23
\$\begingroup\$

Our list of forums gets used in comments quite frequently, including by yours truly --- however, while I like describing it as "our curated list of forums", many of the answers there haven't been updated in ages. This means that statements pertaining to various features like activity, moderation culture and so on may be outdated. Upvotes implicitly recommending online communities can likewise also be based on old information and might not accurately reflect our community's current assessment of those communities.

Since the old forum list is starting to show its age, I think we should "reboot" it --- deprecate it, and start a new one from a clean slate. Information that is known by an active user to be still valid from the old forum list can of course be salvaged to the new one. Not having old votes as historical baggage and the requirement to actually revisit any statements made about the forums would make me feel more honest about referring to the list as a "curated" one.

Addressing comments

nitsua60 asked in the comments about the alternative of simply curating our current forum list:

Not knowing quite where you're coming from makes me wonder about the utility of this approach as opposed to, say, featuring the post and asking people to take the summer to make a concerted community push to update, comment, evaluate, reconsider votes, &c.

I think there are several advantages to deprecating and replacing the forum post over repairing it:

  • People seldom revisit their votes when an answer (or the subject of the answer) changes. Votes are therefore not a reliable assessment of the quality of the answers (or the forums they concern). Essentially the votes say nothing apart from "someone thinks this answer was good at some point of time", which is still worth something for a post that has existed for a week and whose subject matter is unlikely to have undergone a drastic change, much less so for posts that have existed for years. Starting voting from scratch in a new post would handle this automatically without the need of extra effort from the community.
  • The community effort to revisit voting and content is hard to enforce and requires extra oversight and bookkeeping to ensure all answers are indeed reviewed. By my suggestion, no particular oversight is needed. New and salvaged answers both go through the normal quality checks, without having to assign someone the ungrateful task of vetting existing posts and trying to contact people who could verify their content. Content no one can verify will be pruned emergently, exactly as it should be.
  • Periodic new questions is how we handle our own site's "How is the community doing?" reviews and community ad choices. It has worked well for both cases. Those posts concern a similar subject matter and have similar difficulties to the forum list --- they have to stay in flux because the communities concerned are always in flux. A permanent Q&A with enforced curation wouldn't add anything there either, except difficulty.

There is only one non-trivial disadvantage I can think of:

  • Some people'll have to update their pre-made comments.

Furthermore, old comments pointing people to the forum list post will of course point to the deprecated post, but we can perhaps add a signpost link to the most up to date forum thread as needed and close it as a duplicate of the new post.

Summa summarum, I believe trying to keep the current forum list up to date and reliable requires lots of effort and still produces a worse outcome than just periodically starting over from scratch. Let's deprecate it and start anew.

\$\endgroup\$
8
  • 3
    \$\begingroup\$ When you say "show its age," are you saying that just because it's been inactive for a while, or are there entries in there that you noticed and facepalmed, saying "whoa, that forum's been stone-cold for five years now, people. You've got to go to $newplace these days." Not knowing quite where you're coming from makes me wonder about the utility of this approach as opposed to, say, featuring the post and asking people to take the summer to make a concerted community push to update, comment, evaluate, reconsider votes, &c. I'm of two minds. Or maybe zero. In any case, good to bring up, thanks! \$\endgroup\$
    – nitsua60 Mod
    Commented Jul 26, 2019 at 19:37
  • 1
    \$\begingroup\$ @nitsua60 I do not frequent most of these forums, and therefore cannot know the truthiness of any claims made of them. Without active updates, it is impossible to know whether an item on that list has been verified and when. I think creating a new post (periodically, if needed) is advantageous over trying to "repair" the old one: the voting baggage gives undue attention to old answers and the effort needed to create a new forum list is marginal. Do you see any significant benefit in fixing the old post, as opposed to just creating a new one? \$\endgroup\$
    – kviiri
    Commented Jul 26, 2019 at 19:46
  • \$\begingroup\$ Furthermore, the clean slate approach is much easier to manage as it doesn't require any oversight to check that all items are actually updated. \$\endgroup\$
    – kviiri
    Commented Jul 26, 2019 at 19:48
  • \$\begingroup\$ I think I'm in a similar position to you: with the exception of Role-playing Games Chat I'm essentially non-exposed to any of them. So I have no basis for evaluating how much the "old" votes are good information we shouldn't throw away vs. moldy strawberries sitting in the back of the fridge. I'll be interested to see what the rest of the community says =) \$\endgroup\$
    – nitsua60 Mod
    Commented Jul 26, 2019 at 19:49
  • 3
    \$\begingroup\$ It is a sad side-effect of our voting system that votes are timeless (I've probably upvoted stuff that I would disapprove of now) as well as independent of the actual content (an upvote given to one of those forums will remain there even if the forum or the answer's treatment of it changes completely) :( \$\endgroup\$
    – kviiri
    Commented Jul 26, 2019 at 19:54
  • \$\begingroup\$ Here that's sort of combining with the slow pace (of eyes and votes) on meta to exacerbate, I suppose, the possibility of some advice becoming woefully out of date/utility. Pondering. \$\endgroup\$
    – nitsua60 Mod
    Commented Jul 26, 2019 at 19:56
  • \$\begingroup\$ @nitsua60 I collated justifications for my idea in the post proper. \$\endgroup\$
    – kviiri
    Commented Jul 27, 2019 at 9:52
  • 1
    \$\begingroup\$ Cool, thanks. I think your second-to-last paragraph's concern could be alleviated if the "old" forum list were closed as a duplicate of the "new" one, should that come to pass. \$\endgroup\$
    – nitsua60 Mod
    Commented Jul 27, 2019 at 13:34

4 Answers 4

22
\$\begingroup\$

Curate, update, and consolidate existing list into one article

What I always intended for that list was to consolidate all of what it had collected into a single large post with a more natural ordering. The votes were always intended more as a peer check that a given site mattered, was worth including, and was well described.

I feel the votes have outlived their usefulness though, yes, and combining them all into one single CW post would solve the votes’ staleness.

Once it’s a Community Wiki, the descriptions can be updated. The vast majority will be unchanged — forum cultures and activity for the major ones generally doesn’t change that fast at all. That leaves the smaller ones that need checking, and those are less important to keep super accurate. But as a CW, we can iteratively improve the list.

For usability, too, a single post would be better. We can group it by major general sites, then specific topic sites, then niche sites. It would be easier to read a well-organised list, and wouldn’t take much on our part to organise.

That would also keep existing links to our list accurate, and not add a duplicate notice that will only get in the way of new people unfamiliar with our site—its main audience.

We should combine them all into one post, delete the individual posts, and update the descriptions as necessary at our leisure, in that order.

\$\endgroup\$
8
  • 1
    \$\begingroup\$ Aha, another use for a CW. Sounds like a good idea to me. \$\endgroup\$ Commented Jul 28, 2019 at 23:10
  • 3
    \$\begingroup\$ Yup. The worst thing about the existing list is definitely the lack of categorization -- what's the D&D stuff, what's other-game-specific stuff, what's "general," &c. This would fix that up nicely. \$\endgroup\$
    – Alex P
    Commented Aug 1, 2019 at 12:24
  • \$\begingroup\$ Any chance you'd have an hour at some point to devote to that project? The idea seems to have strong community support, and you certainly have a clear vision of what could result =) \$\endgroup\$
    – nitsua60 Mod
    Commented Aug 6, 2019 at 16:30
  • 1
    \$\begingroup\$ @nitsua60 Later today, if someone can start it! As a wrinkle, the question is my own, so an answer started by me couldn’t be pinned to the top, I think even as a CW. \$\endgroup\$ Commented Aug 6, 2019 at 18:11
  • 5
    \$\begingroup\$ Oh, good point. I'll instantiate an answer that you can do 99% of all the work on, and reap all that good-good meta karma for myself =) \$\endgroup\$
    – nitsua60 Mod
    Commented Aug 6, 2019 at 18:42
  • 2
    \$\begingroup\$ @nitsua60 First attempt at consolidation done! \$\endgroup\$ Commented Aug 7, 2019 at 23:19
  • \$\begingroup\$ So... a question that totally should have occurred to me two days ago: if someone wants to add a new entry to the list, what do they do? Just up and edit it in to the big list, or pose it as another answer which can get folded in when it's got some love from the community? \$\endgroup\$
    – nitsua60 Mod
    Commented Aug 8, 2019 at 2:23
  • \$\begingroup\$ @nitsua60 Good point. We haven’t had much experience using CW for this kind of thing… I think, after considering, that either would work? We should just pick one. I imagine, if the question is left as-is, people might default to adding new answers. Maybe that could be made more explicit. \$\endgroup\$ Commented Aug 8, 2019 at 7:34
9
\$\begingroup\$

Deprecate the forum list and start a new one

Arguments for this course of action are presented in the question, I won't repeat them here. Here are the concrete steps I would take:

  • Create a new forum list post
  • Insert a deprecation notice in the old forum list and close it as a duplicate of the new one
  • Repopulate with new answers as well as answers from the old one that are verifiably true
  • Repeat periodically (eg. once per year) to maintain a reasonable level of freshness in recommendations
\$\endgroup\$
3
  • 2
    \$\begingroup\$ I've got to admit: this intrigues me a bit. From the post I wasn't seeing much utility in "throw out the old list, make a new list exactly the same way." But here you're not actually suggesting that, you're saying "throw out the old process (make a list and trust updates to keep it evergreen) and replace it (with make a list every year)." There are implications on drawing people toward meta that catch my eye here, though I'm still worried about throwing away the wisdom of ages. \$\endgroup\$
    – nitsua60 Mod
    Commented Jul 27, 2019 at 13:50
  • \$\begingroup\$ @nitsua60 It wouldn't as much be thrown away as archived, similar to old community ads or "how is the community doing" chats :) \$\endgroup\$
    – kviiri
    Commented Jul 27, 2019 at 13:55
  • \$\begingroup\$ good point--that would have been a better way to describe it. Thank you =) \$\endgroup\$
    – nitsua60 Mod
    Commented Jul 27, 2019 at 14:13
3
\$\begingroup\$

Update the current list

If the problem is that the list needs new input, then those who have new information ought to post it. Also, those who have new sites and recommended forums/reddits ought to post them there. It has served as a decent kind of FAQ and is worth leaving as a touch point.

How? Put that Q in Hot Meta Posts. Leave it there for a month or two.

If a diamond mod can manually insert a meta question onto Hot Meta Posts, as they can now do at SO, then do that to attract attention to that question.

My two cents.

\$\endgroup\$
10
  • 1
    \$\begingroup\$ The problem is not (just) the lack of new input, but the fact that the current list is not demonstrably up to date, and the voting no longer demonstrably represents our community's assessment of the answers/forums in question. What advantage does keeping the current list instead of deprecating it and starting a new one offer? \$\endgroup\$
    – kviiri
    Commented Jul 27, 2019 at 12:46
  • \$\begingroup\$ (Note: "not demonstrably up to date" doesn't mean "demonstrably not up to date", but neither is desirable for the forum list since it is supposed to be "curated", not "correct by happenstance") \$\endgroup\$
    – kviiri
    Commented Jul 27, 2019 at 12:50
  • 3
    \$\begingroup\$ @kviiri why not simply comment on issues you think are present with each answer on the list so that the original author can review their answer and update accordingly. From a cursory review, all those individuals are still active on RPG.SE \$\endgroup\$ Commented Jul 27, 2019 at 13:01
  • 1
    \$\begingroup\$ @Pyrotechnical That would necessitate me babysitting people, and wouldn't solve the problem of moldy old votes. I am also not an expert in determining the correctness of those answers, since I don't frequent most of those forums. Can I also repeat the question: what advantage do you see in clinging to the old post instead of starting over? \$\endgroup\$
    – kviiri
    Commented Jul 27, 2019 at 13:11
  • \$\begingroup\$ @kviiri I agree with you, but to answer your question. The big advantage to using the old post is there already exist links to it, granted though that problem is somewhat solved when we close it as a duplicate of the new one. \$\endgroup\$
    – willuwontu
    Commented Jul 27, 2019 at 13:27
  • 1
    \$\begingroup\$ @williamporter Links in comments are temporary, though, and links on Meta can be easily changed. And yeah, the deprecation notice & duplicate closure can also point towards the new iteration to allow finding the updated version :) \$\endgroup\$
    – kviiri
    Commented Jul 27, 2019 at 13:29
  • 1
    \$\begingroup\$ @williamporter maybe it's just because I'm a Catholic, but "we'd be throwing away the wisdom of our forebears" strikes me as more of a concern than "it could break some links." Call me a small-c-conservative: until I'm convinced something is bad and must be changed, I tend to trust the judgment and experience of those who were here before me. (But things obviously evolve quickly online, its perfectly possible that the list is stale, and I'm glad to see the conversation happening.) \$\endgroup\$
    – nitsua60 Mod
    Commented Jul 27, 2019 at 13:40
  • \$\begingroup\$ @Korvin any thoughts on how to ensure this updating is happening? It seems to me that kviiri's core contention is that this is already the de facto solution, and that the necessary updating (of recommendation and of votes) isn't happening. \$\endgroup\$
    – nitsua60 Mod
    Commented Jul 27, 2019 at 13:43
  • \$\begingroup\$ @nitsua60 Given how few people participate on meta as a percentage of the site users, not rally. I suggest that if Mods can put a q into the 'hot on meta' the way that SO can, then Put That Q&A into the "Hot Meta Questions" \$\endgroup\$ Commented Jul 27, 2019 at 13:45
  • 1
    \$\begingroup\$ Yeah, we can feature something so that it shows in the (right) sidebar on main--that's a technical possibility. \$\endgroup\$
    – nitsua60 Mod
    Commented Jul 27, 2019 at 14:38
-3
\$\begingroup\$

Delete all answers from the old one and start again.

  • Any answers that are deleted can be flagged for undeletion by their poster. This lets dead answers stay dead, while any that are still relevant can come back to life.

  • New answers are able to be posted on the old post still.

\$\endgroup\$
2
  • \$\begingroup\$ Not a bad idea at all, although in the interest of resetting voting I would still prefer my solution. \$\endgroup\$
    – kviiri
    Commented Jul 27, 2019 at 13:42
  • 2
    \$\begingroup\$ @kviiri I do too, this just seemed like an option that people may want as well. \$\endgroup\$
    – willuwontu
    Commented Jul 27, 2019 at 13:43

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .