Requested feature overview description.

We currently have tag groups for Artists, Characters, Species, General, and Meta. I've noticed that "General" can be very large sometimes, so if there was a way to divide it up further, that would be nice.

This proposal is for creating a new tag group called "Colors", and I propose placing it between General and Meta. These tags can often be clutter when trying to look at the list of tags if you only care about the main topics, or vice versa if you want to browse more characters with the same colors, it would be easier if the colors were separate.

Why would it be useful?

I'll use post #2338448 as an example, since it involves characters with many different colors. The following tags (16) would be placed into the "Colors" tag group:

black_body black_feathers black_sclera blue_eyes blue_mane blue_tail_feathers green_body green_feathers grey_body grey_feathers orange_body orange_feathers tan_body white_body yellow_body yellow_feathers

This would leave the following tags (39) in the "General" tag group:

3_toes 4_ears anthro anthro_on_anthro anthro_penetrated anthro_penetrating anthro_penetrating_anthro blurred_background cunnilingus detailed_background eyes_closed feathers female female/female female_penetrated genitals group male male/female male_penetrating male_penetrating_female mane multi_ear nude oral penetration penis pussy sex tail_tuft toes tongue tongue_out tuft vaginal vaginal_penetration wings

So the "Colors" section for this image would have a bit under half as many tags as "General", a sizeable section indeed. In the current state of the color tags being merged with "General", the color tags are about 30% of the tags.

Of course, not all posts would be like this one. Some posts would have more, probably most would have less, but I think this is a big enough category that it's worth splitting off.

What part(s) of the site page(s) are affected?

Primarily the left sidebar's list of tags when viewing a post, but also the tag would change color when browsing/searching through posts.

WDYT?

Maybe you could give your support to [Feature] having tag groups or more categories

Yeah, I think having a 'color' tag category (and possibly a 'clothing' tag category) would be beneficial. However, since it took a while for the 'lore' and 'meta' tag categories to be implemented, it'll probably be a while before any new tag categories are implemented.

Bump, I still think this feature would be very useful. A post can have lots of color tags, and it seems silly to have these mixed in with the non-color tags. I added some color tags on post #3159710 and now most of the general tags are color tags.

Yep, looking for general tags in a clutter of color tags can be pretty annoying, especially if colors are well tagged. Take a look at this beauty: post #3143453!

Would definitely be a bit easier on the eyes if they had their own group.

I don't see how it would help much since the tags would still be visible. If anything, it would make it worse since it's yet another tag header (lengthening the tag list with extra space and more lines), and another tag type, tag color, etc, to manage. Also would monotone_*, two_tone_*, multicolored_*, rainbow_*, etc, count as "color" tags? They're not color themselves, just an indication of how many colors are present on the thing. With some of the tagging jobs I've seen, having a separate Colors group wouldn't help the clutter; people who would tag colors a lot also tend to tag every possible minutia of the image, overloading the tag list in a way that no amount of separate groups would help make sense of the list.

watsit said:
I don't see how it would help much since the tags would still be visible. ...Also would monotone_*, two_tone_*, multicolored_*, rainbow_*, etc, count as "color" tags?

Tag headers can be collapsed. Checking general tags for correctness would be easier without my eyes tripping on colors, and vice versa. I think all color-related tags should be put into "color" category.

waydence said:
Yep, looking for general tags in a clutter of color tags can be pretty annoying, especially if colors are well tagged. Take a look at this beauty: post #3143453!

I'm sorry for being thorough with my tagging.

waydence said:
Tag headers can be collapsed.

Which is a deliberate action you have to do, they're still there by default. And if you're looking over the list for missing or invalid tags, the color tags are just as relevant to check as anything else.

waydence said:
I think all color-related tags should get their own category.

So socks_(marking), gloves_(marking), striped_body/fur/scales/feathers, spotted_body/fur/scales/feathers, dipstick_tail, dipstick_ears, heterochromia, and other tags like that should be in the "color-related" tags, since they're related to colors? How many tags would actually end up in this category?

thegreatwolfgang said:
I'm sorry for being thorough with my tagging.

Thank you for your service. While at it, may I introduce you to countershading?

watsit said:
Which is a deliberate action you have to do, they're still there by default. And if you're looking over the list for missing or invalid tags, the color tags are just as relevant to check as anything else.

Maybe your brain works differently, but I find it disruptive having to scroll below the picture to check. Search doesn't really help with that.

watsit said:
So socks_(marking), gloves_(marking), striped_body/fur/scales/feathers, spotted_body/fur/scales/feathers, dipstick_tail, dipstick_ears, heterochromia, and other tags like that should be in the "color-related" tags, since they're related to colors? How many tags would actually end up in this category?

Perhaps tags that directly imply colors? The main source of clutter, like *_body (fur, feathers, scales, hair), *_background, *_clothing etc. But not stuff that technically works with greyscale, like markings. Hopefully more people will express their opinion.

waydence said:
Thank you for your service. While at it, may I introduce you to countershading?

Perhaps when I learn the difference between it, light_body, dark_body, light_countershading, dark_countershading, and any other body part specific subtags they have first.

waydence said:
Maybe your brain works differently, but I find it somewhat disruptive having to scroll below the picture to check. Search doesn't help all that much.

I find it a problem, too. If I have to scroll away from the picture to check if a tag is there, or if I have to scroll back to the picture to check the validity of a tag, and the shear number of things being tagged, that makes it harder to ensure proper tagging, and especially when each tag looks very similar to the ones its next to, which makes me prone to misreading or overlooking them. However, this is a more general problem of overtagging, and tagging obvious/expected things (we generally don't tag expected things, or tags that have a direct opposite, like eyes_open/eyes; eyes being open/visible is going to be true 99.9% of the time, so there's no real use in tagging it, and there's the eyes_closed tag if you want to find the outliers).

post #3143453 and post #3325236 are good examples here. There are so many individual things tagged, many things which are obvious/expected (like monotone_leash or monotone_glans, as if there's such an abundance of multicolored leashes and glans and they're necessary to separate from the plain leash and glans tags), that it makes verification impossible. As a result, for example, many posts that have monotone_eyes are mistags (it's tagged on a lot of posts that have monotone pupils, again another obvious/expected thing, but have different color sclera and irises, rather than truly monotone eyes where the sclera, pupil, and iris are the same base color), but they're hard to realize as its often buried in long tag lists along side a bunch of other monotone_* tags. The second post is a proper use of monotone_eyes, while the first one is difficult to tell given its size and blurry background details (the clearer foreground details at least would not warrant the tag), but is tagged anyway, and is easy to overlook in the tag list to correct.

I don't think putting color-related tags into a separate group will actually help all that much, since when the color tags become a problem, the tag list as a whole is going to be excessive; there's still going to be a lot of both non-color and color tags that require scrolling back and forth even with the other sections collapsed. Tags that are easy to overlook next to similarly-named tags, are going to congregate into the same groups anyway, keeping it unwieldy regardless.

waydence said:
Perhaps all tags that reference colors?

What do you mean by "reference colors"? To me, that would mean only tags like white_*, red_*, black_*, which are colors, and not monotone_*, two_tone_*, etc, since they aren't referencing any particular color. There's also tags like red_theme, blue_theme, greyscale, etc, which are color related, but are meta tags. Would they stay meta, or moved to this "colors" group?

Never thought overtagging could be an actual problem.

watsit said:
What do you mean by "reference colors"? To me, that would mean only tags like white_*, red_*, black_*, which are colors, and not monotone_*, two_tone_*, etc, since they aren't referencing any particular color. There's also tags like red_theme, blue_theme, greyscale, etc, which are color related, but are meta tags. Would they stay meta, or moved to this "colors" group?

Tags that directly imply colors on particular things. Red_body says that body is red, multicolored_body implies that there are colors present on body. Meta tags are referring to the whole picture, so they can stay as they are, I think.

watsit said:
I don't see how it would help much since the tags would still be visible. [...]

This group would make sense to have at the very bottom of the tag list, maybe even below the Meta group, or it could be above Meta. The point is to de-clutter the General group. Even when not collapsed, just moving these tags farther down will increase visibility for the other tags.

Overtagging is not the problem. Keeping lots of tags organized is the problem. This proposal helps with organization.

aaronfranke said:
The point is to de-clutter the General group. Even when not collapsed, just moving these tags farther down will increase visibility for the other tags.

You can make this argument about any tag group. Weight tags, genital tags, sex tags, pairing tags, etc; moving them down would increase visibility for the other tags. What makes color tags so special?

aaronfranke said:
Overtagging is not the problem. Keeping lots of tags organized is the problem. This proposal helps with organization.

If color tags are causing the tag list to be unwieldy, it is indeed an overtagging problem. You can organize the tags all you want, but ultimately there will be a lot of tags that make it nearly impossible to check over no matter how many separate groups you make (and the separate groups in fact making it worse). Most posts don't have a problem, but those that do are because someone added tags for every minute detail they could think of. If you aren't looking at the tag list, it won't matter, but if you are, separating the color tags is just as arbitrary as separating the clothing, sex, or fluid tags.

watsit said:
If color tags are causing the tag list to be unwieldy, it is indeed an overtagging problem. You can organize the tags all you want, but ultimately there will be a lot of tags that make it nearly impossible to check over no matter how many separate groups you make (and the separate groups in fact making it worse). Most posts don't have a problem, but those that do are because someone added tags for every minute detail they could think of. If you aren't looking at the tag list, it won't matter, but if you are, separating the color tags is just as arbitrary as separating the clothing, sex, or fluid tags.

I don't know why you are painting the whole thing as being a "problem".
Detailed tagging is very useful in finding posts through individually coloured body parts, especially when trying to find the originals from cropped artworks.

If it is unwieldy to the eyes, have it be collapsable as a tag group.
If it is hard to moderate, let those who are actively using the colour tags to fix any mistagging they find on their search results.

thegreatwolfgang said:
Detailed tagging is very useful in finding posts through individually coloured body parts, especially when trying to find the originals from cropped artworks.

Checking tags is also important, to ensure there are no glaring omissions or incorrect tags, which is hard to do when there are a lot of tags and each one looking similar to the several others around it. And with the more extremely specific tags that are only on a handful of posts that it could apply to, when there are slightly less precise but plenty useful tags that are just fine by themselves, aren't really serving their purpose.

thegreatwolfgang said:
If it is unwieldy to the eyes, have it be collapsable as a tag group.

Which as explained, won't help (there will still be a lot of tags to look through, and the similarly-named ones that make looking through it difficult will tend to congregate into the same group).

thegreatwolfgang said:
If it is hard to moderate, let those who are actively using the colour tags to fix any mistagging they find on their search results.

How many of these tags are actually being used for searching, and aren't just needlessly padding out the tag list for the sake of having more tags? How many are only used for searching on rare occasion, but have been filling up with mistags which went unnoticed by people who would've fixed them? As it is, monotone_eyes has been made useless by its rampant misuse, with monotone_breasts not too far behind, which no one who bothers to fix tags was able to catch since many misuses are buried alongside other similarly-named tags, and by the time it was noticed, it's too much to clean out. So rather than being useful for the occasion someone might've needed it, it's there on normal non-monotone instances where it serves no functional purpose.

That's why I'm painting it as a problem, because it's causing tags to go on being misused to the point of ruining them, and people like me who have a habit of checking the tags, are being hindered in fixing problems, be it missing or misused tags. If that's not a problem, I don't know what is.

watsit said:
How many of these tags are actually being used for searching, and aren't just needlessly padding out the tag list for the sake of having more tags? How many are only used for searching on rare occasion, but have been filling up with mistags which went unnoticed by people who would've fixed them? As it is, monotone_eyes has been made useless by its rampant misuse, with monotone_breasts not too far behind, which no one who bothers to fix tags was able to catch since many misuses are buried alongside other similarly-named tags, and by the time it was noticed, it's too much to clean out. So rather than being useful for the occasion someone might've needed it, it's there on normal non-monotone instances where it serves no functional purpose.

That's why I'm painting it as a problem, because it's causing tags to go on being misused to the point of ruining them, and people like me who have a habit of checking the tags, are being hindered in fixing problems, be it missing or misused tags. If that's not a problem, I don't know what is.

You cannot expect every single tag to be actively used in everyday searching, it would be very useful in times where examining a post in detail is necessary.

As for mistagging, I don't understand how something as simple as <color>_body_part would get misused. It is simple to understand and the people who tag them would most likely be the only ones using them.
Your assessment of monotone_eyes and monotone_breasts being misused is based off what definition? There is no wiki written for the both of them.
If we are following the conventional use of monotone_*, I would say that that would be the eyes and breasts that are monotone, whilst not considering the colour differences of the pupils/sclera or nipples/areola.

watsit said:
You can make this argument about any tag group. Weight tags, genital tags, sex tags, pairing tags, etc; moving them down would increase visibility for the other tags. What makes color tags so special?

Not really. They are special because they are a very large and easily identifiable group in general category. For every bodypart or object tag, there's a set of subtags in standart color palette.

Here's a definition I came up with:
Color tags are tags that directly describe coloration of things within picture.

So if it can only be identified on colored image, it's a color tag: red fur, rainbow body, green markings etc.
If it can be identified on greyscale image (or only suggests coloration), it's not a color tag: markings, countershading, heterochromia etc.
Metatags are metatags.

NGL though breaking up tags into further categories sounds appealing, I'll have to try it. Because the smaller the list, the easier it is to identify problems. Have a "colors" category, "actions" category, "bodyparts", "objects"...

thegreatwolfgang said:
You cannot expect every single tag to be actively used in everyday searching, it would be very useful in times where examining a post in detail is necessary.

When is that necessary to examine a post in detail to make it useful for searching? If someone's searching for a post, they're generally only going to have a view pieces of information available, like the artist name, species, sex/genitals, and perhaps some vague details. We can't and shouldn't expect or encourage people to search so precisely so as to be able to get only that one specific result they're after. After all, if you're searching for something you won't have all the details, and some of the details you do have could be wrong, so encouraging hyper-specific tag searches would be a hindrance to actually finding things, rather than using more vague details and trying different alternatives if it's not coming up in the results. The more specific you try to be, the harder it will be to work with.

thegreatwolfgang said:
As for mistagging, I don't understand how something as simple as <color>_body_part would get misused. It is simple to understand and the people who tag them would most likely be the only ones using them.

The people adding tags are the ones searching for them? That sounds quite backwards to me. Unless you mean the only use these tags have is to be added to posts, which means it's just extra tags for tagging's sake. As for how, well just look. monotone_eyes is being used on posts that have more than one color in their eyes. monotone_breasts is being used on posts that have more than one color on their breasts. People started tagging them without considering how they'd be interpreted, and didn't consider which ways they would be most useful, so now the usefulness they would've had for searching is gone and they're just tags to pad out the tag list.

thegreatwolfgang said:
Your assessment of monotone_eyes and monotone_breasts being misused is based off what definition? There is no wiki written for the both of them.
If we are following the conventional use of monotone_*, I would say that that would be the eyes and breasts that are monotone, whilst not considering the colour differences of the pupils/sclera or nipples/areola.

Based on common interpretation, and a basic assessment of how they'd be most useful. monotone_eyes would mean the whole eye is a single color, not just the pupils, just as monotone_fur would mean the whole fur is a single color, not just some arbitrary section of fur. Your interpretation of only considering the pupil is next to useless since the vast majority of eyes are single-color pupils distinct from the sclera (where even if you do search it, would include so many posts that it won't narrow the results down by any useful measure), and creates absurdities like post #3327670 being "monotone eyes" despite there being two distinct colors in the eyes. Consequently, on the occasion where someone needs/tries to search for true monotone eyes, like in post #3295801, their search will fail because it's flooded out with eyes that have multiple colors.

Similar for breasts. Nowhere does it say the the areola/nipples aren't part of the breasts. In fact, there is the featureless_breasts tag which indicates when the breasts are lacking the features of areola and nipples, meaning they are otherwise considered part of the breast when present (for X to be a feature of Y, X needs to be part of Y).

watsit said:
When is that necessary to examine a post in detail to make it useful for searching? If someone's searching for a post, they're generally only going to have a view pieces of information available, like the artist name, species, sex/genitals, and perhaps some vague details.

Again, useful for cropped artworks with the absence of all of these identifying information, especially in the case of cropped yiff artworks used for memes.
E.g., An extreme closeup on a character's face where the only things that are identifiable are the select few facial features would make these detailed tagging useful.

The people adding tags are the ones searching for them? That sounds quite backwards to me. Unless you mean the only use these tags have is to be added to posts, which means it's just extra tags for tagging's sake. As for how, well just look. monotone_eyes is being used on posts that have more than one color in their eyes. monotone_breasts is being used on posts that have more than one color on their breasts. People started tagging them without considering how they'd be interpreted, and didn't consider which ways they would be most useful, so now the usefulness they would've had for searching is gone and they're just tags to pad out the tag list.

I meant that as people who are experienced with tagging in that detail, or in order words, people who are actively using these tags would be the ones who would know about it.
No ordinary user will go above and beyond to tag minute details like nose colours or finger colours, most likely they would just settle with <color>_body and call it a day.

Based on common interpretation, and a basic assessment of how they'd be most useful. monotone_eyes would mean the whole eye is a single color, not just the pupils, just as monotone_fur would mean the whole fur is a single color, not just some arbitrary section of fur. Your interpretation of only considering the pupil is next to useless since the vast majority of eyes are single-color pupils distinct from the sclera (where even if you do search it, would include so many posts that it won't narrow the results down by any useful measure), and creates absurdities like post #3327670 being "monotone eyes" despite there being two distinct colors in the eyes. Consequently, on the occasion where someone needs/tries to search for true monotone eyes, like in post #3295801, their search will fail because it's flooded out with eyes that have multiple colors.

Similar for breasts. Nowhere does it say the the areola/nipples aren't part of the breasts. In fact, there is the featureless_breasts tag which indicates when the breasts are lacking the features of areola and nipples, meaning they are otherwise considered part of the breast when present (for X to be a feature of Y, X needs to be part of Y).

I disagree, for your definition of monotone_eyes, it should be renamed to avoid any confusion.
If you are looking for uniform colours throughout the sclera, eyes/iris, & pupils, I wouldn't call it monotone_eyes.
If you look in the eyes wiki, colours are sectioned off into three sections - eye/iris, sclera, and pupil. The eye is considered separate from the sclera and pupil.
E.g., Empty_eyes (only has iris, no pupil or sclera) is not the same as no_pupils (has iris & sclera, but no pupil) or no_sclera (has iris & pupil, but no sclera).

Same for breasts. If a character's body is entirely white in colour, having white_breasts but with pink_nipples, I would not call it two_tone_breasts.
Or with white_breasts + pink_areola + red_nipples, I would not call it multicolored_breasts.

Sir, this is a feature request.

thegreatwolfgang said:
Again, useful for cropped artworks with the absence of all of these identifying information, especially in the case of cropped yiff artworks used for memes.
E.g., An extreme closeup on a character's face where the only things that are identifiable are the select few facial features would make these detailed tagging useful.

I still challenge the usefulness of such tags, like monotone_eyebrows, teeth, red_mouth, your interpretation of monotone_eyes. If you're needing tags like that for a search, you may as well just forward to https://e621.net/posts. More over, with a little as these hyper-detailed tags are used compared to their slightly more vague cousins, the likelihood of such tags aiding in the discovery of the desired post is next to 0, while slightly more vague but more widely used tags would increase the chances of finding what they want. Again, we shouldn't expect, or even encourage, people to search with as highly detailed particulars as they can so they can land perfectly on the precise result they're looking for. Such a thing won't be possible.

thegreatwolfgang said:
I meant that as people who are experienced with tagging in that detail, or in order words, people who are actively using these tags would be the ones who would know about it.
No ordinary user will go above and beyond to tag minute details like nose colours or finger colours, most likely they would just settle with <color>_body and call it a day.

That comes across as rather pretentious and somewhat clique-ish. That just because you add a lot of hyper-specific tags to posts, whereas I just try to ensure good and correct tagging with somewhere around 30 to 50 or so tags per post, you get to decide which tags should go on posts and what the tags mean. I may not add 100+ tags to every post I upload, but when I see tags getting ruined because a lot of apparent fluff tags are hiding their misuse, I think it's pertinent to speak up about it.

thegreatwolfgang said:
I disagree, for your definition of monotone_eyes, it should be renamed to avoid any confusion.
If you are looking for uniform colours throughout the sclera, eyes/iris, & pupils, I wouldn't call it monotone_eyes.
If you look in the eyes wiki, colours are sectioned off into three sections - eye/iris, sclera, and pupil. The eye is considered separate from the sclera and pupil.

Colloquially, people refer to the iris color as the "eye color", because in normal general real-life usage, the eye has white sclera and a black iris, leaving the only "color" to the pupil. But the eye is composed of the sclera, pupil, and iris, and with art there are more options for different-colored sclera and irises. So just as the chest isn't the only part of the body, the pupil isn't the only part of the eye. Saying a character that has monotone_pupils should be tagged with monotone_eyes even when the sclera is a different color, is like saying a character that has monotone_fur covering their chest should be tagged monotone_body even when the scales covering their stomach/abdomen are a different color.

I strongly believe using monotone_eyes to refer to the eye as a whole (iris, pupil, and sclera, whichever happen to exist) all being one color is a perfectly reasonable interpretation and not hard to grasp when seeing it used that way (in fact, I had been trying to think of what to call/tag it when the eye as a whole was just one color, and when I happened to spot monotone_eyes on some post, it was a head-slap "duh" moment for me; until I saw the state the tag was in, which really set me off). Considering monotone_pupil is even less useful than white_sclera (which is aliased away due to being so common), having monotone_eyes effectively be an alias of monotone_pupil makes it painfully unnecessary to use that way, while people seeing it apply to images like post #3368532, post #3368307, and post #3327670 would come away less sure about what it actually means.

thegreatwolfgang said:
Same for breasts. If a character's body is entirely white in colour, having white_breasts but with pink_nipples, I would not call it two_tone_breasts.
Or with white_breasts + pink_areola + red_nipples, I would not call it multicolored_breasts.

I disagree. If I had a cropped picture of just some character's breasts, and it had red nipples and pink areola, with white covering the flesh/fur/scales around it, I would not expect to find it under monotone_breasts. I would not do such a search with monotone_breasts because it makes no sense to not consider all parts of the breast for whether it as a whole is monotone or multicolored.

Take a look at this post: post #4111967

Here are all of the colors tags on this post:

That is NEARLY 300 COLOR TAGS.

This means that almost half of this post's 650 tags are just for colors.

Really, congrats to the taggers who did all of that. I am shocked that there are people who care this much about the fine intricacies of color, and I salute you.

I don't know if all of these tags are worth keeping, that's not my call. I do think that many of the tags are worth keeping. Regardless, I am not trying to get rid of them or invalidate them, yet there is a big problem here that needs solving. All of these color tags are massively bloating the tag list, and making it difficult for people to read the general tags, because there are hundreds of colors mixed in.

So, again, I restate my original proposal: Add a colors tag group. Let's group together all of these color tags so that they do not take over the entire general section.

aaronfranke said:
Really, congrats to the taggers who did all of that.

The tagger that did that got banned for doing it.

Also, FYI, the chances of getting a new tag category (at all, not even just for colors) is next to zero. If anything I'd take a bet on getting a category for contributors first, and that's been an ongoing discussion for years. Our developer has said he can't just add categories, due to how complex adding them is. I'm making an educated guess here, but essentially every single post on the site needs to be reimported into the elasticsearch index, which is agonizingly slow. Multiple days, if not weeks slow. And that's not mentioning the actual updates that need to be applied to the database, and anything else I'm not thinking of.

donovan_dmc said:
The tagger that did that got banned for doing it.

honestly, honestly? we should probably discuss nuking and/or invalidating a bunch of the tags that Kog created/populated. we need a more bespoke approach to tag creation than what lead to all that nonsense being a thing. I'd argue that all of the monotone_* tags and 90+% of the <color>_<body_part> tags are entirely pointless, with like, the eye colors, and like maybe the genital colors and a few others being kept.

dba_afish said:
honestly, honestly? we should probably discuss nuking and/or invalidating a bunch of the tags that Kog created/populated. we need a more bespoke approach to tag creation than what lead to all that nonsense being a thing. I'd argue that all of the monotone_* tags and 90+% of the <color>_<body_part> tags are entirely pointless, with like, the eye colors, and like maybe the genital colors and a few others being kept.

We (internally) tried to have a discussion around invalidating a bunch of these tags (for the last ~6 months), but that's gone nowhere. Most of the tags he created and populated DO need to be dealt with, but no one wants to put in the effort. I'd personally like to get rid of all of the light/dark/glistening tags, but that seems to be an unpopular opinion.

donovan_dmc said:
I'd personally like to get rid of all of the light/dark/glistening tags, but that seems to be an unpopular opinion.

I'm on board with nuking the light_* and dark_* tags, given how vague and subjective they are (it depends on a point of reference, the same thing can look light or dark when placed in a darker or lighter environment; and someone who deals with darker things more often will be more likely to see something is "light" that other people wouldn't, and vice-versa). As well as most monotone tags, of course.

I agree with nuking those tags.

The best approach is probably to think about what tags to keep rather than what tags to get rid of. At least for body parts, are there any color tags that we need to keep other than for skin, fur, scales, belly, hair, and genitals?

Also keep the *_feathers tags.

oozeenthusiast said:
The best approach is probably to think about what tags to keep rather than what tags to get rid of. At least for body parts, are there any color tags that we need to keep other than for skin, fur, scales, belly, hair, and genitals?

Nose, pawpads, eyes, tongue, claws - accent colors that are usually different from the main body (I like searching by them, and they're useful for identifying characters)

nimphia said:
Nose, pawpads, eyes, tongue, claws - accent colors that are usually different from the main body (I like searching by them, and they're useful for identifying characters)

Really? I wouldn't expect nose to be useful for anything.

As for clothing and accessories, my first guess would be that they could all be grouped into top, bottom, underwear, headwear, and jewelry. Maybe shoes too.

oozeenthusiast said:
Really? I wouldn't expect nose to be useful for anything.

Hello, occasional [color]_nose searcher here 😅 I like when characters have unusually colored noses (and pawpads, claws, tongues, etc)

oozeenthusiast said:
As for clothing and accessories, my first guess would be that they could all be grouped into top, bottom, underwear, headwear, and jewelry. Maybe shoes too.

I think I'm fine with going one layer deeper than just that. so, like attaching colors to stuff like shoes, boots, sandals; skirts, shorts, pants; gloves, wristbands, bridal gauntlets; jackets, dresses, suits would be fine.

dba_afish said:
I think I'm fine with going one layer deeper than just that. so, like attaching colors to stuff like shoes, boots, sandals; skirts, shorts, pants; gloves, wristbands, bridal gauntlets; jackets, dresses, suits would be fine.

There are a lot of different pieces of clothing that could be tagged if we allow that though. One particularly colorful and accessorized character could get 30~50 tags.

dba_afish said:
I think I'm fine with going one layer deeper than just that. so, like attaching colors to stuff like shoes, boots, sandals; skirts, shorts, pants; gloves, wristbands, bridal gauntlets; jackets, dresses, suits would be fine.

Yes please, I like clothing tags like this. I am very interested in fashion. Things like the tags for fingerless_elbow_gloves bug me though. But things like pink_dress, green_socks, yellow_bracelet, I think are good and perfectly fine

oozeenthusiast said:
There are a lot of different pieces of clothing that could be tagged if we allow that though. One particularly colorful and accessorized character could get 30~50 tags.

This isn't an issue on its own, it becomes an issue when it's redundant and fills the tag list with walls of monotone_x. The issue isn't how many tags are present, but the quality of those tags, which the random body part and monotone tags don't provide whatsoever.

I don't think going scorched earth on tagging colors except the very basics is a good idea here. The focus should be on reducing low quality and redundant color tagging.

oozeenthusiast said:
There are a lot of different pieces of clothing that could be tagged if we allow that though. One particularly colorful and accessorized character could get 30~50 tags.

I mean, maybe? but the opposite would be treating skirts and pants, and dresses and suits, and bracers and gloves, and helmets and baseball hats, and scarves and neckties as more similar than they really are.

and certainly don't want to entirely do away with specificity all in one go like that.

also, I want to say that it'd make tagging a bit easier with the implications.

dba_afish said:
I mean, maybe? but the opposite would be treating skirts and pants, and dresses and suits, and bracers and gloves, and helmets and baseball hats, and scarves and neckties as more similar than they really are.

and certainly don't want to entirely do away with specificity all in one go like that.

also, I want to say that it'd make tagging a bit easier with the implications.

I would say that you could search something like blue_bottom skirt but...

nimphia said:
Yes please, I like clothing tags like this. I am very interested in fashion. Things like the tags for fingerless_elbow_gloves bug me though. But things like pink_dress, green_socks, yellow_bracelet, I think are good and perfectly fine

This isn't an issue on its own, it becomes an issue when it's redundant and fills the tag list with walls of monotone_x. The issue isn't how many tags are present, but the quality of those tags, which the random body part and monotone tags don't provide whatsoever.

I don't think going scorched earth on tagging colors except the very basics is a good idea here. The focus should be on reducing low quality and redundant color tagging.

Yeah, you are right. It's easier to alias the tags later if they prove to be a problem than to bring them back if they prove to be useful.

Speaking as That Guy Who Obsessively Adds Color Tags To Things, yeah that list is way too much. The only time specific body parts ought to be tagged is if (A) they're a different "material" from the overall body (e.g. noses, tongues, and pawpads on a furry character; beaks and talons on a bird; etc.) or if they're a separate color from the surrounding area in such a way that already has its associated marking tag (so X_tail_tip for dipstick_tail, X_feet for socks_(marking) etc.) I've run across posts tagged with things like "white_balls" and thought "Eh, 'white_countershading' + 'countershade_crotch' ought to do it" but left it alone because I wasn't sure where we stand on that.

Also there's probably no need to keep the two-tone_X tags around; who cares about the number of colors something has? Plus just today I saw an orange and white fox tagged as "two-tone_fur" despite also having brown hands, feet, and ears, so I'm not sure people even know how to use it properly.

errorist said:
Speaking as That Guy Who Obsessively Adds Color Tags To Things, yeah that list is way too much. The only time specific body parts ought to be tagged is if (A) they're a different "material" from the overall body (e.g. noses, tongues, and pawpads on a furry character; beaks and talons on a bird; etc.) or if they're a separate color from the surrounding area in such a way that already has its associated marking tag (so X_tail_tip for dipstick_tail, X_feet for socks_(marking) etc.) I've run across posts tagged with things like "white_balls" and thought "Eh, 'white_countershading' + 'countershade_crotch' ought to do it" but left it alone because I wasn't sure where we stand on that.

Also there's probably no need to keep the two-tone_X tags around; who cares about the number of colors something has? Plus just today I saw an orange and white fox tagged as "two-tone_fur" despite also having brown hands, feet, and ears, so I'm not sure people even know how to use it properly.

Two_tone_* also kind of frustrates me, tbh. Especially on clothing. I usually don't bother with it and just add multicolored_* instead.

For the technical hurdles of a "Color" tag group, note that e621 just added a "Contributor" tag group: https://e621.net/forum_topics/54179

So the process to add that tag group can be repeated, it should be easy to make the change on a technical level, check the commit that added "Contributor" and repeat those line-of-code additions for "Color".

aaronfranke said:
So the process to add that tag group can be repeated, it should be easy to make the change on a technical level, check the commit that added "Contributor" and repeat those line-of-code additions for "Color".

Just because it could be done once doesn't mean it can be easily repeated. Actually, the Lore category was added soon after the site was rewritten, but adding the Contributor category wasn't as simple as repeating what was done for the Lore category, so creating some kind of Color category (whatever such a category would look like or contain) wouldn't be as simple as repeating what was done for the Contributor category. FWIW, the Contributor category was added by a now-former staff member, and as we see with the recent alias requests, there are still bugs with it. Hopefully they or someone else is still capable of fixing it, but creating this category was obviously not simple or smooth.

watsit said:
FWIW, the Contributor category was added by a now-former staff member, and as we see with the recent alias requests, there are still bugs with it. Hopefully they or someone else is still capable of fixing it, but creating this category was obviously not simple or smooth.

I heard that the bug isn't with the code, but with things running on outdated code, which means it's purely up to Dari/Cinder to fix. This was also the reason artverified:false was borked on arrival.

My main concern is the number of category changes needed if we are ever to create a colors category. Admins are kinda overloaded already and AIBURs eat through them like candy