Given recent renewed interest in creating segregated “rooms”, and the recent suggestion that Discourse’s categories be used to implement this, I tried to do some digging on what Discourse categories can, and cannot do.
What it appears categories can do
- It seems possible to create as many categories as you like, and even a category for every topic (rendering ‘tags’ increasingly redundant).
What it appears categories cannot do
- Segregate who can see and respond to a category, other than by Trust Level.
Non-members cannot see ‘Side Conversations’, and it seems it is possible to create a category that new members (Trust Level 0) cannot see, etc.
Yes, it would be possible to create a new category called ‘ID Conversations’, for example, and make that category invisible to non-members – but it would then be indistinguishable, except by name, from the existing ‘Side Conversation’ category – and would still be visible to all members. The only difference it would make is the increased work of implementing it and maintaining it.
It would of course be possible to assign a higher minimum trust level to the new category, but that would (i) require ensuring that all those interested in that category’s topic (e.g. ID) were given at least that trust level, and (ii) mean that all other members of that trust level and above would still see threads in this category, whether they were interested or not.
It is therefore hard to see what practical purpose creating new, topic-specific, categories would serve.
(Of course, I am not an expert on Discourse software, so if I’ve made any errors, or important omissions, please point them out.)