[Discussion] - Raising Tag Consideration Minimums

Summary:

Raising article requirement for tags to be considered by Tech Team

[Discussion] - Raising Tag Consideration Minimums

Rounderhouse 01 Sep 2021 00:06

Mirror of : http://05command.wikidot.com/forum/t-14184007/raising-tag-consideration-minimums#post-5071971

babby's first tech proposal — posted with permission from aismallard and stormbreath.

Problem

Currently, the metrics for attribute/character/GoI/canon tag requests to be considered by tech team is the 5x3 rule, i.e a minimum five articles by three authors. That's been the historical precedent for quite some time, at least three tech captains ago. It ensured that only concepts notable enough for navigation/organizational purposes received a tag, because getting to that bar was kind of a big deal when you get like 10 articles a week. object classes face a similar issue - esoteric object classes are more popular than ever, and the informal metric for an object class tag was 15x3.

But with the overall increase of articles being posted to the wiki by an increasingly diverse authorbase, it is easier and easier for a concept to reach that bar. This means that concepts that don't really warrant a tag for navigational purposes end up getting it.

Solution & Implementation

Going forward, the requirements for a attribute/character/GoI/canon tag request to be considered will be 10x5 — a minimum of ten articles posted by five authors. We'll also codify the requirement for an object class tag proposal as 25 articles, author count irrelevant. We think this is a reasonable bar to scale tag navigation with the wiki's current activity level. Posts currently in the tag request thread that have not yet been addressed will be judged on the new metrics, but this only extends to two posts (one of which more than fulfills the 10x5 req anyway).

Potential Issues

I don't really foresee any technical implementation issues arising from this. We do have a number of tags that would no longer meet the minimum requirements were they raised - as such, all current tags that do not fulfill the 10x5 or 25 rule will be grandfathered in for now.

---

This discussion is open to everyone!

Last edited on 01 Sep 2021 00:13 by Rounderhouse Show more

Re: [Discussion] - Raising Tag Consideration Minimums

stormfallen 01 Sep 2021 00:52

I'd recommend having at least a three author minimum for object class tags, if only to prevent very unlikely shenanigans.

Also, I think we're already at the point where 10x5 feels low; there's far too many attribute tags as-is. I say make it 15x5.

Re: [Discussion] - Raising Tag Consideration Minimums

stormbreath 01 Sep 2021 01:17

I'd recommend having at least a three author minimum for object class tags, if only to prevent very unlikely shenanigans.

We discussed this in Tech Chat, and ultimately it came down to the following the idea that if somebody writes 25 SCPs all with the same object class that all manage to stick, that passes a level of dedication that it doesn't feel like shenanigans anymore: at that point, it can really only be legitimate effort. If you were really doing shenanigans, you could easily get a few more authors to satisfy that.

---

can I get some nice "no signatures on my forum"
sigma-9 css machine broke
understandable have a nice day

Re: [Discussion] - Raising Tag Consideration Minimums

Kothardarastrix 01 Sep 2021 01:27

+ show block

I don’t quite grasp the logic of increasing the minimum to account for more content. The importance of a tag should be based on the absolute number of articles/authors, not relative to the size of the site as a whole. If there were a new group of interest or canon to finally meet the tag requirement (say, the Red Actors Troupe or LV-Zero Scenario) I’d still want for there to be a tag so I can find the other articles, even if there are only five and that’s just a tiny chunk of the site. I say this especially because, without a tag, the only other reliable way to find more stuff about that same group is by checking backlinks. If somebody out there wrote another Red Actors Troupe article but it wasn’t crosslinked to any existing ones and it didn’t have a “red-actors-troupe” tag, I might never know about it until I wrote another RAT article that conflicted with that one and somebody else who had, by, chance, encountered both told me that in the comments. I have actually run into this problem twice already - once during my 039 rewrite, where I discovered another RAT article halfway through the process (though it fortunately didn’t derail me that time) and again while writing my 6000 contest entry, where I discovered two more tales set in the LV-Zero canon and had to adjust what I initially planned to write to account for it. Tags prevent this kind of thing from happening. If I want to write an article about the class of '76 or Researcher Talloran, I can very easily access all related materials through the tag, and proceed with my writing knowing that there isn’t some crucial bit of information that I’ve missed.

For another example, take the Department of Abnormalities. I understand that the lack of a tag for it is thematically appropriate for the DoA’s unique mojo, but it still illustrates my point. Without a tag, the only reliable way to find more DoA articles that aren’t directly linked to by SCP-3790 is by checking the backlinks, something that many readers (myself formerly included) wouldn’t even think to do/know is possible. With an increased minimum, most new GoIs and characters (assuming that they, unlike the DoA, aren’t supposed to be hard to find) will suffer this problem, making it difficult for readers to find more of what they like.

Furthermore, an increased tag requirement will make it harder for new GoIs/characters/canons to get off the ground. It’ll be harder for people to find the articles, and therefore harder for them to get votes or attract people who just want to contribute to the GoI or canon.

That said, attribute tags could probably stand to be raised. They’re not as useful or as easy to assign as the other kinds of tags, and they potentially apply to FAR more articles. So I wouldn’t object to the threshold being increased for attributes, and probably not for object classes either.

GoI, character, and canon tags are important, and they do a lot to both make an author’s job easier and to connect readers to more stuff that they might like. I think that raising the minimum requirement interferes with that and is therefore a bad idea.

Last edited on 01 Sep 2021 19:05 by Kothardarastrix Show more

Re: [Discussion] - Raising Tag Consideration Minimums

Ethagon 01 Sep 2021 12:50

These are pretty much my thoughts as well. I don't think this improves the tagging system in a good way (at least for GoI, character, and canon tags, I can understand the logic for raising object and attribute tags)

A few minor things I want to add, as Kothardarastrix already pointed out everything major:

• This would have the weird result of GOIs and canons being allowed to have a hub (5x3 rule) but not a tag (10x3)
• In case this is a workload problem, from what I've seen GoI, character, and canon tags are from what I've seen very quick to tag and what costs more time are the attribute tags
• This is also fairly minor, but the more articles are required the more likely it is that the person making the tag suggestion will miss an article

I don't really see why this is necessary or how it helps for GoI, character, and canon tags.

Re: [Discussion] - Raising Tag Consideration Minimums

Sirslash47 01 Sep 2021 05:51

Open (some proabable poor language used due to being heated)

Heavily agree with Kothardarastrix. Especially the "The importance of a tag should be based on the absolute number of articles/authors, not relative to the size of the site as a whole." since the site is always growing, do you want to keep updating stuff after more growth as well? We already have high barriers of entry for new authors no need to discourage them even more.

As far as the containment class requirement goes, no issue but for the GoI and character tags I absolutely disagree. Some of them are already really hard to find and deserve more eyes on them. Hell, didn't you guys ask "how can we get more eyes on other new/short articles" a while ago and put "shortest SCP's in the last 30 days" to encourage more reading? Then why do a reverse motion here?

If the problem is having too many tags to check, fix the attribution list and ask authors to maybe include character tags people should know about in an author post or let them do it themselves after tag team did their thing. I know multiple authors that would love this to happen, myself included. Please, just ask them!

One thing that also was brought up is the authors (and new authors) that don't easily can reach out for other people to write for them. What is the problem with more different content on a website about writing said content?

If, and even if we hypothetically raise the amount of articles needed and not the amount of authors that is bad as well. To clarify: I am English second third language and it takes on average about a month (yes, one month) to write an article in a decent state. And take into account _we do not only write for the things we want a tag for, we also do contests and stuff__. If the change goes through I'd see this personally as some kind of punishment for even trying to make some cohesive narrative over some of my SCP's and creating a hub. Should I just use the character everywhere then? Last addition to this point; if you think being ESL etc should not matter; this has been actively been used against me by some poor individuals (not anyone invested in this conversation though).

Another thing that came up in is the "they hinder navigation and should not be a thing that shows success". Quoting Plague here; "Tags are for navigation and organization, it’s not a mark of success. A mark of success is other people writing your idea. The tag just helps organize them" Then I'd like to officially request the deletion of the featured tag. If you want to keep an easy way to check which ones already got featured you can always make a list of them with backlinks etc. and refer to that list (which shouldn't be that hard especially since they are tagged atm). I'd even do it myself if needed, I could.

Also to quote Calibri Bold and Rounderhouse;
Calibri: "So you guys don't care about someone else's GoI having a tag, big deal. Someone else might care about it, and I don't think you guys being indifferent to it outweighs their investment in it"

Rounderhouse: "Essentially that. We're the ones approving them - so we're not going to approve tags primarily on the basis that they represent a big personal success to the author. If it matches the new metrics then I think it represents a considerable enough concept which a large relative to the growth of the site amount of people have contributed a large relative to the growth of the site amount of pages for. I don't think "barring people from getting tags for their canons" is even a considerable issue because we're scaling it up past a bar set in literally 2013 when you had to walk uphill both ways to the heritage collection
I'm pretty sure the only ones even getting hit by this are like ancient gois from the tenth century-oria, etc

Whilst proabably some truth in this, I really think this is a bit disconnected and a poor comment. Also calling other people their characters/GoI their "pet whatever/pet goi" and other comments not great as well (willing to provide evidence). Please, be more professional but to be fair I am really emotional atm (pissed off) and a bit disconnected as well which probably is visible in the poor quality of this comment. My apologies for the latter but I feel personally disrespected for trying to be constructive before, it is a shame this comment is so shitty. At least it is not 6 AM now and the placeholder is gone.

Also Rounder I really, really don't like you calling me a mf when trying to engage people to post here and calling me out on "not having time" in off site chat when I already reserve time in my limited scheduke in the morning to discuss policy at a regular basis. Screenshots available for staff. (Just to be a bit fair it was a general term to address people not directed as an insult towards me but still very poor.)

TL;DR No problem with classification requirements, heavily against character and goi requirements

Edit: I highly agree with Corq's comment on O5

Last edited on 06 Sep 2021 17:34 by Sirslash47 Show more

Re: [Discussion] - Raising Tag Consideration Minimums

Zyn 01 Sep 2021 06:51

Noting for records-keeping that I've seen the above post, and will bring this forward with staff.

Re: [Discussion] - Raising Tag Consideration Minimums

aismallard 01 Sep 2021 18:43

Thank you for expressing your views on the proposal. One thing to consider with tag approvals is that, from a staff perspective, tags do have a cost. The arguments you put forward could also be used to lower the tag requirement, but this would obviously have issues: there are already a large number of tags, and navigating more is not helpful if they do not bring a reader to many more things to reader. And on the processing side, when MAST applies tags, having many more smaller tags makes the time cost per article harder. This is one of the motivations of the tag rework, and cleaning up many attribute tags which are not well-used or notable.

Regarding Rounder's statement, from my understanding of the conversation, it was a general, casual exhortation for people to engage with policy on this thread rather than an attack on your character.

Re: [Discussion] - Raising Tag Consideration Minimums

morhadow 01 Sep 2021 10:08

I think the problem here is more "tags where the attribute/character/GOI/place is mentioned only instead of being relevant in the article beyond a few words" than "we have a shitton of tags and we are going to raise the requeriments".

If we'd do that, the problem would be at least partially solved.

Re: [Discussion] - Raising Tag Consideration Minimums

Kothardarastrix 01 Sep 2021 13:45

In that case, it seems to me that the solution would be to make it clearer what qualifies an article for such a tag, not this.

Re: [Discussion] - Raising Tag Consideration Minimums

Kothardarastrix 01 Sep 2021 18:58

If the issue is workload on the tagging team, maybe a better solution would be to reduce the number of existing tags, particularly attribute tags. If I'm looking for a a story I'll enjoy, I doubt that a tag like "ectoentropic" or "telepathic" will help me, because it's only relevant to the base properties of the object, not the story being told. And, honestly, I'm not sure that tags like "corrosive" or "liquid" are really all that informative about the object itself either, as broad as they are. What is the purpose of attribute tags, anyway? Am I missing something?

Last edited on 01 Sep 2021 19:06 by Kothardarastrix Show more

Re: [Discussion] - Raising Tag Consideration Minimums

Petrograd 02 Sep 2021 14:27

Just noting that I've used them to try to find similar articles or articles I've forgotten the links to when I'm writing or critiquing. That said, I'm not saying that's a very common use or that it's an argument to keep all the attribute tags.

Last edited on 03 Sep 2021 07:27 by Petrograd Show more

Re: [Discussion] - Raising Tag Consideration Minimums

Ellie3 01 Sep 2021 19:06

Noting that this is really only commenting on character tags.

Tags, at their core, are for organization across the wiki. Shouldn't it not matter how many authors there are for things like character tags, which don't have hubs and are more widespread across the wiki? By binding them to a tag, they pretty much get their own "hub", and that character is organized together. If you don't have that "hub"… well, good luck using the broken wikidot search function to look for mentions of a character. Of course, there should still be a goalpost of needing 3 authors, but I dont see the need to raise it to 5.

I've seen some criticisms that there are too many character tags and that it can be difficult for tagging staff to check, "oh, does this character have a tag?", because they can't exactly memorize every character that has a tag. That's… pretty much unavoidable, though. And it can be solved by leaving character tags up to authors. Personally, when I'm writing something with a character, it's focused on that character, and I usually use their tag beforehand to read all about them. I usually apply character tags myself, and I don't see why everyone can't do that, except for the people who apply tags for small little tiny mentions which… is a whole other problem.

Also, raising the limit to 5 will just make it harder to learn about a character. Characters tend to start out as passion projects, really— dudes that two or three people really invest themselves in (again, unless they're Personalityless Doctor Number 39243 who just gets tiny mentions, but again, that's a different problem). A tag is usually a collective effort between those people. Getting two or three people really invested in a character? That's easier than getting five, especially when you don't have a collective "hub". Once a character gets a tag, it makes it easier for people to collaborate when it comes to said character. And then, naturally, they grow even more.

Anyways! I kind of made tons of different arguments up there, but yes. Character tags should be kept as they are. Raising the author limit to 5 for character tags will just hinder modern efforts to develop newer characters and organize them together.

Re: [Discussion] - Raising Tag Consideration Minimums

Moximillion 01 Sep 2021 21:38

Mirroring Kothardarastrix and others, though with the caveat that while I don't think the requirements for a tag should be raised, the requirements for GOI/Canon hubs specifically should be raised to the proposed 10x5.

Re: [Discussion] - Raising Tag Consideration Minimums

Petrograd 03 Sep 2021 07:44

Having thought more about this, I also see issues with things like character tags and especially GoI/canon tags.

If you up the limit for a tag from 5x3 to 10x5, like Moximillion says, it wouldn't make sense to keep the threshold for the goi, canon, and MTF pages at 5x3.
(Isn't having a tag required for the GOI list?) However, this was supposed to avoid tag bloat, not MTF/GOI/canon bloat.

I'm all for reducing the terrible tagging mess, but I'm not for making it harder to promote those collaborative projects unless there is evidence that list bloat is also a problem.

Last edited on 03 Sep 2021 07:48 by Petrograd Show more

Re: [Discussion] - Raising Tag Consideration Minimums

LemonBee12 03 Sep 2021 08:38

I don't see a good reason to raise the requirements on GoI or Canon tags, for the same reasons croq provided.

Last edited on 03 Sep 2021 08:38 by LemonBee12 Show more