Proposal to change how TRS and Image Pickin' work

Yeah, this whole making a living thing can really cut into your day. It's still on the list for as soon as I can get some wiki programming time .

I'd be curious as to what actually is on the list for wiki improvements/fixes.

See the Wikitech Wishlist forum.

I would also like clarification on this. If there's going to be a thread cap and a requirement for mod approval, I think it's likely that TRS will basically be crippled.

The thread cap goes away.

Hm, what would be the reasons for rejecting a thread?

Repeating an old argument over and over. Spam threads. Claiming that a trope is misused without offering evidence, etc.

Also, in Image Pickin' there are are a lot of threads based on "I don't like it" or spurious "I don't get it" stuff, mainly by people who don't ever seem to like or get anything.

Right, so how does this work? Would it be like a closed mod thing where it's only the mods can see the queue? Would OPs get notified their thread was rejected? Would they be told why? I assume the mods are the only ones who can decide these things but I do appreciate some amount of openness in how the mods do what they do.

edited 3rd Feb '14 10:16:04 AM by treelo

The threads don't allow for posts until a moderator approves them. That is how it works in OTC.

If it works like OTC, then we could open a thread long enough to post about why it won't be approved before locking it.

For the record, in Image Pickin' no-open decisions would probably be contested in the IP policy topic. I proposed to have a TRS meta-discussion topic for TRS.

Several of the thread reasons in IP are going to be easy approvals, and for those that aren't, I'm sure that myself, Maddy, lu, and Telcontar will discuss them since we're the mods that are in IP most. For "Poor Image", whoever makes the thread is going to have to present a really good case for changing if the main gist is "this is a better pic" rather than actual issues in the current, and suggestions will pretty much have to be a requirement.

edited 3rd Feb '14 6:55:05 PM by Willbyr

I've been thinking...we should probably change the "Ambiguous Name" tag to "Misleading Name". Ambiguity in the title isn't a problem unless people are actually confused by it. Having it as a preset tag gives the impression that our standards are lower than they actually are. Just about any name could be interpreted in multiple ways; on its own, it's not valid grounds to drag a trope into TRS, or else we'd have renamed the entire Elemental Powers index by now.

Technically speaking, a lot of "misleading name" cases would be covered by "misused" (since misuse statistics are the usual method to demonstrate that a name is misleading). "Ambiguous Name" ought to cover the remaining stuff.

True, it's usually the misuse that's the problem, not the ambiguity. We're generally not concerned about ambiguous titles that don't also have another problem like misuse or underuse. Either way, it's not a very useful thread tag.

Wow. I got disillusioned with the TRS methodology back in mid-June.

Now I find that, in mid-October, so has the Boss, at least to an extent.

What I would've suggested would have been that each Troper could suggest a thread to repair, but only one at a time. When a "bay" opened up in the Shop, it'd be up to the mods to look at the oldest "ticket" in the queue and call the Troper in to comment on the proposal. No comment within 72 hours? Boop, gone!

So, yeah. I applaud the move towards moderator approval in TRS. I do have some thoughts on the rest of the thread, though. For instance, Post #111 had an "interesting" statement:

The problem with the statement is that we do need to get a concensus opinion on what the problem is and how to fix it if it exists. Otherwise, the shop would be getting the same issues every six months as those with their own solutions would see them as "just as valid" and propose them.

Tuefel does make a later point about advertizing, but there's already a banner that goes up when a page is addressed in the various fora. However, part of the problem is that sometimes the issue is not so much about one Trope as it is in the relationship between Tropes. And yet, we can only name one Trope as having the problem. Perhaps triggering a banner on Trope / Audience Reactions / Trivia pages Potholed in the individual topics (or at least the OP and/or any Crowners involved) would generate the necessary interest to proceed towards a solution.

One concern I see with requiring the OP to do a Wick Check is that said OP might not do a random wick check, instead presenting 50 examples of Misuse with no balance. In essence, Confirmation Bias. Now, if any page gets 50 improper related-to wicks, there's a problem whether the page has 90 or 90,000. Now, in the case of the 90,000, the problem might be remedied with a cleanup effort, but the fact that a problem exists would remain.

As a solution, if a TRS topic is proposed for "misuse" and the page that's accused of "misuse" has at least 400 wicks, then the attending Mod should send a "Wick Check Request" to a "Wick Check Request" thread for a third-party Troper to take up. If there's less than 400, then the Troper can do it themself; even if they do "cherry pick" and come out with a "100% misuse" report, that's still a confirmed 12.5% of the actual wicks misused.

Anyway, nice to see that we're heading towards true reforms and not just a "cap".

That's what experience and the archives are showing me right now. I used to think it would be possible, but trope families may be too nebulous a concept to work with when it gets to the nitty-gritty of "what is the problem" and "how to fix".

Okay, here's a theory (with some sandboxy elaboration): good threads tend to tackle certain questions in a certain order, and skipping too far ahead increases the odds of a thread turning bad. Do these observations seem accurate?

I'll note that providing a non-random wick check and Confirmation Bias are not the same thing. Really, randomness in wick checking makes a difference in only the few cases where a previous wick cleanup was attempted, and that is rare.

i think it would also be a good idea to mention that if you try coming into a TRS with a firm Step 4 in mind, you might be blinded going into step 2 and 3. let the discussion guide the flow a bit, as a wiki, we work on consensus.

edited 13th Feb '14 7:00:05 AM by crazysamaritan

Is "almost the entire example section is composed of Zero Context Examples" a TRS issue, and if so, what thread tag would it get and how would it be fixed (aside from "cut and re-YKTTW", which seems like a drastic move if there's no larger underlying issue like a premature launch)? How often does it happen and isn't a sign of a larger underlying issue anyway?

The rest of the example section of phase 1 covers the thread tags Misused, Not Thriving, Complaining, and possibly Unclear Description. Most of the remaining tags seem to presume the existence of a problem before the fact, and I wonder if your theory would imply getting rid of them. Not Tropeworthy and Duplicate Trope, though, and to some degree Really a Useful Note, describe problems that may not be evident from looking at the page itself (especially if you're not familiar with what a trope is supposed to be) but describe ways where the larger context would seem to preclude the existence of the page.

edited 13th Feb '14 9:57:14 PM by MorganWick

Personally, I would not put "entire example section is made of ZCE" in the repair shop. Example collection is not what TRS is good at, as we've seen. "Underlying problem" would be a suitable reason, though.

As for occurrence, I know that certain tropes (appearance, for examples) tend to lean towards ZCE. It's a complex function of trope name and trope concept, sometimes it's a sign of an underlying problem, sometimes it's not.

I just hopped into the discussion so I'm not sure if this is the way things should be and TRS has just been allowed to fill up due to it needing a revamp, but I think once a change has been done as suggested in TRS, the thread really needs to be closed/removed so more threads can come in. If there was a way to automate it, that would be wonderful!

For example: there's an open thread about renaming Spoony Bard. When Spoony Bard was renamed (maybe through a button on the page) the system could auto-close the thread. Alternatively, people could vote to close a thread sort of like YKTTW. This should help the threads self-moderate a little.

Another thing that might help TRS is having a few checkboxes on the post page that match reasons why a page is being talked about in TRS. To make a thread you need to check these boxes. If none of the above apply, it probably shouldn't be in TRS which means that moderators will need to approve it. This should help auto-moderate out some of the threads that aren't very useful by keeping people on track while reminding people what TRS is for.

Example checkboxes (based on FAQ): Needs better definition, Miscategorized Useful Note, Page Mostly Complaining, Not a Trope, Duplicate Trope, Lack of Examples/Inbound Wicks, Misused Trope: Examples do not match, Needs a Rename, Snowclone, Other (this one gets moderator attention!)

edited 14th Feb '14 3:53:01 PM by lamoxlamae

Renames are not as simple as simply renaming the page; all the links pointing to the page need to be changed as well, which makes it harder to automate closing the thread. Threads are closed when everything that needs to be done is done; show me a thread where absolutely nothing else needs to be done, and I'll show you a thread the moderators just haven't gotten around to closing yet or no one has bothered to holler for a lock for yet. There's been some consideration of moving pending-final-action threads (where the work to be done isn't super-simple) to Short Term Projects, but some people are worried the work won't actually get done that way.

Most of what we're talking about isn't so much people bringing the wrong problems into TRS, but people bringing what they think are the right problems but insufficient justification that they're actually problems. As I hinted above, I'm not sure "needs better definition", "needs a rename", and "snowclone" are problems in themselves if they're not causing actual problems, and some people may cry "misuse" or "duplicate trope" based on a misunderstanding of the trope description or, in the former case, purely anecdotal evidence.

The more I think about this, the more I wonder if we should get rid of Unclear Description, Ambiguous Name, and Snowclone as thread tags. All three presume the existence of a problem without implying the need to establish one, and it's hard to tell what actual problems they would cover that aren't covered by other tags, usually "misused". If we do that we might also want to change "misused" to something like "example mismatch".

edited 14th Feb '14 5:01:17 PM by MorganWick