Topic: Aliasing Away ***_threesome

Posted under Tag Alias and Implication Suggestions

The bulk update request #7680 is pending approval.

create alias mmg_threesome (1) -> threesome (56158)
create alias mfa_threesome (3) -> threesome (56158)
create alias ffg_threesome (1) -> threesome (56158)
create alias ffm_threesome (1470) -> threesome (56158)
create alias mmf_threesome (1332) -> threesome (56158)
create alias fff_threesome (13) -> threesome (56158)
create alias mmm_threesome (22) -> threesome (56158)
create alias mff_threesome (21) -> threesome (56158)
create alias male_female_male_threesome (0) -> threesome (56158)

Reason: Fully expanded out into all possible combinations, this is some 325 tags. Only mmf & ffm are actually populated, presumably by either specific users adding the tags across posts, or tags being brought over from other sites/boorus. These should either be aliased away or established. They've been hanging inbetween for years now. This compact form of one letter also bumps into issues with andromorph/ambiguous and male/maleherm.

definitelynotafurry4 said:
Would consider agreeing if/when https://e621.net/forum_topics/43865 is resolved. If it's denied, I think these should be established.

If they are established they need a new format, the one letter format poses problems. if we use m for male, mh for maleherm, and h for herm, is mhm_threesome maleherm, male, and a missed third? male, herm, male? it can be easy to miss a single letter like that, and all of a sudden the post is completely incorrect

The tags like male/male/female could be an okay target for something like this, though that will definitely be misused. I'd opt for spelling out the genders if we want the _threesome suffix

this is another case for ratio tags.

in theory male_majority threesome m/mh should, more or less, return the same results as male/male/maleherm_threesome but without the creation of like 94 new active tags (and what like 512, if you include aliased tags?) and instead just needing 8/9 new tags.

This is definitely one of the worst approaches we could take to this. If we include all 8 genders (male, female, intersex, gynomorph, andromorph, herm, maleherm & ambiguous_gender) that's actually 512 combinations.

Gender tags from 1 to 10 would (obviously) only be 80 tags. A much lower amount.

Now what if people wanted to search for foursomes? There would be 4096 *_foursome combinations. Obviously completely unsustainable, and will never be permitted, leaving us with a seemingly arbitrary threesome cut-off.

But with character counts? Still the same 80.

Ratios less again, but I don't think 80 tags is that big of a deal for the search precision it gives. The male/male, male/female tags and etc. already use 64 different tags, along with another 64 from *_penetrating_* set of tags, or over 1200 from the holding_* tags. Even with 1200 tags nobody's ever really pointed out that one being a problem, have they?

Updated

faucet said:
This is definitely one of the worst approaches we could take to this. If we include all 8 genders (male, female, intersex, gynomorph, andromorph, herm, maleherm & ambiguous_gender) that's actually 512 combinations.

You're over counting some of them, it's not 8^3
as mff, fmf, and ffm all count as 1

It's 8 + 2*28 + 56, which is 120

Oh, just realised because of dba afish that this is basically choosing 3 genders out of 8, with replacement. So it's 10 choose 3, which is 120.
So the number of combinations of an n-some with m genders is m+n-1 choose n

Updated

snpthecat said:
You're over counting some of them, it's not 8^3
as mff, fmf, and ffm all count as 1

It's 8 + 2*28 + 56, which is 120

Argh, math, my weak point. Either way, it's still significantly more than gender counts just for threesomes alone and it's not really sustainable to go any higher.

---

Another point I forgot to mention in my original post - threesomes are only for sex. It would be useless for posts that don't contain sex, but gender counts would still work just fine.

post #4486557

  • 5_gynomorphs 1_male
  • male gynomorph gynomorph_majority
  • this isn't a threesome or sex, so the ***_threesome tag group is useless here, despite involving more tags than the other options

My original count didn't include intersex and incorrectly accounted duplicates, resulting in 325

const genders = ["m", "f", "g", "a", "h", "mh", "am"];
const list=[];
for(const g1 of genders) {
  for(const g2 of genders) {
    for(const g3 of genders) {
      list.push(`${g1}${g2}${g3}_threesome`)
    }
  }
}

const total = Array.from(new Set(list)).length; // 325

There's multiple problems with that code, so using a bit of help from chatgpt and @SNPTheCat (because I'm dumb and don't like math), it gave me this:

const genders = ["male", "female", "gynomorph", "andromorph", "herm", "maleherm", "intersex", "ambiguous"];
const combinations = new Set();

for (const g1 of genders) {
    for (const g2 of genders) {
        for (const g3 of genders) {
            const combination = [g1, g2, g3].sort().join("_") + "_threesome";
            combinations.add(combination);
        }
    }
}

const total = combinations.size; // 120

So as far as I can see, the total count is 120 (I don't think we need intersex honestly, without it we'd have only 84, but I know we can't really go without it at this point)

Updated

donovan_dmc said:

There's multiple problems with that code, so using a bit of help from chatgpt and @SNPTheCat (because I'm dumb and don't like math), it gave me this:

const total = combinations.size; // 120

Hmmm, it does look slightly suspicious to an outside observer; I gave you help and the corrected code now spits out 120

Updated

the normal paring tags are triangle numbers, the formula for which is (x*(x+1))/2 which gives us 28 for 7, accounting for i/m, i/f, i/?, and i/i gives us a total of 32 active pairing tags.

adding another factor would make it a tetrahedral number, the formula for which is (x*(x+1)*(x+2))/6 which gives us 84 for 7, and adding the 10 intersex ones it's 94 total active threesome tags.

  • 1