Apologies if there’s already post on this, I spent time searching but couldn’t find anything.
Request: With some of the rules (like minimum_post_length), I wonder if it would be better for Discourse to tell users about the rule the first time they break it, but only enforce the rule (e.g. prevent them from posting) the second time? Personally, I think this would improve the UX, the vibes, the feelings about using Discourse. It would reduce the sense that, as a new user, Discourse often tells you off unexpectedly.
Context: I’ve loved Discourse ever since watching one of Jeff’s talks several years back, I recently re-watched it and there was a part where he said “People don’t like to be told what to do” or something like that. Having setup a Discourse forum recently and started testing etc. I was surprised at how often it did “tell me what to do”, even though I appreciated why it was doing that each time, it was sometimes fairly annoying.
Example: Having connected a Wordpress-based resource library to Discourse for discussion purposes, I was editing the Discourse description of one of these resources in its discussion thread. The resource in question had a short (but clear) title. When I went to save the new description, Discourse prevented me because the title was too short.
^ I now realise this isn’t a new user example, and the feature request wouldn’t actually solve this, maybe that’s more of an admin override request, anyway…
I’m not sure. I think this would create a lot of spam, because new users are instantly allowed to type “Hi” if the minimum characters is 10 or something.
I set my minimum_post_length to be at least 2 characters and STILL had complaints that people couldn’t do 1 character posts
To me, a behavior should either be allowed or not. I don’t see the benefit of a single “get out of jail free” card. At a certain point, a line has to be drawn. I guess it’s ultimately an optimization problem balancing the single user experience with the quality of discussion being produced
On the spam, I don’t think it would make much difference, it’s just foregoing the rule once, and many of the people would probably edit their message after seeing the prompt. I think it would just create less initial friction/frustration. Using an analog community metaphor: you’ve only just started and already someone is telling you you’ve done something wrong and you can’t do that, and the thing in question is fairly inconsequential, i.e. “Hi” vs. “Hello I’m new around here, this is cool”—not a massive difference in my opinion.
On complaints, I don’t know whether it would affect these, but the suggestion isn’t really about that. It’s about the experience that new users have and how they then feel about the software/forum. “Allowed or not” is quite binary thinking and not really how ‘real life’ communities work, from my experience at least. How users feel about the forum/community is ultimately a big part of the quality of the discussion, so I don’t think it’s a user vs. discussion quality calculation.
Overall, I think it’s more human to inform someone the first time (and not enforce), and then enforce the second time (because they have been informed).
Anyway, it’s just my thoughts from my understanding of UX and initial interactions with Discourse.
I completely agree with you on this, and for UX reasons, I think this should be between infringing message 1 & 2, rather than 0 & 1, giving the stakes are quite low and people’s feelings are important
Think about how this would look from the other side and the confusion it would create. Thousands of first time posts which don’t comply. Users asking why X can do something they can’t. Would those post need to be labeled as the “freebie”, adding to the noise?
Isn’t it more jarring to be told you can’t do something the second time around having just established that there’s no technical obstruction?
The whole principle here is that topics and replies to topics need to be meaningful. Short responses are absorbed by likes, reactions and votes. Notifications now are far improved because we can suppress pings from ephemera. Being notified by a topic because someone used their free post to +1 rather than throw a like on it totally undermines that.
I understand the principle (and the function of likes, reactions and votes) AND I’m suggesting a tweak to support better UX.
I don’t think it’s more jarring to be prevented the second time when you have be informed the first time.
This and your last point read like catastrophising. We’re talking about a proportion (probably a significant minority) of first posts from new users being less than the character limit. There’s an obvious explanation in the unlikely event that anyone complains, plus it can be made as an option: forgive_first_post_limit etc.
I think a character counter (like on post-voting comments) on first x posts could be a softer way to still enforce a minimum on all posts, but have some pre-warning that a minimum character limit applies before hitting the red bouncy clang warning. Maybe have that apply to users that come under the ‘new user’ extra restrictions.
