Forums are not Dead !!! Change your vision, this is needed

[Nev_Dull]

Should I do what Nev_Dull says - try to make my forum - continue to welcome discussion from all sides.
or should I turn it into a right wing echo chamber? and thereby remove all the drama

I guess that depends on your goals for your forum. Do you want to provide a place for people to rant and spew or do you want a place where people can find all viewpoints and gain a more rounded understanding of the topic? What happens when there is no more opposing thought and all your members are left staring at the screen with nothing to rant about?

 

[Pete]

I think the whole this is more complex than that, though. One of the kickers you will run into is the practical reality of serving video (not so much images) from regular shared hosting.

Discourse doesn't run on your average $3.99 GoDaddy type hosting; it needs a flat minimum of a VPS or better, at which point your options for what you do with it go up. But the PHP end of the market in particular? The open source stuff is designed to run on a shoestring and catering to rich media is out of their capabilities because they're trying to cater to the folks who use terrible hosting that *can't* support richer media.

 

[Kaelon]

Discourse doesn't run on your average $3.99 GoDaddy type hosting; it needs a flat minimum of a VPS or better, at which point your options for what you do with it go up. But the PHP end of the market in particular? The open source stuff is designed to run on a shoestring and catering to rich media is out of their capabilities because they're trying to cater to the folks who use terrible hosting that *can't* support richer media.

I agree that the "PHP end of the market," if you mean "1990s / 2000s script kiddies" are out of their element to run modern software. But I disagree that modern software can't run on shared / simple hosting.

The problem with relying upon a dying platform is that there hasn't been anything even remotely approaching innovation for quite some time. Yes, the script-kiddies who migrated from PERL to PHP and don't really know how to run their own webservers are going to be challenged (and probably end up throwing a lot of money for hardware that they don't really need) to solve essential software-stack configuration concerns But you do not need "a flat minimum of a VPS or better" if you're an administrator who understands how to configure and optimize the right types of dependencies.

Some basic examples, using federated social media for images and video:

1. Pixelfed (federated Instagram-clone for decentralized social media): HTTP, SQL, PHP-FPM, and a number of dependencies; can't use a service that doesn't give you SSH access, but any shared service / non-VPS that gives you control over your own environment (there are many) can achieve this.
   
   
2. PeerTube (federated YouTube-clone for decentralized social media): nginx, postgres, redis in prod with a typical 2-core, 2 GB RAM instance. 1 Gbit/s uplink accommodates 200 concurrent 1080p streams. Totally doable in a shared load-balanced environment.

It's easy to rag on Discourse when you don't know/understand how to manage your own Discourse environment, and in my experience, the 1 GB RAM minimum swap config. translates to a 3 GB un-optimized / unconfigured RAM setup. That's easily the difference between an admin who does know how to administer their Docker boxes or configure their Ruby / Redis / Postgres configurations, and an admin who is looking for a one-click installation solution.

In short, you're right that modern and innovative software doesn't run on a potato the way that old PERL and PHP scripts did. But if you look at Docker pre-installs, DigitalOcean installation scripts, and the like; and, you know a little bit about how to manage, secure, and maintain your own instance, you can easily start to experiment with modern software. And while for some people this is seen as a nuisance, if not completely impractical; it's important to recognize that most of the applications being offered today - from Discord to Reddit, Facebook and Twitter and Instagram - are all essentially optimized backends with standards-compliant front-ends running in a load-balanced distributed scenario.

 

[Nev_Dull]

Damn you Kaelon for making me think so much!

This is my key issue: there is widespread scorn and ridicule re: social media, synchronous forms of communication (like Discord), or even the tiniest incremental experiments (like Discourse), as deeply undesirable.

Again, I find myself agreeing with you. There are some who rail against social media sites or Discord out of frustration and perhaps envy over the audiences they command. I think most of us, though, resist those other platforms because

• we like the pace and flow of that asynchronous communication that forums provide, and
• we enjoy the challenge, and yes, control, of running a self-hosted platform.

I don't think those of us in that camp are opposed to any software that proves better at providing those things. I'm quite certain I could reach a wider audience with a facebook group or by starting a discord. I also know I wouldn't get the same enjoyment from it, and as a hobbyist, that's far more important to me than numbers.

I think it boils down to something very basic: you need to design for what the audiences want, and the audiences want things to be simple.

I can't agree with this statement. Users don't always need simple; they always need clear and easy to use. I find this often demonstrated in mobile apps. While some are quite good, others try to make the app simple (or simplistic) which can create confusion by the lack of information and direction.

Looking at the start page of a well-designed and organized forum, I don't see anything else that is more clear or easier to use. There is (should be) a clearly visible tagline or statement explaining what the forum is about. Following that, is an organized list of the forum sections, each with a title explaining what topic(s) or type of content can be found within. Each of those sections also offers a visual cue as to whether or not there is content the (logged in) user has not seen. And inside each section, the pattern is repeated with the list of thread titles and cues indicating new content.

This basic structure fits with the principles of usability. All content is never more than a couple of clicks away. Users know what to expect at each stage, and are guided by visual or text cues towards the content they want to see. There is always an easy path back, if they make a mistake or don't like the content they chose.

Of course not all forums are well-organized; some are confused messes. As I said earlier, though, that is the fault of the forum manager, not the software itself. Forum interfaces are not particularly pretty, but it does what it says on the tin when it comes to delivering content to users.

I can't really speak to the management side of things, since I don't run a discord or facebook group for comparison. I know forums have plenty of room for improvement. Still, as a system designed primarily for text, I don't see huge leaps forward showing in the navigation and interfaces of social media sites that would make them fundamentally better or easier for new users, regardless of their age. Even the mobile experience of forums, while perhaps not as attractive as in a social media app, still provides an easy path to content.

I just don't buy the argument that forums are too difficult or offer too many barriers to attract younger users. I think there are plenty of other more plausible reasons, such as:

• perception - Younger users see forums as something from the past, so they can't possibly be of interest to them. Younger people have always been attracted to the new.
• findability - Google hasn't been kind to forums, pushing them down in the search results. Since few users ever go beyond the first page, it's not surprising that forums don't get traffic that way any more.
• patience - Younger users tend to have shorter attention spans than older users. They don't see value in longer discussions or waiting for answers, even when those answers may be more thoughtful and accurate. They also can't be arsed to look for a different forum when they have questions about a different topic. They prefer the convenience of the one-stop shop of social media, reddit, etc.
• social grouping - Younger users want to be where their peers are and their parents aren't. At one time that was forums, then it became facebook or twitter. It's become a slowly moving target, as new platforms arise and become mainstream.

 

[Kaelon]

There are some who rail against social media sites or Discord out of frustration and perhaps envy over the audiences they command. I think most of us, though, resist those other platforms because

• we like the pace and flow of that asynchronous communication that forums provide, and
• we enjoy the challenge, and yes, control, of running a self-hosted platform.

I agree, obviously. There is the hobbyist element to what we, as creators, want, and it comes down to control, like you said. We want a carefully curated experience, and social media - by virtue of it being largely distributed from a user-generated content perspective - dilutes the power of the administrator considerably and makes the experience much more of a lowest common denominator factor. You've seen social media innovate not through content, but through interface, and that's where (as an observer of these platforms) I find myself intrigued.

Users don't always need simple; they always need clear and easy to use. I find this often demonstrated in mobile apps. While some are quite good, others try to make the app simple (or simplistic) which can create confusion by the lack of information and direction.

I agree in principle, but you and I are going to disagree in the example you've given below:

Looking at the start page of a well-designed and organized forum, I don't see anything else that is more clear or easier to use. There is (should be) a clearly visible tagline or statement explaining what the forum is about. Following that, is an organized list of the forum sections, each with a title explaining what topic(s) or type of content can be found within. Each of those sections also offers a visual cue as to whether or not there is content the (logged in) user has not seen. And inside each section, the pattern is repeated with the list of thread titles and cues indicating new content.

This basic structure fits with the principles of usability. All content is never more than a couple of clicks away. Users know what to expect at each stage, and are guided by visual or text cues towards the content they want to see. There is always an easy path back, if they make a mistake or don't like the content they chose.

The forumindex is a perfect example of a failure in usability. Categories (forum names) and their details (number of threads and number of topics) along with the latest contribution is, frankly, information overload. How often did you judge the quality of a forum from the sheer number of topics? And do you really care about topics vs. replies/posts? Sure, as admins we do, and even as niche-content-divers, we do. But categories don't mean nearly as much to users as we think they do. Categories are organizing taxonomies for the convenience of a singular stakeholder group, and in this case, it's largely been for the convenience of the administrator/moderator.

Users generally care about very basic things, judging from what has made Reddit the largest and most popular forum clearinghouse on the Internet:

• The date and timestamp of the last contribution; and,
• The number of endorsements / upvotes that the contribution has received.

Categories are really meta, and all of the meta-information around categories only drives speculation, rather than awareness. So, if you look at what drives activity - on Facebook, Reddit, Twitter, etc. - it's:

1. Frequency of contributions; and,
2. Quantity of endorsements / upvotes / re-shares.

That's it. The forumindex, in my view, is the perfect example of a layout that hasn't changed in over two decades, that doesn't substantially contribute to more participation (if anything, it has the potential to dissuade potential new participants, who will get either confused, overwhelmed, or simply misunderstand the taxonomy presented), and undermines the key statistics that are of interest to your users: the number of active users that you have and how frequently they are contributing timely and relevant content. Tags instead of categories would allow users to dynamically re-organize and re-sort content on their own terms. It's also one of the reasons why Drupal, Wordpress, and taxonomically tag-driven publishing platforms tend to have substantially higher readership and commenting than contemporary forums.

It's also why so many Big Boards ultimately shut down and move to different unitary content management systems: because maintaining the categorical taxonomy, the organization of posts-and-replies, and the nurturing of community members to "stay on topic" runs contrary to the natural impulse of digital natives: to consume, rearrange, and contribute to content on their own terms.

It's both humbling and enlightening when you realize that social media did more than just displace forums as digital townsquares; it unseated forum admins as the authorities and conveners of content and it placed this content squarely in the hands of users.

 

[Pete]

But you do not need "a flat minimum of a VPS or better" if you're an administrator who understands how to configure and optimize the right types of dependencies.

Let me know when you find shared hosting that can run Ruby apps, Postgres, Sidekiq and the stack that Discourse runs on, where the server is managed so that the end user doesn't have to do significantly more than upload and go.

At least there are providers who will consider shared hosting for Node so NodeBB isn't entirely out of the water.

It's easy to rag on Discourse when you don't know/understand how to manage your own Discourse environment, and in my experience, the 1 GB RAM minimum swap config. translates to a 3 GB un-optimized / unconfigured RAM setup.

I was perfectly content with equating the 1GB option as the lowest tier on Digital Ocean or Linode, both VPS providers. Again, let me know when I can find a shared hosting provider that supports this and means that I don't have to think about server setup just to run a forum.

Or I could install the other packages that don't require system admin knowledge.

you can easily start to experiment with modern software

You or I can, but the average forum owner who installs from Softaculous or Fantastico (as still happens regularly)? Not a hope in hell. And they don't *want* to. They just want to get going with the forum stuff.

You really should hang around the support forums of the forum packages sometimes. You get to learn the kind of people who are running forums and barely treading water; the ones who _can_, _will_ and won't appear in these spaces, but you get to interact with the ones who are struggling, the ones who just want to run a forum about their hobby, aren't technical and should not be expected to learn a *mountain* of material to get going.

are all essentially optimized backends with standards-compliant front-ends running in a load-balanced distributed scenario.

They're all backends the user doesn't have to run, though, or think about how to set up and that's the key factor here. If you seriously think that getting set up on these things is not a consideration for people just because it's not a problem for you, sorry but you're utterly delusional. This is why SaaS services are so attractive - because they're next to zero effort to get started and next to zero effort to keep running (barring moderation).

There is a reason WordPress - for all its many, many faults - is the number 1 CMS out there, bar none. Because it'll run on a potato, any hosting in the world can run WordPress, and people can get it installed either with one-click install packages or they'll figure out how to get it to install. And if it can't, people will go somewhere that can.

So, if you look at what drives activity - on Facebook, Reddit, Twitter, etc. - it's:

You forgot 'newness' in there.

The forumindex, in my view, is the perfect example of a layout that hasn't changed in over two decades

Many things haven't changed in over two decades. Cars, for example, cars are largely laid out the same way now as they were two decades ago, heck, other than the EV (admittedly a major change), car layouts have broadly been the same for best part of a hundred years. And even EVs don't *wildly* change the game externally or from any of the interface standpoints.

Form follows function, always has. My point is that just because it hasn't changed doesn't automatically imply that it should.

Even the platforms that go in for tag-based categorisation by default (like Flarum) find themselves coming back to approximate the forum index style.

if anything, it has the potential to dissuade potential new participants, who will get either confused, overwhelmed, or simply misunderstand the taxonomy presented

Thus it has ever been - a bad index absolutely confuses people for sure. Which is also why the forum index shouldn't be the front page of a site, and should be the front page of 'the forum' component rather than 'the site' component, to help with getting people in the right place.

But if you're arguing that categories and boards are a problem, please explain to me how categories and channels in Discord is fundamentally not the same thing. The only differences there are you get *less* information on Discord than you do on a forum, not more, and that every design challenge every forum owner had to figure out has been/is being rediscovered by Discord admins.

In some corners there are literally circulated documents on 'how you should best set up your Discord for <type of community> servers'. And even then it's still incredibly confusing as a newcomer.Too many servers do too many weird things with bots in an attempt to stand out and show off (or, conceptually, automate their admin away, reimplementing a number of behaviours that forums have had for 20 years)

Tags instead of categories would allow users to dynamically re-organize and re-sort content on their own terms.

The problem with tagging is that it doesn't work like that. Unless you're expecting every user to do their own tagging (which they won't, no-one has that kind of time), you're reliant on other people doing tagging. Or, more realistically, other people doing tagging and then moderators standardising the tagging so the previously-freeform tags mean approximately the same things to people. This is a recurrent problem that Stack Overflow has about people being too enthusiastic with tags, or not being enthusiastic enough, meaning that questions don't get shown to the right audiences.

Or you could go to the hybrid form that something like Flarum uses which allows you to tag a topic in multiple preset categories, but you still have to define those categories meaningfully and even then there's no guarantee people will get it right.

Or, you know, it's the same thing as people posting in the wrong board. It's going to happen, and giving more latitude not less to making it happen is going to make it happen more. That's just life.

It's also one of the reasons why Drupal, Wordpress, and taxonomically tag-driven publishing platforms tend to have substantially higher readership and commenting than contemporary forums.

You see the effect but not the cause. The reason these things have more meaningful and more impactful taxonomy is because it's curated by design. Few authors/publishers -> many consumers, the consumers aren't the ones doing the tagging or curating. But in the forum world, it's many authors -> many consumers, and the tagging and taxonomy is going to be flakier by design because it can't not be. Different people, different points of view, not everyone is going to agree on that taxonomy at any given point in time.

And if you're going to hold up WordPress and say 'WordPress is more popular than forums', well, um, yes. That's like saying 'cars are more popular than skateboards' - without thinking about any of the context of what any of that means, or whether it is in any way a useful comparison (it isn't).

it placed this content squarely in the hands of users

Who rely on other people (or algorithms) to deal with the messy business of classification for them. How often do you hear about people just scrolling on Facebook? They're not choosing what they're seeing except in the most broad terms, they're seeing what the algorithm chooses to show them - often as a subset of what is out there. Curated for what works best on Facebook, not about what content there really is.

 

[Nev_Dull]

Pete got in there first and covered a lot of the things, so I won't go over them again. But there are a few items that stand out here.

The forumindex is a perfect example of a failure in usability. Categories (forum names) and their details (number of threads and number of topics) along with the latest contribution is, frankly, information overload. How often did you judge the quality of a forum from the sheer number of topics? And do you really care about topics vs. replies/posts? Sure, as admins we do, and even as niche-content-divers, we do. But categories don't mean nearly as much to users as we think they do. Categories are organizing taxonomies for the convenience of a singular stakeholder group, and in this case, it's largely been for the convenience of the administrator/moderator.

So I explained how the basic forum structure complies with usability standards. You're saying it fails because it contains section titles and extraneous numbers. I think I missed something.

I specifically didn't mention thread/post counts because that is internal information that doesn't matter to users and should really only be seen by admins and mods. I also specifically referred to forum 'sections' rather than categories because that is what they represent to users, in the same way a TOC refers to chapters or sections of a book. Regardless of that, what it says is the important bit to users. If I'm looking at a forum and see a section called "Grooming your Bernedoodle" I can be reasonable sure what sort of discussions I'm going to find there.

"Categories are organizing taxonomies for the convenience of a singular stakeholder group"
Well, yes they are an organizing structure for convenience. As for the rest, you're assuming facts not in evidence. I was specifically using the example of a well designed and organized forum. Any forum manager wanting to create such is going to have at least a basic understanding of who their primary audience is and what they are interested in. So the forum sections (or categories, if you like) are a reflection of what that audience is looking for, not the forum manager. As that understanding grows over time, the forum structure will evolve to better suit their needs.

Tags instead of categories would allow users to dynamically re-organize and re-sort content on their own terms.

First, tags are categories by another name. So, if they are called categories they are confusing, and if tags, they're good?

Secondly, I'm all for taxonomies. Done right, they are indeed powerful tools for organizing content. But I cannot fathom the notion that these younger users, who you've argued aren't bright enough to figure out how to use forums, could somehow manage to develop and maintain taxonomies that would in any way be better than what forum categories offer. I guess we can look at the efficiency of hashtags on social media as an example.
#useless, #wtf, #confusing

 

[Pete]

To add to that, if you look at some of the trending hashtags on Twitter you do not find the subject in question but the exact opposite of you are observing anything political. E.g. In the UK the prime minister Boris Johnson is a very polarising figure, #BackBoris is a regular fixture - but to my estimation half or more of the recent users are using it in tweets aimed at ousting him precisely because “I don’t #BackBoris” which is clearly a subversion of intent if not in context.

I would even point out that the number of topics/posts in a board can be useful - a site with many boards and little content is a problem.

 

[imandings]

Unfortunately, I truly believe Discord will take over from forums due to the lack of innovation from forum software creators.

How I hear you say? By building their own forums system inside of Discord.

 

[Pete]

That doesn’t prove that *forums* are dead though, it just proves the current incumbent implementations of the concept need to up their game. This is normal however, just as vBulletin was once top dog, things evolve.

 

[Nev_Dull]

This is just lipstick on a pig. This isn't Discord taking over from forums, it's using Discord to create a forum. The only "innovation" at play here is leveraging the current popularity of Discord to (hopefully) gain users more quickly. In the bargain, you have to give up the control that self-hosting provides.

 

[Web Diva]

I believe rethinking the "forum" requires "re-imagining" one's purpose. As Nev_Dull mentioned in post #111 (yes, I like post number designations too) his key word was innovation and beyond that, asking yourself, "what is your ultimate goal", simple niche or world domination? I think if you ask that and are honest, you will be happy with your purpose, happy with your development and mostly pleased that you are providing a nice place for people to engage on a wide range of topics. Isn't that why we do this for in the first place?

 

[DigNap15]

If Kaelon thinks that forum index's are the problem and that forum are too complicated, then....
Why don't they remove all categeories from their forum and just have one long list of posts.

Then show it to us, and then we can see how it goes.

The sad thing to me is that there are only the same old same old members contributing to this topic which just goes round and round.

We all know that forums are under pressure or dying, and we all want to blame someone.
As if some magic new feature will save us all and bring thousands of new members.

To say that people find forums hard to use or navigate is just ridiuclous, most people are computer and phone literate these days.

The fact is that there is only so much time a person can have onlline and there are so many apps to compete for their time

Youtube - which has thosands of topics you can subscribe to and comment, Facebook Groups, Reddit, Discord, you favorite online newspaper,

 

[Kaelon]

If Kaelon thinks that forum index's are the problem and that forum are too complicated, then....
Why don't they remove all categeories from their forum and just have one long list of posts.

Then show it to us, and then we can see how it goes.

The sad thing to me is that there are only the same old same old members contributing to this topic which just goes round and round.

We all know that forums are under pressure or dying, and we all want to blame someone.
As if some magic new feature will save us all and bring thousands of new members.

To say that people find forums hard to use or navigate is just ridiuclous, most people are computer and phone literate these days.

The fact is that there is only so much time a person can have onlline and there are so many apps to compete for their time

Youtube - which has thosands of topics you can subscribe to and comment, Facebook Groups, Reddit, Discord, you favorite online newspaper,

The solution to the "better forum" is already out there: it's Reddit and Facebook Groups.

Discord is also presenting a very compelling anti-Forum experience that has displaced most gaming and hobby-related forums.

So, yeah, the solution is there, and the age of the independent self-hosted forum operator has effectively gone the way of the BBS'es from the early 1990s being displaced by the web. Forums were displaced by social media.

Game over.

 

[Pete]

Thankfully not everyone agrees with you.

 

[Kaelon]

Thankfully not everyone agrees with you.

Whether people agree with me or not is beside the point; the facts are undeniable. Audiences have moved on, and they're never coming back to legacy-style forums. The fact that most posters here are clinging to their precious UBB-style linear forums which haven't changed in 20 years like an indignant matron might cling to her pearl necklace is classic head-in-sand-style passivity. Users have moved on, generally.

Just like callers generally moved on from BBS'es. Yes, there are still BBS'es around and operated as curiosities and for nostalgic purposes. Some of them are even very active; but their volumes are tiny compared to social media, and they are far from seen as ubiquitous.

There's been proactive guidance on how to remain relevant, including:

• Focus on micro-content (around ~250-300 characters, not words)
• Simplify the user interface, remove extraneous statistical data
• Go mobile-first on accessibility
• Make listings all synchronous so that users can see replies / updates without having to refresh or navigate
• Integrate with broader multi-forum networks, and leverage the ubiquity of social media

There is a huge loss of control, of nuance, of long-form, asynchronous content control. But that's where audiences have gone.

The fact that this thread exists means that people agree that this is a problem, are (depending upon who are) bitter or disappointed or frustrated or nostalgic for the loss of users, and are looking for solutions. Simply saying "I disagree, this isn't a solution, I don't want to be like Reddit or Discord or Facebook, I just want my forums from the 1990s and early 2000s" is only going to hasten the demise.

 

[Pete]

You keep repeating these facts as if they're true. The evidence does not support your conclusions, sorry.

 

[DigNap15]

Kaelon
Why would I want 250 character posts or memes on my forum?
I dont want them, and I don't encourage them
Moat of my members like to debate specific topics

Sure I could start a Facebook Group or a Reddit, but then I would have no control, and would be open to their whims and censorship etc.

 

[DigNap15]

If I removed my statistical data would I get more members?

 

[Kaelon]

You keep repeating these facts as if they're true. The evidence does not support your conclusions, sorry.

Which facts are untrue? What evidence does not support which one of these conclusions?

1. Most user populations have moved on from most forums.
   
   
2. The majority of forums have either closed or gone completely inactive over the past 20 years.
   
   
3. Many forum developers have either slowed or halted meaningful development on their platforms, and are opting for Enterprise-focused SaaS models instead of self-hosted solutions, in an effort to recapture lost revenues.
   
   
4. Advertising revenue for forum-based page-views have plummeted, and most search engines now penalize (down-rank and hide) forum content, further decreasing visibility.
   
   
5. Reddit, Discord, and Facebook have largely supplanted forums as the intended user audiences for news and discussions, community-building, and group engagement.