Discussion in 'Community Organization' started by Wayne Luke, Sep 15, 2005.
what do you mean with the term forum structure or methodology.
Like Wayne mentioned in his starting post, that the two column index page on Sitepoint (which is no longer there) is not really innovative. Restructuring the same data isn't innovative in my book. Or the methods of displaying a discussion or having a "thread" be the start of a discussion. Those methods are basically the best methods for a forum. I can imagine something slightly different on the discussion viewing and workflow, but in effect, my ideas are still just a threaded discussion.
Discourse is a great example. They've taken away the view of categories with sub-categories in the forum index and instead, have them in a flat structure. The restructuring of the categories isn't innovative and it is actually restrictive. Although I don't think a lot of categories is a really good idea, not being able to have sub-categories is also the pits.
It's disappointing that forum UI hasn't improved a whole lot in those nine years. However, I'm hopeful. I think things are looking up and several developers/development groups are stepping back and trying to rethink things from a UX perspective. In the old days, I think features were added just because they sounded cool, not because they helped a forum achieve its goals. That's starting to change, spurred on, I think, by new entrants to the forum software space.
I wish there was a nice list somewhere of the new guys that one could look to to crib ideas from. I'm aware of Discourse, NobeBB, Muut, ModernBB, and esoTalk which all have clean default themes (too clean some ways, making them forgettable) and seem to be trying to add features in a thoughtful manner. They make missteps too, but I guess they're new enough that they haven't accumulated a lot of cruft yet. Anybody have any other favorites to look to for UX ideas?
But the goal is useful online communities, NOT forums. The forum companies need to see themselves beyond just forums (threaded discussions).
For any community wanting to do or accomplish something .... threads are terrible data structures. Forums that rely on just chit chat will increasingly be ignored by Google and that low level chit chat will be done on facebook and twitter.
Agreed. That's not innovation ... but they have been creative about thread viewing.
Right. That is what I was getting at. Forum software, is forum software, is forum software. A dime a dozen now. If anyone wants to innovate, they need to think of online communities as much more than just a forum.
I wouldn't say it's the greatest thing since sliced bread, but (with admittedly little hard data) would defend the proposition that a default Discourse installation is better at facilitating discussion among today's computer users than a default install of most other forum software packages. And that's a good thing.
That is a good thing.
Exactly how is discourse doing this ?
They actually still have categories view in place, but its disabled by default with their reasoning being it should be used to solve problems that may arise as your forum activity grows (like need for real segregation of discussions).
Twitter Dev forums are example of Discourse installation that uses this mode: https://twittercommunity.com/
It's what they're passionate about. AFAIK, *none* of this is exclusive to them, but they put it together in a nice default package. Here are a few key things off the top of my head:
1) They try to get you started quickly--log in with existing credentials...google, facebook, twitter, yahoo,
2) Easy overview of who is posting in threads (that ugly avatar column), attempt to show you what's hot (they're tweaking things in that regard) without including data you don't care about. (i.e., don't distract you from the threads themselves).
3) Attempt to show discussions topics intelligently--infinite scrolling (not a huge fan of this myself, but it makes sense to do it), by default collapse very long threads to the most important posts, reading time summary, remembering exactly where you left off in every thread and instantly returning you there when you go back to it,
4) Include affordances that are both useful and the facebook generation already knows--e.g., likes, at-mentions, etc., but omitting things that don't help
5) Eliminate surrounding elements that don't matter as much to the conversation (postbit, the way they display dates, etc), but they still give you context in the thread of how posts relate to each other in a nice way.
6) Good reply tools. Can reply while reading. Nifty cards can get auto-embedded in your posts for various links in an actually useful way. Easy way to reply and fork to a new thread when it makes sense.
7) Everything is realtime.
They're working towards making things a pleasure to read and respond to, plain and simple, as well as helping you find what's mostly likely going to be interesting for you to read. We'll see whether they're on the right track in doing that, but if you read meta.discourse.org, it is obvious that they have a particular focus, an opinion if you will, and they're trying to get to that.
EDIT--I forgot a Discourse feature I'm a big fan of--their Trust Level concept, which *might* be unique to them, at least in a default install. It makes a whole lot of sense to increase what your users can do automatically as they demonstrated their value to the community.
It's a step in the right direction, although I believe that redesigning the forum software or user interface is not the most crucial step. One must first analyze how people use forums and other social networking platforms. What do they want? What are they thinking? How do they feel? In other words, I believe that to move foreward, one must look at things from an entirely new perspective.
Right Zordoz--that's what I meant by "rethink[ing] things from a [User Experience] perspective"--but once you've done that you do actually have to build interfaces informed by those thoughts. It's a difficult task and some folks seem better suited to it than others.
It's very difficult, much more difficult than programming, as I continue to learn each and every day. I've been conducting a sort-of informal use-case analysis and the results have been eye opening... to say the least.
I'll have to see how they do this.
Got an example ?
will have to see what you mean.
more than what Xenforo does ? ie. you click a thread title you've been to ... and it takes you to the newest post..
I saw that. I thought it was cleaner.
You mean like alerts and search results ?
I like the idea.
What is an example of automatic escalation of a user's ability ? (ie. 5 posts = can posts links ?)
>Re: collapsing threads
Here's a random example from a discussion whether Discourse should have signatures. It's probably not the best because people were deliberately trying to mess around in that thread, but the collapsed view lets you see what's important in the thread and hides most of the messing around (so maybe it *is* a good example. ) I chose it because it happens to be one of the longer threads on that board.
>Re: reading time summary
In the same thread, if you turn off the summary (button in the gray area below the OP), you will see that the text changes from "You're viewing a summary of this topic: the most interesting posts as determined by the community." to "There are 101 replies with an estimated read time of 14 minutes." BTW, the area right above that is pretty helpful too. It shows links from the thread as well as counts as to how many times thread reads clicked on each link.
>Re: returning you to last post read
Yes, like Xenforo. As I said, not exclusive to Discourse, but its inclusion in the fairly feature-lite Discourse package is a sign of its focus.
>Re: trust levels
For the full details, which they're still designing/building out, see this thread. But in summary, brand new users can't flag posts, attach images, post attachments, etc. You quickly get these restrictions removed--nothing new yet, right?--but then dedicated users automatically eventually get trusted with things like moving/renaming threads, and you can even eventually get to being able moderator-like abilities such as editing posts, closing/pinning/unpinning threads, etc.. That last trust level is currently only available via a manually promotion--they're working on making that automatic in a way that makes sense too, but as you might imagine it's tricky. Remember, the guy behind Discourse is the same as the guy who is behind Stack Exchange, a top 100 site, so having community driven features is probably going to be in Discourse's DNA. At Stack Exchange, things are very much community driven. Moderators generally come from a community process, not from on high, and just like Discourse, as you get more rep there, you're allowed to do more on their sites (including editing other peoples questions, etc), same as Discourse. It works there, although many thought it wouldn't. It's hard to argue with the results (being top 100 as already mentioned).
Xorlof - Thanks for your post on Discourse thread features. There are lots of features I like in StackExchange, so it won't surprise me to like Discourse as it moves forward. My friend runs a Discourse site now, I'm going to have to check it out more.
I'll say right off .... I really like the concept of [Summarize this Topic] . Alot of my criticism of threads is they become less helpful over time and can be close to useless to a new joiner to a 100 post thread. Of course, implementation is everything ... so I'll have to see how it works in practice. I'm pretty excited about it. Certainly s.molinari and Rafio should evaluate this feature for themselves.
I see Discourse has:
Comments in Posts (ie. Reply to Post (not Topic)) Vilandra
It is a hated feature in vB5, and almost universally trashed as a feature by Xenforo users (see my comments in posts threads on Xenforo.com).
Xorlof - have you thought of using it ?
I love Comments in Posts
Yes, and world peace is going to happen at some time in the foreseeable future. Because that's basically what your dream social network/forum hybrid is, some sort of one world order utopia which simply cannot ever happen.
The reason people set up so many different forums across the internet is because people by nature cannot ever all co exist in one space. Opinions are too diverse for it ever to be possible, and no one can ever run a system that pleases all (or even most) people. By definition, the policies of any one community cannot ever be applied to all others. It's also why the world is and always will be full of different products and services, because no one solution fits all.
That and well, even if you did somehow succeed, it would just as easily splinter the minute a group of forums didn't like being associated with a bunch of others, or disagreed with some policy enforced by the network for whatever reason.
Back to the main point though; forums don't have a lot of 'innovation' in their design, because for the most part, they don't need it. And even if they did, you couldn't sell it, because people are resistant to change. It's why stuff like Discourse will never become as popular as they claim, since the minute a big forum/site switches over to it, the memberbase would quite literally revolt. Probably with their own spinoff site using the old style and software.
Same reason for why companies and developers don't embrace such change. What, you going to try and force a new type of design on your customers, somehow? On the assumption that they won't just switch to a different script that offers what you used to offer? If IPB or XenForo or whoever say, went the Discourse route, their sales would probably just fall off a cliff due to existing owners being reluctant to upgrade or switch to the new version.
Basically, forum software doesn't innovate because:
1. Many people don't see it as necessary
2. The software makers don't want to risk another vBulletin 5 fiasco
3. Existing sites + massive change = full fledged member revolution
If any major changes are going to happen to community software, it'll probably involve a script that's meant to be entirely different to forums, with a different purpose and structure. Sort of like how people didn't try and force upgrades on Usenet or chat rooms, they simply came up with new solutions for similar (but still noticeably different) problems.
That's fear talking, fear of unknown consequences, hence resistance to change. Remember Myspace? It was hugely popular and people thought it didn't need to be improved, then this little-known startup called Facebook came along and changed the world.
Ok, thanks for the info. I stand corrected on the categories with sub-categories.
And I don't. Not for the general public, at least. They tend to dampen forked off discussions. I can see why some people would like that, but discussions are exactly the reason why forums have been so popular and good and to basically force discussions into a dead end is not a good sign of progress.
Discourse's comments, on the other hand, aren't really comments either. They are a listing of replies related to that one post. It is different to vB's comments, which basically breaks the flow of conversation. The only issue I see with Discourse's related reply view is it confuses a new user and makes them read content twice. As one user on Sitepoint said, they are now getting used to ignoring the "X Replies" button, which means they are using Discourse like a XenForo, by using quoting to reference a post in a reply and to keep the reply in context.
In other words, Discourse's supposedly innovative "replies to this post" view is a feature nobody needed or wanted and thus nobody uses.