[..]

Personally I hate nested implications because there is no quick indication what implicates what, and having things like female_alligator & female_reptile on the same image, with both implications to alligator, reptile, female, would offer a lot of room for errors.

We'd very much rather have a way to add tag groups, for example based on characters, where you add tags to a specific character in the image, then go through all characters and have the server "add" the proper tags to the full tag-list. Once the database and site actually are able to handle a system like that we wouldn't require any additional upkeep with implications and aliases at all. (Ignoring the initial start where we'd have to actually make those character tags on a million old posts.)

• complexity of tags increase exponentially as an individual set increases
• e.g. [8! gender tags] * [any set with >1 element] = 40 230+ commutative possibilities
• 8! * [5! body types] = 4 838 400 commutative possibilities
• 8! * [10+ species tags] = 146 313 216 000+ possibilities

NotMeNotYou said:

Let's go at this from the other angle, we do have something planned in the far future to have character specific tags, but not soon and not your system, for a very simply reason: the sheer amount of tags people have to memorize.

We have 7 gender tags and ambiguous_gender, further we have 5268 species tags at the moment, but most of that is probably redundant because overlap, lorespecific names and what have you.
However, let's say 50% of those tags are useful, we now need to make designations for those 2600 tags, for all 8 possible genders (male, female, herm, maleherm, dickgirl, cuntboy, neuter, ambiguous), we are now looking at 20,800 new tags with 2 implications each, so 41,600 implications.

Even if we assume you have 100% determination to get this done, and you make no mistakes with suggesting them, we still face issues of the tag names themselves. Do we go with "normal" names like cow? Or uniform with bovine_female to accept things like bovine_dickgirl?

Personally I would love to have a filter like that where you can make sure that only a specific gender of a specific species shows up, or specific character of a specific gender, but your system is simply nonviable with the sheer amount of work required for the initial setup as well as prolonged upkeep.

[..]
Personally I hate nested implications because there is no quick indication what implicates what, and having things like female_alligator & female_reptile on the same image, with both implications to alligator, reptile, female, would offer a lot of room for errors.

Genjar said:

abadbird said:
The project of linking tags on pre-existing posts would be tremendous and likely never completed, of course.

It definitely wouldn't. Most posts are missing even the basic tags. I've been working on adding the penis tag to relevant posts; on estimate, that's missing from at least third of the posts that should have it. Here's an example: post #56669, which was rated questionable for six years.

I've got about 2% done in a month. And that's just one basic project out of hundreds: for instance, while I've been doing that, the number of posts that are missing all species tags has grown by about 5000. The amount of posts that are lacking the form tags (anthro, feral, humanoid, etc) is at 120000 and growing rapidly. At least half of all posts are missing the background tags. Gender and orientation are missing from vast number of posts. And so on.

Nobody's got time to even keep up with the already existing tags. Starting to tag things per character will only take time away from tagging the basics, and therefore it will make the overall tagging worse.