The bulk update request #2250 is pending approval.

create alias dragon_wings (704) -> membranous_wings (110697)
create alias dragon_wing (3) -> membranous_wings (110697)

Reason: Dragon wings and membranous wings are effectively synonymous, so these tags can be aliased together. Dragon wings have no canonical appearance beyond being membranous due to being fictional creatures, so there’s no reason to keep it as a distinct tag. Also, the very similar drake_wings has already been aliased.

scaliespe said:
Reason: Dragon wings and membranous wings are synonymous, so these tags can be aliased together.

What about for feathered_dragons or furred_dragons? Their wings aren't always membranous.

watsit said:
What about for feathered_dragons or furred_dragons? Their wings aren't always membranous.

Either we alias to membranous and clean up occasionally, or alias to wings in general and lack the proper tags.

watsit said:
What about for feathered_dragons or furred_dragons? Their wings aren't always membranous.

feathered_wings exists for what you may see on feathered dragons when they have wings that aren’t membranous.

I’m not sure what furred dragons could possibly have that aren’t membranous wings, unless they have feathered wings for some reason. Even furred wings should be membranous, as the fur would presumably be growing out of a membrane such as they do on bat_wings (the only example in nature of a winged creature with fur - note also that bat_wings implies membranous_wings). If it’s something else, it’s not something that can possibly exist in nature. Not to say that it can’t exist in art, but it’s rare.

Even so, I think the chance of this causing a mistag is so low that it’s worth the alias. About 250 tags for dragon_wings so far, and I can’t find any that aren’t membranous.

scaliespe said:
feathered_wings exists for what you may see on feathered dragons when they have wings that aren’t membranous.

Obviously there is a feathered_wings tag for feathered wings, just as there's a membranous_wings tag for membranous wings. My point is that "dragon wings" tells you nothing about the wings being membranous or feathered or insect or bone or mechanical or cosmic or ???. You can't know what the wings are made of from that tag, just that they're on a dragon (which come in a variety of types), so aliasing it to a specific type of wing isn't unlikely to produce mistags.

scaliespe said:
Even so, I think the chance of this causing a mistag is so low that it’s worth the alias. About 250 tags for dragon_wings so far, and I can’t find any that aren’t membranous.

Except a mistag you removed a few hours ago, which has magical flaming wings. And I found another, which appear to be feathered wing arms, and another, which appear to be some goopy liquid substance, both on the first page of results for dragon_wings. How many removed uses were non-membranous, rather than just the remaining ones? Or put another way, how likely is someone to try using the tag in the future in reference to non-membranous dragon wings, and cause a number of mistags due to it getting aliased to membranous_wings with all its implications? Given that there has been some uses of the tag for non-membranous wings that have since been cleaned up, it seems likely to happen again.

watsit said:
Obviously there is a feathered_wings tag for feathered wings, just as there's a membranous_wings tag for membranous wings. My point is that "dragon wings" tells you nothing about the wings being membranous or feathered or insect or bone or mechanical or cosmic or ???. You can't know what the wings are made of from that tag, just that they're on a dragon (which come in a variety of types), so aliasing it to a specific type of wing isn't unlikely to produce mistags.

Dragons canonically have a certain type of wing (with some variation) that are always membranous. Most people will be able to recognize that feathered wings are not standard for a dragon. I don’t know what would possess someone to tag bird wings on a dragon as “dragon wings,” but so they did precisely once, as you pointed out below. However, that being said…

Except a mistag you removed a few hours ago, which has magical flaming wings. And I found another, which appear to be feathered wing arms, and another, which appear to be some goopy liquid substance, both on the first page of results for dragon_wings. How many removed uses were non-membranous, rather than just the remaining ones? Or put another way, how likely is someone to try using the tag in the future in reference to non-membranous dragon wings, and cause a number of mistags due to it getting aliased to membranous_wings with all its implications? Given that there has been some uses of the tag for non-membranous wings that have since been cleaned up, it seems likely to happen again.

I’ll start by stating that the primary reason I’m arguing for this is because the alias system isn’t only for getting rid of unwanted tags. It’s also for helping users find the actual tag that they’re looking for. For every user who uploads a post here, there are probably hundreds, if not thousands more who are just using these tags for searching. So, as it turns out, “membranous” is not a word in the average English speaker’s vocabulary. Out of 529 cases, the user was apparently trying to find membranous_wings in 525 of those cases (see below), but they either didn’t know that membranous_wings existed, or they forgot the word for it. They’re going to keep doing that, and aliasing this tag to wings isn’t going to be much help. Most of them will be lazy and leave it as “wings.” Most users probably won’t go to the wings wiki page to try to find the precise tag that they’re looking for. And so, they never find it, we lose precision in our tagging system. As for the question of whether or not this will cause mistags, I think even a very small number of mistags can be justified for helping users find what they want, especially when it comes to searching. User wants to find membranous wings, can’t think of any name but “dragon_wings.” The search term gets aliased to “wings” and they probably get a majority of avian wings. User can’t find what they want, and they don’t know where to look for this information. A simple alias could solve all of that.

As for whether or not this would cause too many mistags to manage, I’ll get to that in a second. First, I’d like to note that I didn’t remove the tag from post #2784668 as I figured it’s close enough. They have the shape of typical membranous wings, and the goo appears to be forming a membrane structure. My understanding is that we tag goo creature anatomy the same as if it wasn’t made of goo. Like how goo_penis is still a penis, I’d say a membrane made of goo is still a membrane. This also has a precedent in tagging already: goo_creature membranous_wings

I wasn’t sure what post #2966784 was at all. They don’t look much like feathers, and I thought maybe it could be some really tattered (and poorly drawn) membrane. However, I’m not even sure those are wings on second thought. They look more like blue fur elbow tufts. Regardless, I’ll count that one as a mistag.

So, I tallied up all these posts to try to figure out exactly what kind of ratio we’re getting here. Out of all the other pages, I only found one more mistag - a case of mechanical_wings. Out of all the deleted tag uses, I still found only one true mistag (that being the blue feral with the bird wings - I don’t count post #924345 as those are still clearly membranous; they just have scales on them. You’ll already find plenty of similar things under scales membranous_wings, like post #3234488 as one of the first results. The scales aren’t just free floating, they’re apparently covering a membrane).

With the newly removed tags, we have 245 results for dragon_wings and 284 posts that the tag has been removed from. 284 + 245 = 529. Of those, only four could not reasonably be considered membranous. 4 ÷ 529 = ~0.00756. So, about 0.756% of these were mistags - less than 1%. I think it’s safe to say that mistags would be very rare.

Even so, the alias would help with that. The tag suggestion system displays what a tag is aliased to if you start typing it in. Most normal users, if they see this, won’t use the tag if that’s not what they’re looking for. Likely the only excuse for that is that they’re simply copy/pasting tags over from other websites without checking their validity here, which might cause an incorrect application of membranous_wings - but even so, there’s probably well under a 1% chance of the tag being applied where it shouldn’t have been there. And further still, users tagging posts like that are going to have many incorrect or invalid tags, not just this one, so they’d need to be cleaned up regardless. For the benefit to the average user in pointing to the tag that they’re overwhelmingly likely to be looking for, I’d say the very rare case of a mistag is worth it.

scaliespe said:
Dragons canonically have a certain type of wing (with some variation) that are always membranous.

No, they don't. Dragons are fictional creatures, with various cultures depicting them very differently. There is no "canonical" dragon design, much less a canonical type of wing dragons have, as there are plenty of dragon types that don't even have wings, let alone membranous ones. But even if there was, the fact that there are and have been uses of the dragon wings tag for non-membranous wings means the alias will cause mistags.

scaliespe said:
I don’t know what would possess someone to tag bird wings on a dragon as “dragon wings,” but so they did precisely once, as you pointed out below.

Twice, at least. One post that still had it when I looked, and one post used to have it but was removed some time afterward. There is quite possibly more, since I don't have the wherewithal to check each post, and some posts have been deleted and can't be checked due to having no working source.

scaliespe said:
I’ll start by stating that the primary reason I’m arguing for this is because the alias system isn’t only for getting rid of unwanted tags. It’s also for helping users find the actual tag that they’re looking for. For every user who uploads a post here, there are probably hundreds, if not thousands more who are just using these tags for searching.

And we don't create aliases to the statistically most like option, when there are other completely valid options. That's how you get mistags, and we can't know that a given person searching for "dragon_wings" means "membranous_wings", as opposed to "a feathered dragon with feathered wings" or "a dragon with any kind of wings", not to mention many non-dragons have membranous wings that the search would include, so the search results for "dragon_wings" would incorrectly include membranous wings on non-dragons, while incorrectly excluding other types of wings dragon can have. Such tags either get disambiguated, or aliased to the lowest common denominator (e.g. wings in this case), to reduce the mistags, as the user will need to find better search terms for what they actually want regardless.

scaliespe said:
As for whether or not this would cause too many mistags to manage, I’ll get to that in a second.

Not a question of being "too many mistags to manage", more than it is extra unnecessary mistags that have to be managed on top of everything else.

scaliespe said:
First, I’d like to note that I didn’t remove the tag from post #2784668 as I figured it’s close enough. They have the shape of typical membranous wings, and the goo appears to be forming a membrane structure.

But it's not. A membrane is "a thin layer of tissue that is meant to act as a barrier of sorts". Liquids and other gooey substances are not tissue, so aren't membranous, they're liquid.

scaliespe said:
I’d say a membrane made of goo is still a membrane.

A wing made of goo is still a wing. But a membrane made of goo makes no sense, it's like saying a rock made of water. Sure you can have water/goo in the shape of a rock/membrane, but it's still just water/goo, not actually a rock/membrane. They're two different substances. At most, you can have a membrane covered in and/or dripping goo, but in the aforementioned example, all indications are it's purely goo in place of membrane and not merely a coating.

scaliespe said:
I wasn’t sure what post #2966784 was at all. They don’t look much like feathers, and I thought maybe it could be some really tattered (and poorly drawn) membrane. However, I’m not even sure those are wings on second thought. They look more like blue fur elbow tufts. Regardless, I’ll count that one as a mistag.

I too had difficult even seeing any wings. I initially though it was just extra fluff coming off the arm... but given the assumption there were meant to be wings, I considered it possibly something like post #3212070, with feathers coming off the upper arm and elbow. In either case, it looks way too fluffy to be membranous.

scaliespe said:
I don’t count post #924345 as those are still clearly membranous; they just have scales on them.

That's cheating. Membranous wings are supposed to be for when you see a wing made of membranous tissue, not for wings that technically have a membrane inside them underneath some other material, like scales. You can't see the membrane, so TWYS would dictate not tagging it.

scaliespe said:
You’ll already find plenty of similar things under scales membranous_wings, like post #3234488 as one of the first results. The scales aren’t just free floating, they’re apparently covering a membrane).

For that one in particular, the detail isn't great on indicating whether it's scaled_wings or membranous_wings. Compared to post #924345 where the wing structure has the same sharpness and roughness associated with hard scales all throughout, post #3234488 has wings with a more typical fleshy structure that's made for stretching and catching air, that has some line patterns that are similar to (but not exactly the same as) the lines they used for the body scales. The latter could go either way.

scaliespe said:
With the newly removed tags, we have 245 results for dragon_wings and 284 posts that the tag has been removed from. 284 + 245 = 529. Of those, only four could not reasonably be considered membranous. 4 ÷ 529 = ~0.00756. So, about 0.756% of these were mistags - less than 1%. I think it’s safe to say that mistags would be very rare.

That's quite underestimating it by my count.

There's post #2984966 where you can barely see the wings, let alone tell what they're made of. post #2784665 post #2784668 post #2784673 and post #2784675 are all goo wings, not membranous, IMO. post #924345 is scales, no visible membrane for the wings. post #2966784 doesn't even have apparent wings, so obviously not membranous ones. post #765731 is feathered wings. Just those alone would bring it up to 1.5%, but I haven't checked the remaining ones that can be checked, to say nothing of those that can't be checked. That's not an insignificant percentage.

Why not just alias dragon_wings to wings and be done with it? A search of dragon + wings (or membranous_wings) would give better results anyway, and a non-dragon wouldn't get tagged with dragon_wings either. While modern culture generally regards bat-like membranous wings as traditional for dragons, they are by no means universal for winged dragons. Some taggers will see a dragon wearing wings and reasonably think dragon_wings even though they aren't membranous wings. In the end, rather than quibble over exactly what dragon wings are (besides being worn by a dragon), alias them to something we can agree they are: wings.

clawstripe said:
Why not just alias dragon_wings to wings and be done with it?

That's what I would suggest. There's no way to tell whether someone searching it meant dragon+wings, membranous+wings, dragon+membranous+wings, or something else, so it won't give good results no matter what it's aliased to. Nor can it be known what someone tagging it might mean by it, beyond some kind of wings. So aliasing to wings seems like a much safer bet to avoid mistags.

watsit said:
No, they don't. Dragons are fictional creatures, with various cultures depicting them very differently. There is no "canonical" dragon design, much less a canonical type of wing dragons have, as there are plenty of dragon types that don't even have wings, let alone membranous ones. But even if there was, the fact that there are and have been uses of the dragon wings tag for non-membranous wings means the alias will cause mistags.

In western mythology, dragons do have a roughly canonical appearance. I understand that this is not necessarily the case in the modern world, but traditional depictions of the dragon of western mythology, if not wingless, pretty unanimously had membranous wings. There were variations in size and shape to some extent, but they were seldom if ever depicted as, say, bird wings. They’re so commonplace that any exception would be a notable oddity - even Wikipedia doesn’t mention them in their article on dragons. Besides the West, eastern dragons were always wingless, so there’s really only one style of wing that can be reasonably called a “dragon wing” in a mythological context.

Twice, at least. One post that still had it when I looked, and one post used to have it but was removed some time afterward. There is quite possibly more, since I don't have the wherewithal to check each post, and some posts have been deleted and can't be checked due to having no working source.

If you want to count the one that I think looks more like furry arms, then sure, two. I did check every post, including the ones where the tag was removed (I skipped the ones where the tag was also replaced with membranous_wings in the same edit, as I assume that those are all, in fact, membranous wings…) and I didn’t encounter any more feathered wings. I didn’t run into any cases where the image was deleted that didn’t at least have it available at source.

And we don't create aliases to the statistically most like option, when there are other completely valid options. That's how you get mistags, and we can't know that a given person searching for "dragon_wings" means "membranous_wings", as opposed to "a feathered dragon with feathered wings" or "a dragon with any kind of wings"

I still have to reject the idea that there are other valid options, regardless of what mistags there may or may not be. A dragon with wings that are not the traditional type of dragon wings shouldn’t be tagged dragon_wings, in the hypothetical situation where this would be kept as a valid tag. If it’s a dragon with avian wings, the dragon_wings tag would be removed and replaced with avian_wings (oh, there’s another tag that probably needs an alias. Would you be opposed to aliasing avian_wings to feathered_wings?)

not to mention many non-dragons have membranous wings that the search would include, so the search results for "dragon_wings" would incorrectly include membranous wings on non-dragons, while incorrectly excluding other types of wings dragon can have. Such tags either get disambiguated, or aliased to the lowest common denominator (e.g. wings in this case), to reduce the mistags, as the user will need to find better search terms for what they actually want regardless.

Actually, that’s what I would expect a user would want. If you just want any dragon with wings, you’d search dragon + wings. But that’s not how users are trying to use this tag. Even the current results for dragon_wings include many non-dragons: dragon_wings -dragon
You have wolves with dragon wings, horses with dragon wings, rabbits with dragon wings, and so on. Clearly, the overwhelming majority of this tag’s use is attempting to tag the wings that a traditional dragon has regardless of what species it’s on. Usually it’s on a dragon, but often not.

This is actually making me reconsider the validity of the tag. Perhaps this should be an implication instead, similar to bat_wings? That way, at least the occasional incorrect use of the tag can just be removed rather than having the mistag get mixed in with all the membranous_wings posts. Dragon wings, although fictional, have a recognizable appearance which can be found on other species. On the other hand, traditional dragon wings are essentially identical to bat wings, except that they might have scales instead of fur. In fact, are there any membranous wings that aren’t essentially based on bat wings? Maybe bat_wings should’ve been aliased to membranous_wings instead. The bat_wings wiki says that they have thumbs, but most dragon wings do too. Maybe I’ll suggest aliasing that one too, later.

Not a question of being "too many mistags to manage", more than it is extra unnecessary mistags that have to be managed on top of everything else.

Four mistags by my count, and, what… 8, by yours? In the entire history of the site, out of 3 million+ posts. I don’t think that constitutes a meaningful enough amount to discount an alias, personally. I wouldn’t start considering an alias invalid unless the mistags start approaching the hundreds, at least, depending on the frequency of use, just as long as the alias is helpful for finding the correct tag that people are most likely looking for.

But it's not. A membrane is "a thin layer of tissue that is meant to act as a barrier of sorts". Liquids and other gooey substances are not tissue, so aren't membranous, they're liquid.

A wing made of goo is still a wing. But a membrane made of goo makes no sense, it's like saying a rock made of water. Sure you can have water/goo in the shape of a rock/membrane, but it's still just water/goo, not actually a rock/membrane. They're two different substances. At most, you can have a membrane covered in and/or dripping goo, but in the aforementioned example, all indications are it's purely goo in place of membrane and not merely a coating.

In the real world, that’s completely true. However, there aren’t any such rules in fictional universes where a creature made entirely of slime can exist. I think some leniency has to be given for unusual anatomy like that. Slime in the shape of breasts aren’t actually breasts (it’s doubtful that they can produce milk), but we still tag them as breasts. Goo creatures often have claws and horns, which are explicitly made of keratin in the real world. Horns made of goo wouldn’t be considered to be actual horns normally, and yet we still tag them as such.

But even if you’re right, this isn’t an issue with the dragon_wings tag. People have always been tagging goo wings as membranous. It’s a much broader issue.

That's cheating. Membranous wings are supposed to be for when you see a wing made of membranous tissue, not for wings that technically have a membrane inside them underneath some other material, like scales. You can't see the membrane, so TWYS would dictate not tagging it.

For that one in particular, the detail isn't great on indicating whether it's scaled_wings or membranous_wings. Compared to post #924345 where the wing structure has the same sharpness and roughness associated with hard scales all throughout, post #3234488 has wings with a more typical fleshy structure that's made for stretching and catching air, that has some line patterns that are similar to (but not exactly the same as) the lines they used for the body scales. The latter could go either way.

I already see this all the time with both scale and fur coverings, so that would require quite a big cleanup of the membranous_wings tag. It would also disqualify what appear to be membranous wings where the wings are folded so that you can’t actually see the membrane. But regardless, what people are attempting to tag in these cases is the general appearance of the wing. It seems much more useful to the average user to allow some variation in appearance/material/covering for these wings so that people can find what they want. I rather doubt that anyone searching for membranous wings really cares if the wings also have fur or scales on them, or are technically made of goo.

In nature, there are essentially three kinds of wings: feathers growing out of a forelimb; a membrane attached to a forelimb; or a thin, rigid piece of chitin or something like it that is not itself an actual limb with joints. In the world of fiction, there are bound to be variations on these three basic types: they don’t always fit the definition seen in nature, but to have separate tags for each various kind seems like too much work for little to no benefit. We’d need a tag like goo_membrane_wings (which doesn’t imply membranous_wings, I guess), goo_feather_wings (likewise, since goo feathers aren’t actually feathers?), furred_membrane_wings (again, clearly a fur-covered membrane, but you can’t see the membrane, just the fur), scaled_membrane_wings, and so on. I don’t see how this is useful, though. If it’s just to avoid the membrane_(anatomy) implication, would it be better to remove that? There can be cases where a wing is clearly membranous without the bare membrane itself being visible.

That's quite underestimating it by my count.

There's post #2984966 where you can barely see the wings, let alone tell what they're made of. post #2784665 post #2784668 post #2784673 and post #2784675 are all goo wings, not membranous, IMO. post #924345 is scales, no visible membrane for the wings. post #2966784 doesn't even have apparent wings, so obviously not membranous ones. post #765731 is feathered wings. Just those alone would bring it up to 1.5%, but I haven't checked the remaining ones that can be checked, to say nothing of those that can't be checked. That's not an insignificant percentage.

I still don’t think that’s much. But even so, I think the point of this alias is more to help people find the tag that they’re actually looking for, even if it’s only 98.5% of the time. If it has to be the correct tag exactly 100% of the time, we’ll have to reconsider a lot of the aliases we currently have. bat -> chiropteran is probably going to be the animal 99% of the time, but it does occasionally get mistagged for bat_(object). Since bat does not equal chiropteran 100% of the time, surely you must argue that this alias is invalid. But the only alternative is bat_(disambiguation) which will accumulate hundreds of results that likely nobody will bother cleaning up and of which at least 99% will be chiropterans. The issue with tags like chiropteran and membranous_wings is that these words don’t exist in the average user’s vocabulary, so we need to use more common words like “bat” or “dragon wings” to point them in the right direction. If somewhere in the range of 98.5 to 99.5 percent of uses are attempting to find precisely one thing, I don’t think that’s enough to invalidate the alias.

clawstripe said:
Why not just alias dragon_wings to wings and be done with it? A search of dragon + wings (or membranous_wings) would give better results anyway, and a non-dragon wouldn't get tagged with dragon_wings either.

Well, no, a lot of users have been using the tag for traditional dragon-style (ie. membranous) wings on non-dragons.

While modern culture generally regards bat-like membranous wings as traditional for dragons, they are by no means universal for winged dragons. Some taggers will see a dragon wearing wings and reasonably think dragon_wings even though they aren't membranous wings.

After sifting through all 500-something posts the tag has ever been used on, I only found four cases where they weren’t at least arguably membranous wings, though there are a few other posts on the fence.

In the end, rather than quibble over exactly what dragon wings are (besides being worn by a dragon), alias them to something we can agree they are: wings.

My reason for quibbling this point is mainly this:

scaliespe said:

I’ll start by stating that the primary reason I’m arguing for this is because the alias system isn’t only for getting rid of unwanted tags. It’s also for helping users find the actual tag that they’re looking for. For every user who uploads a post here, there are probably hundreds, if not thousands more who are just using these tags for searching. So, as it turns out, “membranous” is not a word in the average English speaker’s vocabulary. Out of 529 cases, the user was apparently trying to find membranous_wings in 525 of those cases (see below), but they either didn’t know that membranous_wings existed, or they forgot the word for it. They’re going to keep doing that, and aliasing this tag to wings isn’t going to be much help. Most of them will be lazy and leave it as “wings.” Most users probably won’t go to the wings wiki page to try to find the precise tag that they’re looking for. And so, they never find it, we lose precision in our tagging system. As for the question of whether or not this will cause mistags, I think even a very small number of mistags can be justified for helping users find what they want, especially when it comes to searching. User wants to find membranous wings, can’t think of any name but “dragon_wings.” The search term gets aliased to “wings” and they probably get a majority of avian wings. User can’t find what they want, and they don’t know where to look for this information. A simple alias could solve all of that.

clawstripe said:
Why not just alias dragon_wings to wings and be done with it? A search of dragon + wings (or membranous_wings) would give better results anyway, and a non-dragon wouldn't get tagged with dragon_wings either. While modern culture generally regards bat-like membranous wings as traditional for dragons, they are by no means universal for winged dragons. Some taggers will see a dragon wearing wings and reasonably think dragon_wings even though they aren't membranous wings. In the end, rather than quibble over exactly what dragon wings are (besides being worn by a dragon), alias them to something we can agree they are: wings.

Should we though? Because not all dragons have wings. What if there's two characters in an image, one's a bird and the other's an Eastern Dragon but you're only looking for images with dragons that have wings?

pheagleadler said:
Should we though? Because not all dragons have wings. What if there's two characters in an image, one's a bird and the other's an Eastern Dragon but you're only looking for images with dragons that have wings?

This should give you good results:
dragon membranous_wings -solo -insect_wings -feathered_wings -furred_wings