I guess by āthis industry,ā I am referring less to the prevailing current-state of SaaS / PaaS, which Invision has clearly shifted to as you rightly point out, and more the industry of selling software-as-a-framework (into which forums had largely been concatenated, thanks to how people twisted forums to serve as broader CMSāes). Forums arenāt used this way anymore, and that is what I lament: because in the software-as-a-framework approach in the single license approach, it spawned a veritable cottage industry of add-on developers, integrators, solution architects, and all sorts of really fascinating software use cases.
The market absolutely does win. The question is whether the market definitively chose Facebook and Reddit as the winners, though. After all, this is where most of the forumspace and blogosphere went in the 2010s. So even as Facebook reinvents itself to chase the TikTok short-form video-craze, the question is whether interest groups, hobbyist organizations, and even content management frameworks will ever want to exit the SaaS space to something that is more owned, controlled, and managed by the people and organizations that depend upon them.
I see Mattermost as a very fascinating answer to this question. In an age where the pandemic-era āwork from homeā craze has resulted in a mad-rush towards Slack and Teams as the two dominant collaboration and communication frameworks (with RingCentral/Avaya, and even Zoom, trying to jump in at the last minute), there is strong demand in financial services, workforce management, software development, government and security, and cyber defense spaces to have something that people just āown,ā and can accordingly control without relying upon a third-party. Yes, ISO 27001 and SAS Level 2 / 3 audits and certifications exist for a reason.
The next five years will be a very telling time to see how people work and play together, and what tools and systems they will elect to use. I also think it will largely determine the fate of social media platforms, like Facebook, as the arbiters of online content. And I view all of this as an exciting opportunity.
What is interesting in the statistics that you show on software installations for Invision vs. Discourse isnāt so much the absence of a decline; it is the absence of conversion.
If you look at how the hobbyist space collapsed and, as weāve discussed earlier, has resulted in total ecosystems blaming Facebook and the like as the destinations of users, itās also clear that no one has really enticed people back from Facebook Groups or Reddit subreddits. Or presented a credible alternative that has slowed or converted the growth of Discord, for that matter.
This to me suggests a rather permanent exodus, rather than a successful non-decline. Yes, the market-share continues to stabilize and make modest improvements. But the size of the total addressable market (not to mention each companyās serviceable obtainable market) has collapsed.
Completely agree. We are about to enter a new epoch.
BTW, Iāve really enjoyed this deep dive. I canāt think of another platform where we could have had this flow of thoughts and conversation without being distracted by other channels or random sub-conversations cropping up to interrupt.
Yes, forums will never be as dominant as they were, but that is because there are now more and better options for communities that donāt require the scale or substance of a forum.
This is a market correction and I am fine with it. We continue to work with the market than move against it.
We were the first of our āClass of 2000ā to create a substantial SaaS product to safe guard against the loss of the hobby market which was populated with highly technical individuals who knew server management. We also put resources into enterprise sales and marketing early as a lifeboat to allow us to regroup and refocus on new emerging markets.
Given the āprice it low and stack it highā approach from forum vendors in the 00s and the switch to increasing the lifetime spend of a customer with a SaaS product, Iād expect to see a huge decline on raw numbers as communities come to a natural end or as forum vendors withdraw the cheaper self-hosting product, and then a slow and steady growth based on SaaS numbers alone.
I 100% agree with you. This goes back to my sense earlier (and this is something you also rose further above in the conversation) that forums were being used for a lot of weird non-forum use cases. Like CMSāes. Group management. Blogging and comments. And so on and so forth. Facebook Groups, Reddit subreddits, and Slack/Discord/etc. voice-and-chat communities have largely supplanted forums for many of these use cases.
What I think remains is a very compelling use case for the Support Discussion, Question-and-Answer, Ticket-Deflection business use case. Forums are indisputably valuable here, and the theory of change is largely accepted. But so too is to the disruption that Jeff Atwood, and other early entrants, are presenting to the way that forums are structured. It may not be a revolution entirely, but I think itās a very aggressive evolution of the platform that has strong value-based implications, and again, this goes back to Atwoodās philosophical views that I think are gaining traction, and not all of us agree with these traits 100% of the way either (I certainly have mixed feelings about them):
Less conversation ā not more - is essential to a strong signal-to-noise ratio.
Up-voting and down-ranking content should elevate and hide content.
Simpler UI means that some things - like post counts and latest replies - donāt really matter.
Some of the other trends are less controversial, but no less dramatic in their scope:
Mobile-First Design means that the age of high-density content is coming to an end.
Image-less design aesthetics eschew branding in favor of a look-and-feel that is modern and simplistic.
A preference for embedded images and video are driving a yearning to better engage with this content in different ways (such as AI-assisted speech, which is now largely affordable).
And then there are some clear market directions weāre getting that most people will hate ā at first ā and then eventually love. Remember: users hate change. But they are rarely right; you donāt give users what they want. You give them what they need. And that seems to be this:
Convergence of real-time synchronous content with the asynchronous permanence, as evidenced by real-time chat being integrated more substantially in forums now (both Discourse and Invision are offering compelling roadmaps that show this use case).
Elimination of clutter, old and outdated conversations, and the need to drive automation in content management is putting an end to forums as these āgreat archives,ā since the archives are now largely un-indexed and undiscoverable by search engines (and not seen as valuable or timely or relevant by users).
The use of machine learning and genuine artificial intelligence in automated moderation of posts, content, and even users themselves is leading to a future where the role of the moderator is far less administrative and far more a specialized contributor. This has radical implications for how forum-based communities will be built, structured, and governed.
For my niche, which is collaborative content management, it also presents an exciting solution to the traditional scalability of high-touch use cases where my moderators were largely responsible for driving every conversation and/or decision, or acknowledging and rewarding contributors for their participation. Discourse and Invision both have new-world and old-world automated approaches to this, but I see the needle moving very quickly towards automation and an eventual reduction of the human element. This will make the time we do spend conversing far more valuable, and far less dealing with trolls and troublemakers.
I definitely have enjoyed this deep dive, too. There is no question that, for those of us who grew up with technology in the 1980s, forums represent a civilized medium that allows the free-flowing linearity of give-and-take, of a deeper-dive to explore nuances and substance of what people are thinking.
I look at Reddit as a very clear leading indicator of what drives people to asynchronous communication, and I genuinely worry that future generations (at least, people who are younger than you and I are!) are expressing themselves in very different ways. They tend to be:
Shorter-form.
More image-centric, video-driven.
Reaction-based to news/events/trends/content.
I do not see an appetite for walls of text outside of Gen Xāers and early Millennials, and instead see later Millennials and Gen-Zāers seeing conversations like these as āa waste of time,ā or āfree labor.ā Trying to embrace some of the shorter-form, image-centric and video-driven content, and reaction-based entry-level interactivity will be essential to grabbing the attention of new audiences. Because these long-form discussions are higher up on the Engagement Pyramid for people like you and I who are already invested in Rosie Land, the Forums Ecosystem, and esoteric conversations about software, technology, and the economics behind it all.
For new members, they need easier bite-sized chunks to grab their attention and convince them that contribution is worthy of their time. And forums ā whether itās Discourse or Invision ā arenāt doing a good job of making the entry-stage, lower-levels of engagement accessible or effective, in my view. I wonder, though - is that less the job of the forum, and more the job of the blog/CMS?
This rubs the wrong way. I would argue the other way around - vision is expensive, execution is cheap. In the sense that if you donāt have the right vision and strategy, you can end up executing years on end without any results. At that point, even if the execution is flawless, driving the wrong product in the wrong market, for the wrong user, on a wrong problem-solution space will cause missing Product Market Fit, and failure.
I was reading some tweet of a guy that worked at Skype the other day, where he thinks scrum had anything to do with Whatsapp taking over because they started using scrum too late, even if Whatsapp had only a 50 people team, vs 1000+ at Skype. Noā¦ the issue wasnāt execution or how they were working. But what they were working on.
The same is true for companies that eventually figured out the vision, and pivoted (Segment, Twitch, many others). For years on end, they were executing the wrong vision, then they fixed that and succeeded in no time.
This is such an important topic I canāt emphasize it enough.
Are there any other communities where you spend time (time spends you)?
I agree. The more preventive moderation, the better. Iām curious if you think the role of the moderator changing, would mean fewer moderators, or the same / more, focused on a different kind of work?
The approach weāre taking with our product is to charge per moderator seat vs what others are doing as to align ourselves with our communities and help them grow (which would mean more moderators ā increased revenue) making everyone happy.
Iām focused on the synchronous aspect of online communities, yet to be able to position ourselves in the market I actually looked at both async and sync markets, with their specific target groups. And there is some innovation coming for forums too (ignoring the fact that people voted for Twitter as such too - https://twitter.com/andupotorac/status/1554074981855895561).
So thereās Outverse, Heartbeat, Beams.fm (audio), Circle and Twist (workspace).
Which AI platform are you using? I think AI is the next best thing to come for conversations. In fact, I suspect that about 40% of our products, at different stages, will be AI-based. Thereās just soo much potential out there thatās absolutely not being used at all.
We are building our own tools and using the AWS sentiment analysis service. We had a call with their engineers and we had access to some new tools they are developing.
It really isnāt - not in the real world at least.
Iāve seen a lot of people with truly impressive visions but they never made it beyond a spec because without the funds to build it, it is just another idea in an endless sea of ideas.
IMO, this has more to do with the Discourse ecosystem.
Other enterprise platforms have design baked into their onboarding process. Or the partner ecosystem is strong enough for 3rd party firms/consultants to deliver it after the fact.
Discourse doesnāt have that, and UI customization doesnāt seem to be a priority.
So unless an org is willing to fund the work for theme customization, every Discourse instance is going to look identical, aside from the basic logo swap and color tweaks.
Great perspective. This also explains why the forums listed on Discourseās Customers Page consists largely of organizations and institutions that either have in-house design and marketing resources, or the capital budgets to outsource. That said, editing templates and CSS files allow you to create very richly varied looks.
You donāt need to be a Discourse customer like gaming fiants Geerbox, CCP/Eve Online, Unreal, or Funcom to create unique looks. Scroll on down and youāll see Banks (Nubank, especially, has a stunning look!), Wireless / Transmission firms (look at Western Digital and Wyze), and the obvious enterprise leaders across Envato, New Relic, and Zoom. Each looks unmistakably Discourse, but thereās solid variation on appearance (especially when you use imagery as headers, like New Relic and Nubank do).
This is a fantastic discussion and Iāve enjoyed reading the differing perspectives. Iām not going to jump in and make it all about Discourse, outside of one clarification:
No budget required. We build fully customised themes for all Enterprise clients as part of the engagement. The sites that look more vanilla are because those customers opted for that.
I would argue that we are sitting squarely in the Enterprise space these days, along with Vanilla et al. (although not at the expense of our smaller customers or self-hosted ecosystem).
Welcome to the conversation! Iām thrilled you could join. As you can see, Iām quite a fan - something that I started following closely when we started connecting our political and non-profit clients with Stack back in the early 2010s. Discourse has been a fabulous and much needed jolt of innovation to this space.
How extensive is the Discourse help documentation for in-house teams who want to opt for a more custom look? We canāt all have the vast marketing and visual design departments that Blizzard Entertainment has, but you have enough amateur design talent that with enough adoption help for your mid-market and retail customers, they could serve as shining examples that amplify the Enterprise segment.
Iām a bit hesitant to post too much because Iām not sure what @rosiesherryās guidelines are around talking about our own products, but Iāll answer your question in the mean time.
We have a full theme CLI and a number of pretty extensive documents that offer guidance.
