Reinvent Slashdot's meta-moderation. http://slashdot.org/faq/metamod.shtml

Don't treat users like their contributions are equal. Some users make better decisions than others. Reward them.

I think the Slashdot FAQ should be required reading before developing a comment system. There are a lot of lessons they learned ahead of the rest of the internet, problems others still have not solved. Something history, doomed to repeat.

http://slashdot.org/faq/

Thanks.

Community moderation is quite a nice concept. Slashdot has indeed implemented it quite intelligently.