The forum life

HardwareGeek

@GuyWhoKilledBear said in The forum life:

I can't imagine ChatGPT coming up with a useful summary (i.e. one that actually contains the right answer) of a topic in a tech support forum.

It might be able to come up with something like this:

The accepted answer is X, but several users pointed out that it is wrong because Y. The highest-rated answer is Z. There are N other answers, including ...

Perhaps, in addition to summarizing the topic, it could compare with other sources of information to synthesize its own preferred answer (which, of course, you'd need to check carefully for correctness and applicability, because misinformation).

I also can't imagine it coming up with a useful summary of one of our megathreads.

Yeah, our megathreads don't stay on a narrow enough range of topics for a summary, whether generated by AI or a human, to be useful. About the best that is possible would be something like, "Tens of thousands of posts on a wide range of subjects. Some of the most-often mentioned topics are .... Most have some humorous content."

dkf

@HardwareGeek said in The forum life:

Perhaps, in addition to summarizing the topic, it could compare with other sources of information to synthesize its own preferred answer (which, of course, you'd need to check carefully for correctness and applicability, because misinformation).

We need an official looking stamp that proudly says "AI Certified Bullshit" which we can attach to such summaries so that it's really obvious that they shouldn't be trusted.

Arantor

So I said I'd comment on some of the other insights about the forum software world. I hang out in various places - on forums that discuss forums, if you can imagine it - and I have a few things to share, though this is not particularly curated so apologies in advance as it's likely to ramble a little as I try to articulate everything I've heard, seen and thought about lately.

The classic forum platforms are for the most part 'I am a forum and if you want anything else, there's some kind of addon', with a few that have first-party addons for other kinds of content, e.g. forum + gallery + blogs + other content or similar.

Now, one of the ongoing wishlist items/arguments is 'we need better tools for things that aren't forum discussions', because the belief (not entirely unfairly, also supported by reasonable data) is that what brings people to the community is something in the form of an article or other content around which a community came to be. (In TDWTF's case, this is of course the mythical front page.)

To me this represents a very interesting split between the groups who argue for/against; there are certainly those who argue that 'it should just be a forum', and that articles etc. should be left to a proper CMS, and there are those who argue that the forum should be more of a community hub and include some content management functionality.

Then you look bigger picture at these platforms that are a forum first and a gallery/blog/download centre/helpdesk/e-store/whatever and one thing really smacks you in the face if you ever spend time looking: they're not really cohesive.

They're separate things bolted toegther. Yes, you can have a forum with the same theme as the gallery, with the same theme as your downloads area and your staff blog or whatever else, but the content being siloed up like that (historically to sell plugins)

This is certainly something I've spent a lot of time pondering: when we talk about 'building a community space' are we talking about just dropping in a forum as a vehicle for talking or is it genuinely something broader than that?

I know a big recurring theme that came up in this thread is how central 'letting people actually do the talking thing' really is, but there's also a place in a community space for being able to share things that aren't just discussions. Writing communities want places to be able to share and discuss stories; photographic communities to share their images and so on.

Which brings me to the next bone of contention: forum categories.

We have our nice list of categories - something that the old forum guard still insists should be the front page of the forum. That showing people your taxonomy and hierarchy should be the opening gambit for content. There I have to give it to Atwood, for being ballsy enough to break with that tradition by default, though I'm not sure he did the right thing with it.

What I do know is this:

people do rely on the categorisation for being able to find topics that are thematically related
people who are already familiar with a site will have some idea where to go to place content to reach a certain audience within that site, or to put a particular slant on it (e.g. here, my posting this topic in General Help rather than in General)
people who arrive at a site inevitably do not arrive at the list of categories early in their journey and it is not so useful to first-time viewers - other industries understand this; your average e-commerce venue doesn't have a vomited-up list of all product categories of products on ther front page (it'll be curated, with calls to action, funnelling the users towards particular sections)
it is very common for users to join a site to ask a question and be confused where to post that question, especially if the taxonomy is not very clear; this is often met with hostility on the part of the community, and especially if the community skews older with a younger new member
the categorisation is often seen as confusing and unnecessary for 'younger members' (often a thinly disguised bashing Gen Z opportunity) when they're used to posting on social media where such things are not presented as barriers to entry
Discord exhibits much of the same behaviour with channels as categorisation and yet 'Gen Z' don't have the same problem, it's almost as if the problem is actually bad taxonomy and disinclination to improve

I've often wondered if the way to go is actually to not have boards per se but instead treat it as those a given topic can be attached to multiple categories where it might actually fit.

I mean, this isn't without all of its own problems; what if for example a topic were dual-tagged at creation for both the Lounge and General? Is it more logical that this should be visible to all (because General) or restricted (because Lounge)?

I mean, you get the same general arguments/concerns about moving topics between categories as you would re-tagging them (a mod moving something from the Lounge to General is in the same vein as someone retagging a topic from 'Lounge' to 'Lounge, General'), but I don't think this is something that is necessarily a blocker.

I think the biggest part of my reticence about just pushing for this model full-bore is because categories have permissions by default, not just around who can see them, but around who can moderate, and setting this up for more than a few people in a few areas is a massive ballache to do, which is why we end up with the boards-as-a-bucket delineation, and the boundary for permissions.

I've seen the other model, where users can create 'clubs' for their own little areas on a site - for the metaphor, I suppose... imagine if the Garage were its own little area not moderated by the site moderators, but by 'the Garage moderators' who have literally no other moderation powers outside the Garage (the typical model posits that general moderators would not have power there either, other than admins, and that topics being moved to/fro is... sketchy at best).

I dunno. I find myself standing at the weird cross-roads. On the one hand is the old guard - the folks who passionately and violently defend forums as they are - and the new guard who seem to have almost comprehensively rejected forums.

I continue to view this as less about 'the forum model having its time' and more the stubbornness of not evolving and broadening its horizons... it feels to me like the next evolution is somewhere between the classical format, and social media (WAIT DON'T RUN AWAY) where we find a balance between being able to have a voice without jumping through unnecessary hoops, and being able to find that content again through a platform that does not eternally view 'now' as the central most important thing.

I suppose my question here is not even whether this makes any kind of sense, but... is it worth fighting for to try to pursue this avenue? Is it worth it, in 2023, to pursue building a tool to build communities that isn't fundamentally rooted in 2000s internet culture?

(Please note I don't think I'm going full Atwood here. There's a lot of value in not trying to change how people discuss, because that's not the bit that's broken. It's all the fluff around that that's the problem. It's the categorisation model, the presentation model of topics of interest, adjacent topics, related topics, finding more content beyond the content you already have.)