litphoria

No it wouldn't. Other tag based searches get along like gangbusters with this sort of system. All it would need is that curator role to keep the options tidy with synonyms, related options, pruning, merging, and so forth — I know at least a couple people who would love to do that — and a couple search toggles to accommodate those associations. In addition, preferences could also hook into the option relationships and intelligently display not-quite-preferences a little bit differently.

That curation system does something else too. People know that their options might be messed with via curation, that's the norm. As long as there is a culture of respect for the curators' job then there won't be much thoughtless creation. All it takes is a community that will peer pressure somebody out of entering thirty different variations of Vajooboo, or having social consequences for it. That would probably require that options have some meta-data tying them to the profile they were created on though, and performing acknowledgement of a curator's hard work in a public chat every once in a while.

It seems like a pretty thankless and tedious job to categorize all of the possible ways someone can describe their hair or personality through tags.

And I'm still not seeing how this won't make it a total nightmare for preferences. If this goes through, we could hundreds of tags for hair that could all very well be legitimate, but it would be hopeless to even imagine going through all of those options. Amplify that by all of the traits that would have this and you have yourself a real headache in filling out a profile. This is something other systems with tagging - like most boorus - don't have to deal with. They also don't typically encourage people to down to the level of frizzes in the rain.

I'm a fan of this kind of system. It's still searchable (implications, aliases, parent/child tag relationships, whatever kinds of connections we choose to use will help a lot) and I'm pretty confident it'll let our profiles be more expressive & specific in ways that are both useful to reading and to searching.

It also beats a number of the issues we have right now with the trait system's super strict usage requirements.

It seems like a pretty thankless and tedious job to categorize all of the possible ways someone can describe their hair or personality through tags.

That's nobody's job though. Specifically because people will be typing that stuff in when they want it and no sooner, and not need it prepared for them in advance.

If you were referring to the curator's job, there are people who don't find that tedious and who enjoy it for its own benefits. I do tag curation on various systems that empower me to do that and I enjoy it, and there's always multiple other people pulling major weight & loads of other people pulling small amounts of weight, and all of those people are doing it because they want to whether or not they get thanks for it.

And I'm still not seeing how this won't make it a total nightmare for preferences. If this goes through, we could hundreds of tags for hair that could all very well be legitimate, but it would be hopeless to even imagine going through all of those options. Amplify that by all of the traits that would have this and you have yourself a real headache in filling out a profile. This is something other systems with tagging - like most boorus - don't have to deal with. They also don't typically encourage people to down to the level of frizzes in the rain.

Can we set aside the frizzes in the rain thing for a moment? Because I'm not even going to set a preference for that thing myself, and it was just an example.

It poses a logistical challenge if you assume everyone has to fill out all preferences but that's a non-issue because nobody's going to do that nor should they be expected to. You're alarmed by that scenario, which means we need to avoid whatever scenarios are alarming.

I only fill out preferences on things I feel strongly about. This thing lets me set more preferences on more specific things I actually care about, and I wasn't going to fill out the rest anyway. Incidentally that means my list of preferences is going to exclusively consist of things I've actually got preferences on --- other than I got pressured into filling out, which dirties an otherwise really clean signal. The rest I don't set preferences for. (Some people get upset at the idea people don't fill out their profile with a clear stance on absolutely everything, but fuck that. That's not helping me communicate what I actually want, and they can go ask me if they're unsure. I'm still grateful the 100% completion thing went away recently like it did.)

So the answer for me personally is: I'm not going to go through all those options. I'm going to know options I want to talk about, and set preferences for those. I'll add more later as I care about it or find them. Other profiles in general are probably going to be a lot less more complete in terms of preferences, and I'm fine with that, because the ones they actually bother to list will be ones I know are important to them to list. That's good. That means I get a better picture of what they want.

If it's thankless, then we make a point of thanking them anyway. If prominent community members go out of their way to make sure that they're recognized, there will be esteem attached to the job. Positive reinforcement. If we give people positive attention for moderating these tags then people who would never have thought to contribute to that will see a social reward there and participate. So eventually that work load is spread out over time.

Frizzes in the rain was purposefully silly though. It's probably one of those tags that would be trimmed. This can be made a lot easier if there is a checklist system for creating trait options. "Is this not only specific to my character?" "Is this a major salient detail?" and so on. Better wording and more comprehensive than that, but there are other ways to prompt people to think about their trait options.

I'm not sure where the headache would come from though. I'm imagining that all tags would be filtered by default with a text input box. You type brown in the hair box and the system will show you only the tags with brown and its synonyms. It would show nothing until you filter it though. Preferences would work the same way. Synonyms are treated as the same tag and "close enough" variants (not sure what to call them) might be highlighted slightly different. So saying you like brown hair in a preference also includes chestnut, mouse brown, and so on unless you go out of your way to exclude one by disliking it.

Also: what Korra said.

So, all of this complexity and upkeep so that we can describe profiles with terms like mouse brown instead of brown? What exactly is wrong with just using brown and specifying exactly the kind of mouseyness of it in the trait context, like we do for everything else? Why does it need to be that someone can create 30 synonyms for tags already in the system?

Because nobody sets contexts and nobody even reads contexts. People do read trait options though.

Because nobody sets contexts and nobody even reads contexts. People do read trait options though.

Well jeez, this is a pretty serious problem, if so. I'd like to see something that encourages people to use and read these instead of just assuming it's dead weight and designing systems that go without it, though.

Also, nobody, or most? Most aren't supposed to get anything out of contexts. Most people who view your profile won't be interested. It's additional information that helps to flesh out your profile for the people who are interested, though, and it avoids bogging people down with a bunch of details (like attempting to decipher what exactly mousey brown is) when they're browsing through a lot of profiles.

The context thing is beside the point though. The thing is, the system we have now is not scaling well to the amount of trait options and interests we have. As well, the process of getting anything to describe more niche character traits is both cumbersome and overcomplicated if we want any reasonable amount of fidelity. People who are into xenobiology are seriously underserved with the single genital trait option of alien and anything that would serve them in the current system is going to have some serious scaling problems. This bypasses that entire problem by letting the trait options be created and maintained organically. Needs are met as they arise, rather than only squeaky wheels (the ones who propose trait options & get them endorsed) getting their needs filled. The curation system then massages those preliminary solutions as they're put into actual practice.

Rapid, small scale iteration rather than trying to get them all perfect on the first, second, third... try. And with these small scale iterations, actually pitching in on the iteration process is less of a task.

The point with context is salient, though.

If the context is displayed by default, for example, without expanding anything, then you already have your ability to have a custom-named option. As far as:

As well, the process of getting anything to describe more niche character traits is both cumbersome and overcomplicated if we want any reasonable amount of fidelity.

The proposal has someone created a hierarchy (or relating them somehow to one another) for the sake of seaching. Well, that's exactly what traits and options do already. The difference is instead of someone typing in "algoquian dark orc" they would select "orc" from the options and then type in the algonquain stuff into their context. Same thing with the alien for genitals - the options are not meant to have the fidelity, that's for the context.

As for the process itself, well, that can be iterated on as well. People have suggested being able to simply upvote, and it's entirely possible that all that would have to be done is suggest something and then have staff check it before it is added. However, you have an uphill battle convincing me that everyone being able to add 30 options for the same thing and requiring someone to constantly prune the options (potentially every time someone edits a profile!) is at all a better way to approach quality control.

Also, we have the option of doing the system now as a hat trick: imagine someone inputting something into the text box as proposed here, but instead of that automatically creating a new option that someone has to go maintain, it just sets the context for a trait that already exists.

So because there's a lot to this suggestion, I wanted to be more clear about some of the pieces of it, and better collect my thoughts:

• Multi-select traits Sure, though might confuse people in writer profiles (where multi-selection would be used to determine variation)

• Trait options related to what I picked Yes, good, useful, similar to what we're doing now with the creator

• Input box first in the add new trait flow Is where you start to lose me. It is my opinion that presenting this too early would be intimidating. I think a better flow would be to present the options and have an "other" button which then allows custom input. I would also say that I believe this input box should just go right into the context. (I should also say that this is how the site behaved very early on; trait options are technically optional)

• Free-entry of options It is my opinion that this is not necessary with context. The provided reasoning that I can see, so far, is entirely covered by context, and the search-ability of it is already covered by trait options.

  • Streamlined creation of new options Is implicit as part of this free-entry, and imo I think it needs to be a separate suggestion for a separate problem

• Curators We actually already have community managers, which behave in this capacity to some degree already (they help to trim and clean up trait options as one of their duties). However, I did not hire them with the expectation of work described in this suggestion.

If the context was improved so that it always displayed, certain traits were allowed to have multi-select, and we found some way to streamline the creation process of new items, what problems would we all have with that as a solution?

Pixiv is a website which just allows users to create whatever tags they want and it works sort of well for searching. I don't think they use any sort of curation at all though, and it's just a free-for-all. The issues with searching would only get better if the free-for-all was cleaned up a little.

I can imagine that this suggestion would be a lot of effort to implement, not to mention the curation effort, and that is definitely a point against it. I think once that issue is set aside (if it can be set aside), the suggested system here is better than what we currently have.

Right now, the majority of emphasis on profiles is placed on traits and preferences and interests. We recently got text fields added in to combat this a little, and maybe new people will use them, but for example I know I've been too lazy to go back and edit all my profiles to take advantage. Moreover, the layout naturally draws the eye to these things anyway, because they are more quickly digested and better arranged than just blobs of text. This isn't a point against what we have, I really like how this site looks on the whole, but I think it's fair to say that profiles are mostly communicated now through their traits.

Context does a lot of the job that this suggestion is trying to do, but context requires a click. I disagree that "people don't read contexts" I know I sure do anyway, but with how it's implemented now it feels like extra information, something to clarify aspects of a profile. Under the paradigm suggested here, unique features about your profile wouldn't be amendments, they'd just be aspects of your profile. And it would definitely be a much better opportunity to communicate some of your personality as a writer through the basic aspects of your profile.

More importantly, I feel like the current system of feedback for traits just isn't really working. We have a bunch of discussions about which trait options need to be added, then end up creating more feedbacks to remove whole traits that are judged unnecessary, and those go nowhere because people have different views and some people want to be able to implement X character and others feel X character is better implemented like so, etc etc This whole mess disappears under this new paradigm. At worst you'd have people arguing over what's a better name for the same core thing, or specifics about the curation.

So to summarise, where I'm at is, this system seems to me like it'd be an improvement indeed, and make profiles more fun, but I can totally understand if the architecture re-write it'd require behind the scenes is too much.

Eye color is a bad example to think this idea through since it's a quite useless trait. Try the same with gender and the problems get visible.

The first mental gender tag I would add is "woman". Other players might try "girl", "bimbo", "shemale", "tomboy", "slut", "butch", "shota", and so on. So, is "tomboy" a mental gender? Or a gender expression? Is the curator free to decide this or are there guidelines? Since there are players who are into "tomboy" characters it's a useful tag. Is that enough to permit it? If it is permitted will it be subsumed under female, male, androgynus or all three? It's easy to guess that chestnut means brown but other traits are more complicated. How this turns out would depend on which tags you let pass.

@Otaku The feedback system works so, that a decision is made in the end. Accepted or rejected. Plantigrade, etc. was replaced by feet, hooves and paws. With the tag system we would simply have all these and a few more.

Eventually I'll go through and read all of the comments, but I want to start out by saying - this is the kind of system that I really hope eventually this site would move toward.

I am also the kind of person that would happily do such a thankless job as constantly organizing and re-categorizing these associations. Because it's just something I adore doing, honestly. I can't be the only person that would actually like to sit here for a few hours and go through new tags, gather them up, and make sure they're tidy. I find it stress-relieving, but don't feel like I'd have the motivation to ever put myself forward as moderation or community management or anything like that. Simply curating tags for a few hours a week would be a nice way to help out though.

And now I can read comments!

I'd like to see something that encourages people to use and read these

I am one of those people that is guilty of not reading context. The only thing that would help me read context would be if I could search it.

But, let's take something like "breath play" for a moment. Because there are so many types of this thing, although I do enjoy one or two of them, I don't really search on, use, or read context of it because 90+% of it is not relevant to my interests personally. I don't like drowning, or garotting, or weird latex-based suffocation bags, or limited oxygen supply. So I just ignore it when it's on a profile out of the assumption that it's just going to be another thing I don't care for. It usually is, why bother taking the time to read someone's (maybe even well thought out) writings on the breathplay they like only to get to the end and see... ooop! Nothing I'm interested in!

Could this be mitigated in this situation with more specific interests? Sure! But the specific was added right on conception (it used to be choking), and changed into something general because of being "too specific". I'd rather not get into circular discussions like that. Over and over again. They eventually turn into a go-nowhere rut. I'd rather simply be able to add and search my specific without it getting in anyone else's way.

This is talking about traits though, not interests.

Because there are so many types of this thing, although I do enjoy one or two of them, I don't really search on, use, or read context of it because 90+% of it is not relevant to my interests personally.

Though, this is exactly why there is context and they're collapsed by default in the first place though, so the additional specific details don't bog you down when you're scanning through a profile initially.

And they don't bog me down.

But they're also not helpful or useful in a way that is meaningful to me personally. To continue the same example, I cannot find people that are into domineering manual strangulation. I have to search them by breathplay. Then read context. Then find out that's not included in what they're looking for. Then move onto the next person. Then read their context... and so on.

When the interest was first created, this was okay. It was early, it was called "choking", and there was some discussion about being able to present "severity", so I waited to see what might come of that. Nothing did, it was deemed not that important. So finding specifics is still a chore and there are quite a few people that (rightfully, with the way interests are presented now) don't want my specifics that I find important to bog down their search/interest list. Which is great for them! It shouldn't.

But leaves me less and less inclined to use the site's search (because seriously, reading through 100 context boxes to find maybe 1 person isn't fun) and more and more inclined to simply use this as a place to post writing samples.

Alright, but, even if this was implemented exactly as it was suggested, it wouldn't impact your specific scenario that you're outlining, since you're talking entirely about interests, and this suggestion is focusing entirely on traits and how to streamline adding and presenting them.

That's why it's only an example. I don't want to talk about things that I wouldn't have any strong opinion on, and I have no strong opinion on the things traits describe. But filling in context works the same no matter if we're talking traits or interests.

Trait options are exclusive, interests are not. You cannot choose "blue eyes" and "grey-blue eyes" even if there are both trait-option.

Replace the words breathplay with that old trait "ethics" and the scenario around that then, and it's the same basic issue. Some people really, really liked, wanted, and probably used that as intended. It got in the way of a vocal portion, with no way of knowing if it was even a majority. And if litphoria is anything like any other site, it is actually a minority that participate in suggestions forums. That vocal portion had it replaced with something more general (alignment in this case). But there should still be a way to search that ethics thing, because it was obviously important to whoever suggested it in the first place, so it should still somehow be there.

Right now, it's there if you fill in the context. Contexts aren't searchable except by the vague upper-level traits that are supposed to encompass them. Since context isn't searchable, people want specifics added so the thing they find important can be searched.

@Tilly

Well this suggestion is about traits. If a player loves long hair but only "chin length" and not "shoulder length" then he could find characters with his preferred hair length by the player made trait option. I just believe this is rare. And the player who loves long hair in general will have problems to find all profiles with long hair since there is a dozen different trait-options for them, "waist-length" and "floor length", and so on.

For interests it makes more sense since I can set both: "choking" and "breath play". There is the problem though, that we might have an alphabetical list of 5000 interests in the end. Which after all is the reason why "ethics" was replaced by "alignment", instead of simply having both. I liked ethics more than alignment but I also have an interest that the trait list is kept short.