Hi Glenn! Thanks for the feedback.
In short: you're right, but I think you were saying your approach on its own would work better. If so, I think instead a balance between what this forum does and your suggestions is what's wanted. I do personally like the bot/tutorial idea (and also think something like that would work), and like the more-threads/in-thread-topics approach to a degree.
More dets. below.
I agree to a large extent, except I think you're missing one key thing: you'd have too many threads for one-off problems with your approach.
Like Stackoverflow does well, and like this forum does decently well, we want the common problems and the best answers to stand out, and not get lost in the noise.
A balance is key; it's not that what you said isn't important, but it's not the only important thing. Sorry if that wasn't your intent, as I may've misunderstood. Equal parts encouragement to participate and the ability for others to find what's most likely important to them (based on more users than not having the same questions and benefiting from the same answers) is important.
I respond more directly below and provide some further details.
There are problems with that approach, such as assuming that all people will have the same types of problems,
From experience, it is a good assumption. Ideally, not to the exclusion of others; but one-off questions and answers don't benefit tons from a forum, so not drawing attention to them at the expense of more popular topics, but ensuring those asking feel welcome to ask anything is incredibly tough to do.
And, if you let topics organically become popular, you get a feel for what the common issues are. As is done, all topics start out as "one-offs", but if you get many other posts in said topic or about the same or similar question, that's a sign it needs to become a more official topic.
But, that balanced approach has to be the goal.
ThreadTopic decay occurs for what you listed.
Agreed. Just like what you mentioned isn't everything, what this forum does well isn't everything either. It does tend to exclude, I think, concentrating too much on the more common issues, and I'm more certain can overwhelm (a potential solution to that is mentioned below).
Again, balance by far is the key.
The disadvantage to that is having to wade through a lot of different messages to find what you're looking for.
One thing that's not obvious is how to best use the forum to prevent such, let's call it, "topic-wading" . And, I personally really like the idea of having the bot cover this. Although apparently no one really likes the bot, as few use it in its entirety. I thought it was great, but I think it further was something that other forum admins liked too, but also fell flat in terms of actual efficacy (only vague-recollections-of-outdated-posts to "support" this).
To grab the focus back, having an intro. tutorial on how to use the forum is something I would improve upon going forward. In this case how to efficiently search:
- Check the standard course topics.
- Do a search. The search is pretty good.
- Once in a topic, especially a long one, you can do an in-topic search
That last one is probably the one many don't do as it's probably not apparent. No need to wade through topics normally, just check that box off, and you get the equivalent of an in-browser search, as if infinite scrolling wasn't on (that's another thing I'm personally not a big fan of, but I don't think that's avoidable).
Another one that is awesome, but in my opinion only appear on topics once they've become too long, is
From experience, having a Stackoverflow-like ability similar to what that Summary functionality gives , as our forum is primarily questions and answers, would be ideal (I still prefer what Stackoverflow does; having each topic exactly mimic a Stackoverflow page but on our site would be ideal, imo). The above doesn't kick in until the topics get really big, though.
A solution to the problem of disparate issues within a challenge would be to
Yeah, agreed that's great, but that does take too much time to do as a general approach (see below, as I have done this, and yeah, it's a pain to keep it up, which as an overall approach does defeat the purpose. Of course, given more resources, this changes things!). Even so, not a great excuse on my part, as I haven't tried it much, and could've used it more. This is where I genuinely don't know how to find a good balance. 100s of threads, or in-topic-links, where no one can find any commonality or dozens of threads with one or two large ones. Neither is ideal, and a balance--yup, there's that word again!--is for sure needed.
I'm not sure how to best solve this going forward, but the criticism here is well taken. I think we went too far one way, and would need to bring it back to centre, for sure.
Note those in-topic search and summary functionality features do help. When used they make a huge topic more manageable. e.g., a different approach, and if each post about "Problem with parse" had that phrase in it, voila, same difference.
Tough to enforce or maintain either way, and that's a problem I have yet to figure out how to tackle well. I think probably the current approach, with a decent search as exists now, but along with 1) more instructions on how to properly use the forum 2) a guideline and some enforcement of a per-post subject (i.e., first line is your subject within 1.6) would strike a balance nicely.
Even though it was/is tough to maintain, the multiple links idea is handy, no argument there
I've used it rarely, and it'd be best if this was more software supported, or, like my idea above, a compromise that effects almost the same thing.
And, for what it's worth, yeah, I prefer 1.6 type challenges too The current system was in place before I got started, works ok, and I think was done when older versions of this forum had a search that wasn't as good.
and not offering default threads, discourages questions,
I'm not certain if it caused a huge issue, but admittedly, if it did, for reasons you mentioned, I wouldn't for sure know.
My feeling is that it didn't. I took the only-make-a-sidewalk-when-people-carve-a-path-in-the-field approach. Don't waste valuable time, which has an impact on users if you are spending that time which could better be used helping users, making sidewalks no one will use.
Having said that, it wouldn't be too hard to maintain, and I just went with my approach because many topics early in a course would quickly become irrelevant, or obviously one-offs. I should've at least experimented with the other approach, and when a topic became outdated, close it, but make a new empty default topic for that challenge.
And, for transparency, I think the intent with the topic lists, which was a great idea and not mine, was to have one for all or most challenges, as you suggested.
I rambled in a lot of places, have a lot of in-line asides, etc., so if anything is unclear, lemme know.
And in general, if you or anyone have any follow-up comments, please feel free to dive in and share!