the sizes tagged with each breast size-tag seem to have shifted. because of that I came up with a couple sizing scales for breasts based on community tagging habits. opinions? agree/disagree? nitpicks? it'd be good to hear where people think the break points should be.

size scaling on this site seems to follow an odd curve. small breasts are rather small. medium is rather small to rather big(anything that isn't mostly one way or the other gets put here), and big is rather big(like 34f irl) to head sized.

the size categories for flat_chested, huge_breasts, and hyper_breasts seem to be pretty stable and well defined. it's the break points between small_breasts, medium_breasts, and big_breasts that could use a bit of work

when creating these scales I tried to go by the general break points used in what has already been tagged on the site. based on those general community tagging patterns the scale seems to have shifted a bit from the scaling sizes that are currently shown in the wiki scale. imo the tags should apply the way the community is tagging them meaning the current wiki scale should be updated to reflect this shift in tagging.

questions:

• should edge cases be tagged with two size tags? (especially relevant for the medium_breasts tag)
• how much spillage/ambiguity is ok between categories?

-

scale 1:

• note: cup sizes in this scale are US cup sizes with a 34 inch band size assumed as average. vague language is purposefully used surrounding these sizes as they are used as an additional descriptive aid in this scale, not the de-facto measuring metric.

-

scale 2:

• even in a scale like this someone's going to say "but it's exactly on the cut off line? what now?". if you make the end points different characters it would be "but the size is in between those characters...". imo at least some amount of overlap is always going to be involved when a tagging system involves a gradient, a nature of the beast sort of thing. (furry pun? furry pun :3)

-

current size scale:

Watsit said:
You've given some examples, but still haven't explained how you came to those conclusions. The problem being discussed here is objectively categorizing images in a useful way, but all your post is saying is that small is equal to "A to large-B/small-C", etc, and that doesn't help when the size itself is debatable. If we could accurately measure cup sizes in drawn images, this wouldn't have ever been an issue.

I don't have a suggestion for improving categorization beyond using personal judgement (the current method), which is why I think an ambiguous size tag would work better than a medium size tag. That's my suggestion.

thanks ^_^

More depth in explaining methodology:

Watsit said:
You've given some examples, but still haven't explained how you came to those conclusions.

sneezer22 said:
I came up with a couple sizing scales for breasts based on community tagging habits.

going through like 10+ pages of each size looking for the high and low of each size category tossing out obvious outliers. the hardest to get a feel for was medium since its under tagged and just looking at breasts that haven't been tagged for size returns all sizes of breasts. so part of the way I looked for the edges of medium was by using the high of small and the low of big.

while going through each category i took out a few pics that seemed to appropriately represent the high and low break points of a category and what each category looks like most commonly. tbh with this kind of scale the examples feel like a really important part of the guide.
---

---

Watsit said:The problem being discussed here is objectively categorizing images in a useful way, but all your post is saying is that small is equal to "A to large-B/small-C", etc, and that doesn't help when the size itself is debatable.

sneezer22 said:
note: cup sizes in this scale are US cup sizes with a 34 inch band size assumed as average. vague language is purposefully used surrounding these sizes as they are used as an additional descriptive aid in this scale, not the de-facto measuring metric.

the main points used for describing the bounds of each category were the pictures themselves and the break points I came up with. the cup sizes could be taken out but I like having more ways of describing the bounds of each category(specifically between small and big). first I put together break points and pics to go with them, then from that I tried to come up with descriptions for each break point.

the break point descriptions were created to be straightforward and unambiguous while remaining on point for where the cutoffs between each category are

FLAT

-----
break-point: has breasts
-----

SMALL

-----
break-point: do they fill a handful? do they have more than a bit of hang to them?
-----

MEDIUM

-----
break-point: larger than softball sized, can they give better than ok paizuri?
-----

BIG

-----
break-point: larger than the character's head
-----

HUGE

-----
break-point: larger than a watermelon or beach-ball, non-realistic
-----

HYPER

---

---

If we could accurately measure cup sizes in drawn images, this wouldn't have ever been an issue.

that was the idea behind the second scale, a higher level of specificity for the region of the scale between small and big, not based on a clothing size that's hard to visualize and pin down but on a scale where you can go by what you see with a dense spectrum of examples to use as reference. the same break points were used in both scales and the idea was that the two scales would line up and compliment each other.
vvvvvvvvvvvvvvvvvvvvvvvvvvv
--------------------------------
Scale 2:

post #174655

small_breasts: from Digit to Xeanica
medium_breasts: from Xeanica to Angle
big_breasts: from Angle to Nina
huge_breasts: from Nina to Tila
--------------------------------

Watsit said:
using personal judgement (the current method), which is why I think an ambiguous size tag would work better than a medium size tag. That's my suggestion.

ambiguously_sized_breasts

still gotta get around to making a wiki for that. seems useful

Watsit said:
The tag I'd propose isn't just for obscured breasts, but ambiguously-sized breasts. They may be obscured, or could be in full view but perspective may be wonky, the shadows or line work can throw perception off

and obscured_breasts, I see opportunities for tagging that one all the time too

"sneezer22":said:
if a picture shows a character with breasts that are obscured by: clothing, their own body, someone else's body, an object in the scene,... literally anything that stops the viewer from getting a full view of them, especially if its to the point it makes tagging size difficult

sneezer22 said:
going through like 10+ pages of each size looking for the high and low of each size category tossing out obvious outliers.

Categorizing based on what's already categorized doesn't sound so helpful. A person shouldn't have to comb through previous posts to figure out whether what they're currently posting counts as small, medium, or big, instead of some kind of rough objective measure by looking at just the picture in question, especially considering existing posts can be incorrectly tagged giving a false impression of what counts or doesn't. And it's still up to personal interpretation.

Besides, TWYS means not using external information, including other pictures. For example, we don't tag a character's gender based on how other images of the character were tagged. Similarly, we shouldn't tag breasts' size based on how other images of breasts were tagged. So far you've just given us your personal opinion based on some random assortment of existing images.

sneezer22 said:
the hardest to get a feel for was medium since its under tagged and just looking at breasts that haven't been tagged for size returns all sizes of breasts. so part of the way I looked for the edges of medium was by using the high of small and the low of big.

Exactly part of the problem of using existing images (how do you compare to other medium size images to know if it fits, if there aren't images tagged medium size to compare to?), and why there needs to be some rough objective guide or measure. We're also dealing with a pretty loose scale with rather blurry lines, there aren't "edges" to speak of.

sneezer22 said:
the break point descriptions were created to be straightforward and unambiguous while remaining on point for where the cutoffs between each category are

And again, we're talking about pictures that don't always (some might say don't usually) show breasts in a manner that makes their size straightforward and unambiguously apparent. For example, someone could make a reasonable argument that the way the shadow drops below the chest makes it look like a character has medium breasts, while another person says it's just their posture or light angle creating larger shadows under small breasts. Someone could make the reasonable argument that the perspective of the torso makes it apparent the breasts are fairly big for a character, while another says they don't look that big and the perspective makes it hard to compare to other big breasts.

I'm still in strong support of aliasing away medium_breasts. The sizing scale is too loose to have something between small and big, as it doesn't cleanly fit into a scale hierarchy (e.g. how hyper breasts are still considered huge breasts, which are still considered big breasts).

sneezer22 said:
ambiguously_sized_breasts

still gotta get around to making a wiki for that. seems useful

My suggestion would be for breasts that don't seem small enough to qualify for small_breasts and don't seem big enough to qualify for big_breasts. If you can't justify having one of those two tags, consider it ambiguous. Note that huge_breasts and hyper_breasts both imply big_breasts, so if you're simply not sure if they count as huge or hyper specifically, they still count as big for sizing purposes, not ambiguous.