I'd like to make a request to add a genitalia section to the upload page which will make it easy to properly tag genitals.
I'm requesting this primarily because I see just how often tags like humanoid_penis are missed - it's a tag that I often want to exclude from my searches but despite doing that, the page is still completely full of humanoid penises. Of course it's not the only genital tag that's often missed and I think that having a dedicated section on the upload page could significantly improve the situation - it makes sense too since genitals should be tagged on the vast majority of explicit posts so I think it would be very often used.

Now for my idea - I think it could be just a couple of main category buttons like penis | pussy | anus | cloaca | balls
Then after selecting one - for example penis - the category would expand into sub categories with different penis types for example humanoid_penis, equine_penis, canine_penis etc.
After selecting the type it could expand to also include individual parts that are often found on that particular type of genital, so in the case of humanoid_penis it would be glans, frenulum, foreskin, circumcised, urethra - It's possible that this third level would introduce too much complexity so I consider that one optional.

In my opinion this would make it much less likely for people to forget tagging what is arguably some of the most important tags on the website and could also improve the accuracy and consistency of tagging some of the lesser known tags for individual genital parts.

And just to prove how enormous the scale of the problem is - here's the search for penises that excludes every single significant penis type - which theoretically should yield very few results and yet it returns around 120 thousand posts which only go a year back (maximum range of 750 pages permitted by the search at 160 posts per page): penis -humanoid_penis -animal_penis -tapering_penis -hybrid_penis -unusual_penis -mechanical_penis -ridged_penis -ribbed_penis -spiked penis -censored_penis -nubbed_penis -sentient_penis -avian_penis -ovipositor_penis -arthropod_abdomen_penis -spiral_penis -reptile_penis -gastropod_penis -frilly_penis -vacuum_penis

Doing a more detailed count using this technique combined with the date:x_yesteryears_ago I was able to come up with a very specific number:
49% of posts tagged with 'penis' never get tagged any penis type. Since the beginning of 2013 there are now 1 192 698 posts tagged with 'penis' and of those 578 751 or 48,52% don't have any penis type tagged.

The lowest this percentage has been was in 2019 with "only" 43% posts lacking a penis type (45 108 out of 104 341).
The highest it has been is actually right now, followed closely by 2022 (52.3% and 51.5% respectively). This is no doubt due to the fact that over time some of the posts that are wrong now are going to be fixed. In fact if we look at just last 7 days the percentage is at 53.5% - so with with the least time for the community to correct the mistakes we're the closest to the true error rate 'at the upload stage' - which is no doubt a bit higher than that.

+1 - This is one of many things I'd like to see the upload page have at some point in some way or form.

I proposed a similar feature once where the male/female, female/female, etc. tags were expandable rather than always being offered (topic #37844) so if one of these features were added the other one would probably be easy to add at the same time.

faucet said:
+1 - This is one of many things I'd like to see the upload page have at some point in some way or form.
I proposed a similar feature once where the male/female, female/female, etc. tags were expandable rather than always being offered (topic #37844) so if one of these features were added the other one would probably be easy to add at the same time.

Oh that also sounds like a good idea! And just to answer some of the concerns raised in the other thread that might pop up here as well - while implementing these features might be a bit of work, it's something that has to be only done once and it will become extremely beneficial in the long run by improving the accuracy of all the posts going forward and making the mods/taggers have less work fixing the tags on posts.
Another concern was about edge cases - it's a very common argument against implementing anything new - 'but what about that very specific and unlikely case where said thing wouldn't work?'. I say that the fear of edge cases shouldn't keep us from implementing overall positive features and edge cases aren't the end of the world anyway - it's possible to deal with them as they come. Besides the proposed features don't break any functionality of the upload page - everything could still be tagged manually if one so chooses.

Also while we're at it I would like to see the 'Related: tags' feature improved - currently it displays related tags for only the first 25 tags which is a pain when I'm tagging a post with over 100 tags and can't see related tags for the majority of them.

Is bumping allowed here?

angrydraconequus said:
Is bumping allowed here?

angrydraconequus said:

angrydraconequus said:
Is bumping allowed here?

Yeah

I've had that exact same gripe with nobody tagging humanoid penises, so I'm a +1 to this or some other way to make genitalia tagging more frequent.

I was like "HUUUUH -_-" but seeing y'all +1, guess there is a demand.

closetpossum said:
I was like "HUUUUH -_-" but seeing y'all +1, guess there is a demand.

Yeah I think the sheer number of badly tagged posts speaks for itself. I edited the first post with more detailed and accurate calculations. But here's some additional info:

Out of all the posts tagged with penis uploaded in the past 10 years, 48.5% don't include any penis type. 48.5% is the average error, but recently it's been even higher. Here's the year by year breakdown:
2023: 52.3% (71 318 out of 136 360)
2022: 51.5% (111 492 out of 216 342)
2021: 50.2% (87 035 out of 173 383)
2020: 47.9% (65 327 out of 136 524)
2019: 43.2% (45 108 out of 104 341)
2018: 46.9% (48 144 out of 102 755)
2017: 44.0% (40 702 out of 92 547)
2016: 44.5% (36 293 out of 81 550)
2015: 48.9% (31 460 out of 64 363)
2014: 48.2% (21 405 out of 44 412)
2013: 51.0% (20 467 out of 40 121)
Total: 48.5% (578 751 out of 1 192 698)

We're getting almost 10 thousand incorrectly tagged posts like that EVERY MONTH. Fixing that many posts would be a monumental task. Fixing that many posts while there's new ones constantly being uploaded is pretty much impossible. The only solution is to minimize the error rate at the upload stage which is why I proposed a solution that would definitely help.

angrydraconequus said:
Yeah I think the sheer number of badly tagged posts speaks for itself. I edited the first post with more detailed and accurate calculations. But here's some additional info:

Out of all the posts tagged with penis uploaded in the past 10 years, 48.5% don't include any penis type. 48.5% is the average error, but recently it's been even higher. Here's the year by year breakdown:
2023: 52.3% (71 318 out of 136 360)
2022: 51.5% (111 492 out of 216 342)
2021: 50.2% (87 035 out of 173 383)
2020: 47.9% (65 327 out of 136 524)
2019: 43.2% (45 108 out of 104 341)
2018: 46.9% (48 144 out of 102 755)
2017: 44.0% (40 702 out of 92 547)
2016: 44.5% (36 293 out of 81 550)
2015: 48.9% (31 460 out of 64 363)
2014: 48.2% (21 405 out of 44 412)
2013: 51.0% (20 467 out of 40 121)
Total: 48.5% (578 751 out of 1 192 698)

those numbers sound pretty bad, but how many of those posts have penises that can't be sorted into any of the existing categories, either because they are partially obscured or because there is no category that they would properly fit in?

+1 for the feature though, I'm always surprised how often I come across posts that are missing basic tags like penis.

themasterpotato said:
those numbers sound pretty bad, but how many of those posts have penises that can't be sorted into any of the existing categories, either because they are partially obscured or because there is no category that they would properly fit in?

+1 for the feature though, I'm always surprised how often I come across posts that are missing basic tags like penis.

Yeah, simply being unable to categorize is a large part of the problem. Fictional species in particular often have penises that don't fit any of the existing boxes. Ideally features would be tagged as well.

themasterpotato said:
those numbers sound pretty bad, but how many of those posts have penises that can't be sorted into any of the existing categories, either because they are partially obscured or because there is no category that they would properly fit in?
+1 for the feature though, I'm always surprised how often I come across posts that are missing basic tags like penis.

scth said:
Yeah, simply being unable to categorize is a large part of the problem. Fictional species in particular often have penises that don't fit any of the existing boxes. Ideally features would be tagged as well.

There are plenty of categories for even the most obscure types. Besides the majority (56.4%) fall into the humanoid_penis category, most of the remainder fall into the animal_penis category (43.1%). The only other significant one is tapering_penis at 4.6%. Then it's unusual_penis at 0.7% and all the others fall quickly to 0.6, 0.3, 0.2 and 0.1%.*

Penises that don't fit any existing boxes aren't statistically significant, and most of them should still be caught by the unusual_penis tag.

Now the situation where the penis type is unclear is much more common - however if you have a quick look at the search results of penis -humanoid_penis -animal_penis -tapering_penis -hybrid_penis -unusual_penis -mechanical_penis -ridged_penis -ribbed_penis -spiked penis -censored_penis -nubbed_penis -sentient_penis -avian_penis -ovipositor_penis -arthropod_abdomen_penis -spiral_penis -reptile_penis -gastropod_penis -frilly_penis -vacuum_penis then it's pretty clear that a lot, if not most posts have pretty clear view that should be easily categorisable. So lets say 50% of them have that problem - then just divide all my previous values in half - that still leaves a significant problem. And I suspect it might actually be more than 50% but I'm not gonna go calculating that because I think it's pretty clear that the problem is there and that it's massive.

*The percentages add up to more than a hundred because there's overlap on posts - a post can have multiple penis types on it.

Oh and that's only the penis side of things. Pussy is even worse when it comes to categories. Currently there are 748 811 posts tagged pussy while there are only 39640 posts tagged animal_pussy, 14382 posts tagged humanoid_pussy and 400 posts tagged unusual_pussy. Posts tagged with any pussy type make up only 7.3% of all posts tagged with pussy. Although I suspect that the vast majority of the posts should be tagged with humanoid_pussy and the vast majority of posts that have any animal pussy present are correctly tagged as such - I think people tag that way more diligent than they do for penises. I think the problem here is just that the humanoid_pussy tag is so often omitted. So I'd say that humanoid_pussy tag is missed on at least 80% of posts which is also a huge problem.

snpthecat
Yeah

Aight, last message was 24 days ago so throwing a bump in here because this is a good idea IMO

If possible, disallow lone "penis"/"pussy" tags that aren’t followed by a more specific tag (animal penis/pussy, humanoid penis/pussy), to ensure none of them get missed.

Bump

+1!

Changes to the site itself tend to be slow-going, as we only really have one developer working on stuff, and there's a lot on the to-do list. But e621 is open source, so anyone willing and able enough can try submitting a pull request for the desired feature on Github to potentially get it implemented more quickly.

In the meantime, if anyone does want to try to help clean up these posts, there is a penis type tagging project on TagMe.

I know it's not what's being asked, but I made a userscript for this that you can download if you have tampermonkey here (hardcoded version, old no longer updated, see below version)

If you want to see the source of the gist, click here

I don't have anything to upload, so it's untested, but I see no reason for it to not work, if you test this out, let me know if it's working good. It only shows when the post rating is tagged as explicit

The tags and their sub tags were taken from the wiki under the "types" for both penis and pussy.

This is a fully working example of what this could potentially be if added to the site, but it won't have widespread adoption, so it's more of a QOL thing.

Newest updated version:

If you want a version that auto fetches tag implications for the penis/pussy types, and therefore contains more tags and dynamically updates, then download this version instead. When you click on a tag, it will fetch the implications to it and add them. I only fetch after the second tag, this is because way too many tags implicate the top level tags that aren't related to a single genital (think penis_piercing, which implicates penis, we don't really want to see that here). This version does have some lag as it has to request the implications from the e621 api the first time you click on a button. But the additional half second of lag is worth the QOL, imo.

This version has become a more general tag helper UI, rather than just an explicit tag helper. It now can help you add art types like digital_media_(artwork).

Example video can be found here: https://gist.github.com/DontTalkToMeThx/74a4a271e19cbe14195cd2c6671736ba#file-readme-md

definitelynotafurry4 said:
I know it's not what's being asked, but I made a userscript for this that you can download if you have tampermonkey here (hardcoded version)

If you want a version that auto fetches tag implications for the penis/pussy types, and therefore contains more tags and dynamically updates, then download this version instead. When you click on a tag, it will fetch the implications to it and add them. I only fetch after the second tag, this is because way too many tags implicate the top level tags that aren't related to a single genital (think penis_piercing, which implicates penis, we don't really want to see that here). This version does have some lag as it has to request the implications from the e621 api the first time you click on a button. But the additional half second of lag is worth the QOL, imo.

If you want to see the source of the gist, click here

I don't have anything to upload, so it's untested, but I see no reason for it to not work, if you test this out, let me know if it's working good. It only shows when the post rating is tagged as explicit

Here's an example (gif): https://gist.github.com/DontTalkToMeThx/74a4a271e19cbe14195cd2c6671736ba#file-readme-md

The tags and their sub tags were taken from the wiki under the "types" for both penis and pussy.

This is a fully working example of what this could potentially be if added to the site, but it won't have widespread adoption, so it's more of a QOL thing.

This looks nice.

Could possibly add anus, cloaca, and genital_slit? There are some subtypes of cloaca as well - horizontal_cloaca, vertical_cloaca, triangular_cloaca round_cloaca that would be nice to have.

scaliespe said:
This looks nice.

Could possibly add anus, cloaca, and genital_slit? There are some subtypes of cloaca as well - horizontal_cloaca, vertical_cloaca, triangular_cloaca round_cloaca that would be nice to have.

Added to this version

Example video is here: https://gist.github.com/DontTalkToMeThx/74a4a271e19cbe14195cd2c6671736ba#file-readme-md

I've made this script into its own forum topic: https://e621.net/forum_topics/40752

Latest version of this is more general, and you can add your own tags with a small amount of JS knowledge. I've put a comment that starts on line 289 that explains how to add new tag types if you want to do it yourself.