So let's take a look at post #763074.
There is one species present on the image.
There are 10 (ten) species tags on the image at the time of writing this post.

I do understand the idea, but this is getting quite ridiculous. Can we cut down on those?

Could you please explain why being able to search for herding dogs is somehow damaging your e621 experience?

IHateRule34 said:
So let's take a look at post #763074.
There is one species present on the image.
There are 10 (ten) species tags on the image at the time of writing this post.

I do understand the idea, but this is getting quite ridiculous. Can we cut down on those?

Hi there! I'm the one mostly responsible for domestic dog breed tag organization. I do admit it can get a little bloated, and I might end up trimming a few in the future possibly by moving around/getting rid of superfluous ones perhaps. However, even then, a lot would still exist.

I don't see an inherent problem with having more categorized dog breed tags, however.

IHateRule34 said:
So let's take a look at post #763074.
There is one species present on the image.
There are 10 (ten) species tags on the image at the time of writing this post.

I do understand the idea, but this is getting quite ridiculous. Can we cut down on those?

Personally, I don't see why it matters if there's two species tags or twenty, so long as they're correct. For the most part, with that post, they're all implied from collie, so it's not like you need to remember them all for tagging purposes either. They also allow for people to be as specific or vague with their searches as they choose to.

When reading posts talking about species tags, I've often thought it's like complaining a dictionary has too many words in it, or that it shouldn't contain words the poster is unfamiliar with.

I'd get it if you had to add them all manually, had to remember them all, or there was actually a downside to having them. But as far as I can tell it doesn't make a difference, and allows people who want to search for specific things to do so.

As long as it does not slow down searches or make the site hard to browse, what's the harm.

When it comes to tags, the more the better, and I don't think we've reached the point yet where the tags are bordering on redundancy.

Well I get the synonims, but are all those variations of "canine"/"canid"/"canis" necessary? Can I see the data that was behind the decision to introduce all those? Were there really that many failed searches for "canid"/"canis" and the like?

IHateRule34 said:
Well I get the synonims, but are all those variations of "canine"/"canid"/"canis" necessary? Can I see the data that was behind the decision to introduce all those? Were there really that many failed searches for "canid"/"canis" and the like?

I didn't know the difference between canid/canis and figured they were the same, but that's one reason there's a wiki.

If all canids were canine, then I'd agree to remove some, but the wiki lists two other sub-families that are canid but not canine.

And I think it's less to do with data for people searching things that aren't tagged, and more to do with being correct. So that *if* someone wants a specific species, or subfamily, then they can search them.

Honestly, when I look at all those tags I'm just impressed that Ratte managed to sort them all like that. I feel it must have taken a long time to go through them all and include wiki articles for each one as well.

Off topic, but I feel this is a very similar thing to the tagging of ratios. Some people see them as stupid and pointless, and others see them as useful, or at least see how they could be useful to somebody else. But that's for another thread.

Again, with them all being aliased properly, so the site does all that part for you, the only thing you get as an end user is a few more orange tags in the sidebar.

It goes back to my comment on it being like saying there's too many words in the dictionary. Just because you don't find them useful, or won't use half of them, it doesn't mean that others don't find them useful or won't use them.

This change has nothing to do with failed search attempts and everything to do with structural organization of species names. If someone searches mammal they should find all species that are classified as mammals, if someone searches for canis they should find all species of canis genus but not species of the lycalopex genus, and so forth.

Just hijacking this thread to ask a species tags related question. I hope that's ok.

I've noticed that older posts only get updated with the new, implied, species tags after they've been updated.

Could an admin cause an update to automatically add all implied tags retroactively? I feel it's a bit pointless having the tags if older posts don't get updated.

Pupslut said:
Just hijacking this thread to ask a species tags related question. I hope that's ok.

I've noticed that older posts only get updated with the new, implied, species tags after they've been updated.

Could an admin cause an update to automatically add all implied tags retroactively? I feel it's a bit pointless having the tags if older posts don't get updated.

what do you mean? implied tags get added automatically even to older posts.

hiekkapillu said:
what do you mean? implied tags get added automatically even to older posts.

I thought they did as well, but I've seen a lot of times while tagging ratios that several species tags get added as well, tags that are implied by others. I just presumed that I was wrong, and only aliases updated old posts.

I can't find a better example right now, but a few hours ago I added reaction_image to post #100802, and pastoral_dog and sheepdog were also implied.

Looking at the tag history, several others were implied in the edit before mine as well, where it looks like only ace_attorney_(series) was changed to ace_attorney and dog was removed.

Pupslut said:
I thought they did as well, but I've seen a lot of times while tagging ratios that several species tags get added as well, tags that are implied by others. I just presumed that I was wrong, and only aliases updated old posts.

I can't find a better example right now, but a few hours ago I added reaction_image to post #100802, and pastoral_dog and sheepdog were also implied.

Looking at the tag history, several others were implied in the edit before mine as well, where it looks like only ace_attorney_(series) was changed to ace_attorney and dog was removed.

that is just because the application of implicated tags is not attached to any account, and the added implied tag gets stapled in the tag history on whoever changes the post next. its purely cosmetic thing that affects tag history only and the implied tag was there already.

lets take for example
post #5654
as you can see, it has canis and canid tags. if you edit it now, those both will be attached on the edit you made.

hiekkapillu said:
that is just because the application of implicated tags is not attached to any account, and the added implied tag gets stapled in the tag history on whoever changes the post next. its purely cosmetic thing that affects tag history only and the implied tag was there already.

Oooh.. I didn't realise it worked like that. For a long while I've just presumed they needed to be updated for them to get applied.

Thanks for that, I appreciate the explanation.

As a small suggestion for the admins, if it's easy, it might be good to create a user like "implications updater" and have them all attached to that user. I don't suppose it'd matter, with it only being cosmetic, but it'd at least avoid confusion in future.

hiekkapillu said:
lets take for example
post #5654
as you can see, it has canis and canid tags. if you edit it now, those both will be attached on the edit you made.

After posting my last post I noticed you'd edited to add the example, I removed the mammal tag, knowing it'd be implied again, and sure enough the other tags show as me adding them, despite already being there.

Thanks again!

Pupslut said:
And I think it's less to do with data for people searching things that aren't tagged, and more to do with being correct. So that *if* someone wants a specific species, or subfamily, then they can search them.

the point was to get more clear-cut boundaries and expand possible options/flexibility with searching and blacklisting. it also doesn't really hurt you any to learn a new word. some people on here like finding species or group example for reference images given how our search system works, so why not try to make it a little better? it can also help save on tag search limits or unwieldy/complicated blacklist entries.

Pupslut said:
Honestly, when I look at all those tags I'm just impressed that Ratte managed to sort them all like that. I feel it must have taken a long time to go through them all and include wiki articles for each one as well.

for a lot of the weirder ones i did try to go out of my way to make wikis explaining what things were and why it was organized that way. there are a lot of things that get tagged with something even when it's not actually that something, or when a word is so dumbly defined depending on use that the term doesn't really mean anything, so i have to decide how we define it. a good example of this is the fox tag, which was very vague until we defined it as "anything of the Vulpes genus" which would therefore only include Vulpes species and exclude several animals people just commonly call foxes, such as the grey_fox.

IHateRule34 said:
Well I get the synonims, but are all those variations of "canine"/"canid"/"canis" necessary? Can I see the data that was behind the decision to introduce all those? Were there really that many failed searches for "canid"/"canis" and the like?

Let me help.

To pick a random branch:

Domestic dog - Is implied by a huge swath of dog breeds. It implies:
Canis - Contains MOST of the familiar members of the dog family, like wolves, coyotes, and jackals. However it does not include foxes and several other less-well known dog-family members. Implies:
Canine - contains everything in Canis, and also: Raccoon dogs, bush dogs, African wild dogs, foxes, and others. This also includes things like "werecanines" ... it implies:
Canid - contains all of the above, plus several extinct species, as well as a huge array of "dog-like-things" like Eevees, renamons, lucarios, etc

It's not about failed searches. it's about giving people more options.

If you want dalmatians, search dalmatian.

If you dont' care as long as it looks doggish, search Canid.

Look, I literally wrote about this yesterday in another thread:

? brown pelican 7
? pelican 63
? pelecaniform 251
? avian 62917
? bird 44046

brown pelican is a specific pelican.
a pelicaniform is a group of birds that include some animals like shoebills and herons and spoonbills. Full of that big-beak energy.

You want birds with big beaks and throats? Pelicaniform. seriously: pelecaniform -animal_humanoid There ya go.

That's the idea with all this.

Why are peopel so scared of learning a new word OR having to click on the little ? to remember the right word?

You can LITERALLY bookmark any search you like in your browser. You could save them in your profile too. (Protip: if you type a search somewhere and enclose it with {{ and }} it'll turn into a search. That text up there? {{pelecaniform -animal_humanoid}} ...

SnowWolf said:
Why are peopel so scared of learning a new word OR having to click on the little ? to remember the right word?

Imma be real here and say that using a "you just have to learn new words" argument really doesn't help sway people into liking the implication system.

It's much much better to just explain to them that all implications and aliases improve the searching experience for everyone while not requiring any extra effort from both the uploader and browser's standpoints.

You don't really have to learn terms such as canis or canid. If it doesn't make any difference to you then you can just tag/search common terms such as wolf and dog and the entire system takes care of it by itself. Literally the only side effects of having an extensive alias/implication system is that you can be much more specific in your searches whenever you want to

There is absolutely no downside for the user. You don't need to take any extra effort when tagging or searching for pics, it only gives people that want to much more granularity when using the site. I feel that this is a point that should be given much more attention instead of being diluted among other weaker ones

mabit said:
Imma be real here and say that using a "you just have to learn new words" argument really doesn't help sway people into liking the implication system.

It's much much better to just explain to them that all implications and aliases improve the searching experience for everyone while not requiring any extra effort from both the uploader and browser's standpoints.

You don't really have to learn terms such as canis or canid. If it doesn't make any difference to you then you can just tag/search common terms such as wolf and dog and the entire system takes care of it by itself. Literally the only side effects of having an extensive alias/implication system is that you can be much more specific in your searches whenever you want to

There is absolutely no downside for the user. You don't need to take any extra effort when tagging or searching for pics, it only gives people that want to much more granularity when using the site. I feel that this is a point that should be given much more attention instead of being diluted among other weaker ones

But the purpose of being able to click on a tag to search for all posts under that tag is lost if you don't know what the tag means.

PheagleAdler said:
But the purpose of being able to click on a tag to search for all posts under that tag is lost if you don't know what the tag means.

That really isn't the case

If you check out the examples I've pointed to, you'll see that despite all the implications that get added you'll still have the wolf and domestic_dog tags on the post

But let's disregard even that for a minute because what you said is still wrong even only considering the new "foreign" tags. You'd still be able to click on the tag and see the other posts tagged with it (and you're gonna get correctly tagged stuff since a lot of the tagging is doing the work by itself), the "purpose" of the link is still there and is even enhanced since you'll have more (and more granular) tags to search for

If I open a post and it's about a character/position I don't know, it's not gonna ruin my entire search experience and leave me stuck. It's just gonna be a tag that I didn't know about so I wouldn't be looking for it anyway, I could either ignore it or explore it by clicking the link, while without an extensive implication system would just not be possible

There's literally no downside to having more tags on a post (unless you're really trying to force an issue, like having the page so big it doesn't load or breaks formatting, but at this point we both know you're just arguing for the sake of it)

mabit said:
That really isn't the case

If you check out the examples I've pointed to, you'll see that despite all the implications that get added you'll still have the wolf and domestic_dog tags on the post

But let's disregard even that for a minute because what you said is still wrong even only considering the new "foreign" tags. You'd still be able to click on the tag and see the other posts tagged with it (and you're gonna get correctly tagged stuff since a lot of the tagging is doing the work by itself), the "purpose" of the link is still there and is even enhanced since you'll have more (and more granular) tags to search for

If I open a post and it's about a character/position I don't know, it's not gonna ruin my entire search experience and leave me stuck. It's just gonna be a tag that I didn't know about so I wouldn't be looking for it anyway, I could either ignore it or explore it by clicking the link, while without an extensive implication system would just not be possible

There's literally no downside to having more tags on a post (unless you're really trying to force an issue, like having the page so big it doesn't load or breaks formatting, but at this point we both know you're just arguing for the sake of it)

My mistake, I was referring to something different but also happening to the site concurrently. More tags has no downside, unless you want to get to the bottom of the sidebar sooner or something lol

PheagleAdler said:
More tags has no downside, unless you want to get to the bottom of the sidebar sooner or something lol

Except for putting strain on the servers.
The sheer number of tags is what broke the tag history search, and the size of the database puts major limits on which features can be implemented.

Genjar said:
Except for putting strain on the servers.
The sheer number of tags is what broke the tag history search, and so on. The size of the tag database puts major limits on which features can be implemented.

Bad design and the enormous quantity of tag changes over the last 10 years is what broke tag history search. Additional tags per post is negligible.

KiraNoot said:
Bad design and the enormous quantity of tag changes over the last 10 years is what broke tag history search.

Introducing enormous number of new tags and implications is what caused the number of tag changes to explode. It would've broken eventually, but it is definitely one downside to 'more tags'.

Genjar said:
Introducing enormous number of new tags and implications is what caused the number of tag changes to explode. It would've broken eventually, but it is definitely one downside to 'more tags'.

Implications and aliases don't generate tag history items per post like tag edits do. Only when posts are next changed do those changes appear in the tag history. So it didn't make any impact on that area of the site.