Yes, an online forum is a barrier. But in-person ethnography at this scale is an even higher one. So, we go for the lower one. Most people will not get it, and will not participate. That's OK. It's always been that way.
@johncoate can tell you how people got on the WELL: get a floppy disk mailed to you, then install software and configure their modem, then endure dial-up speed and a text-only interface. And pay for the service. Of course, most people thought they WELL users were weirdos and would have not been caught dead doing any of this. That's OK, because the WELL did not need everyone. They only needed a few thousand people to get on board, and they got them. That's OK for everyone, both participants and non-participants. Free country.
A good rule of thumb is not to go out of your way to accommodate people that will not go out of their own. Those early WELL adopters saw a clear value proposition, and accepted to jump through the hoops of 1980s computer-mediated communication in its name. With online dialogue, 95% of people are just not up for it. They are on Facebook. They are on Whatsapp. They wax lyrical about face-to-face contact. They point to UX glitches, but, in my experience, they are just not interested enough. Nothing you can do will get engaged. This is completely understandable, because written, online dialogue is hard work and not for everybody, including many smart, energetic, articulate people. You still do work on your UX, but you do it to accommodate users, not non-users. The 5% who do see the value.
It's been the same for me throughout the years. Visioni Urbane, then Kublai, then Edgeryders 1, then Spot The Future, then OpenCare. We used Ning. We used Drupal. We used Wordpress. We used Discourse. We used Second Life, God have mercy on our souls. Identical results: a small minority gets very active and engaged. Everyone else shrugs and goes tight back to hang out with the cool kids. That's OK, as long as the small minority is, in absolute value, large enough. Which it was, every single time. This is what gets us here, as a coveted partner in H2020 projects.
In practice: Biofab is Discourse: so is Edgeryders. The UX is identical. The tradeoff we are facing is:
If we break the four debates across four completely separate platforms (different codebase, different database, no single sign-up), this is more appealing to the "ontologist in our head". Medieval history has a shelf in the library, business has another. This is a fallacy, because ontology is overrated, but whatever: it is how most people think at a superficial level. So the advantage is that it is initially more intuitive and easier to explain.
If we keep them close and "meshed" ( like different cats in the same Discourse instance: one signup, notifications from every topic that you contribute to, across all cats) we make moving from one debate to another easy, like wandering from one group of people to another at a party. We tried this in opencare, with significant spillover (explained above in this post). So the advantage is that, once people are aboard, it makes for a more diverse and richer conversation, with a stronger interdisciplinary stance.