Topic: How to turn an alias into an implication?

I want to create a BUR which goes like:

unalias lucent_nargacuga -> nargacuga
implicate lucent_nargacuga -> nargacuga

which leads to:
Error: Antecedent tag must not be aliased to another tag (create implication lucent_nargacuga -> nargacuga)

The way I understand that, the problem is that I can't create an implication for a tag that is currently aliased to something. Do I have to do a BUR where request to remove the aliases first, wait for it to get approved, and then do another BUR to add the implications?

naughtycugaowo said:
The way I understand that, the problem is that I can't create an implication for a tag that is currently aliased to something. Do I have to do a BUR where request to remove the aliases first, wait for it to get approved, and then do another BUR to add the implications?

Exactly that. If it was the other way around, you could do it through an alias request (they don't have some limitations BURs have), but I don't believe you can do it for flipping an alias to an implication.

Granted, this should not be an implication regardless. See the conversation about MH Guardians and their implications

bugabond said:
Exactly that.

Thanks! It would be useful to have an own command for that (smth like implyAlias), since the current way looks like a lot overhead for both users that want to create the BUR, and the mods approving them.

bugabond said:
Granted, this should not be an implication regardless. See the conversation about MH Guardians and their implications

I'm not sure about that. If I type in rathalos into the search bar, I'd expect azure_rathalos and silver_rathalos to show up in the search results too (the way it is currently implemented)

watsit said:
...This would otherwise make it impossible to search for posts that have both the guardian and non-guardian variations together in one post (searching "guardian_seikret seikret" would have posts that have just a guardian seikret and be no different than just searching "guardian_seikret", while searching "seikret" would return posts that don't have a normal seikret in sight).

Then again, this is a very valid critique.
The only way to implement both that I can think of would be to create a new tag for the base species of every monster (like nargacuga_base or nargacuga_(species) ); but I think that might be a bit of an overkill.
Without a new tag, imo the way it's been solved for most subspecies currently (the subspecies implies the parent species) is more true to the behavior the user expects

naughtycugaowo said:
Then again, this is a very valid critique.
The only way to implement both that I can think of would be to create a new tag for the base species of every monster (like nargacuga_base or nargacuga_(species) ); but I think that might be a bit of an overkill.
Without a new tag, imo the way it's been solved for most subspecies currently (the subspecies implies the parent species) is more true to the behavior the user expects

This problem has appeared with regional forms and mega evolutions of Pokemon, and it is handled in the way that there is no implications between them, but rather the alternate form implicates a specific tag grouping this variant together (mega evolution, alolan form, etc)

I mentioned this in BUR 55562 but unfortunately there's not that many people handling MH tag relationships, and so I got no feedback

bugabond said:
This problem has appeared with regional forms and mega evolutions of Pokemon, and it is handled in the way that there is no implications between them, but rather the alternate form implicates a specific tag grouping this variant together (mega evolution, alolan form, etc)

I mentioned this in BUR 55562 but unfortunately there's not that many people handling MH tag relationships, and so I got no feedback

I've looked into it, and I'm not convinced the approach for pokemon is the best one.
Imo, there should be one tag for the whole species, that includes the base form + all subspecies.
Another issue: If you take a look at how often for example charizard has to be re-tagged with mega_charizard by some motivated folks:
https://e621.net/post_versions?commit=Search&search%5Btags%5D=mega_charizard&search%5Btags_removed%5D=charizard
you'll see why I think that this seems to be pretty unintuitive for the average user.

The way I see it, there are two options:

1)
Have every subspecies imply their parent species:
The intuitive way, userfriendly. This is the current way things are handled for most MH subspecies at the moment. Can be inaccurate if both the base form and subspecies are within the same post.

2)
Have a dedicated tag for every species, which is implied by every subspecies and the base form:
The more correct approach. Lets you filter more precisely. Will very likely lead to more work, since it's unintuitive for uploaders, and might be confusing for the average user while searching

I'm motivated to spend some time cleaning up the tags for Monster Hunter atm, but I'm not looking forward to keep removing the parent species from posts since people don't understand the second approach tbh

naughtycugaowo said:
I've looked into it, and I'm not convinced the approach for pokemon is the best one.

It's not just pokemon, it applies to characters as well (link_(wolf_form) and link_(rabbit_form) don't imply link). It's the general way "forms" are handled, and just as you wouldn't have link_(rabbit_form) imply link_(wolf_form), the normal nargacuga form tag shouldn't be implied by its other form tags. People should be able to search for posts with the normal form with any of its other identifiable forms.

watsit said:
People should be able to search for posts with the normal form with any of its other identifiable forms.

You mean without?

I take your not a friend of my first approach. What are your thoughts about the second one?

naughtycugaowo said:
You mean without?

No, with. People should be able to search nargacuga <other_nagracuga_form> to find both the normal nargacuga and the other nagracuga form in the same post. An implication would mean a search like that is identical to just searching the other nargacuga form, giving results with and without (primarily without) the normal nargacuga form.

naughtycugaowo said:
I take your not a friend of my first approach. What are your thoughts about the second one?

A separate "base" tag that's implied by each form would be the ideal way to do it, yes. Similar to lycanroc which isn't itself a distinct form, and is implied by midnight_lycanroc, midday_lycanroc, and dusk_lycanroc, which are the only three official forms. The idea has been floated around in relation to pokemon, though there haven't been good options for how to name these "base" tags, so there hasn't been much movement. The "base" tag can't be the normal name people are used to tagging for the normal form (people are used to tagging just nargacuga for the normal form, so the normal form will largely only be tagged that, which if it's also implied by its other forms, brings us back to the same problem of the normal form not being uniquely tagged), and a _(base) suffix isn't intuitive and looks kind of ugly.

Yeah, that's the exact same conclusion I came to.
I think using a new tag to group each species would be the best option. This way, the normal variant still has the tag people are used to, only the subspecies will get mistagged every now and then.
It should be the species name with a suffix (e.g. nargacuga_(group) ), and since the grouping tag will always have more or equal the number of posts the base tag has, this way it will show up higher than the base tag in the results the autocomplete returns when typing in the name.