"detexted" just sounds slightly off, could the alias be reversed?
Posted under Tags
"detexted" just sounds slightly off, could the alias be reversed?
ghostrigger said:
create alias textless -> detexted
i would guess that whoever removes the text is not absolutely significant.
If textless referred to the original image and detexted referred to photoshopping the image to remove the text, then yes it would be rather significant who removed the text. One by its very nature is photoshopped to cut out the text and is a best guess of what is beneath the text, the other is what actually depicts what was beneath the text before they overlaided it on top of the image. At minimum detexted rightfully should implicate photoshop, while textless wouldn't.
NWF_Renim said:
If textless referred to the original image and detexted referred to photoshopping the image to remove the text, then yes it would be rather significant who removed the text. One by its very nature is photoshopped to cut out the text and is a best guess of what is beneath the text, the other is what actually depicts what was beneath the text before they overlaided it on top of the image. At minimum detexted rightfully should implicate photoshop, while textless wouldn't.
Good points.
So:
implicate detexted -> photoshop
implicate detexted -> textless
NWF_Renim said:
If textless referred to the original image and detexted referred to photoshopping the image to remove the text, then yes it would be rather significant who removed the text. One by its very nature is photoshopped to cut out the text and is a best guess of what is beneath the text, the other is what actually depicts what was beneath the text before they overlaided it on top of the image. At minimum detexted rightfully should implicate photoshop, while textless wouldn't.
Wouldn't the orignal image without text just become the parent post of the version with text?
Why do we need a textless tag for it? The fact that image has no text is not a feature of the image. The fact that the child post (scan or whatever) has text added on top is a feature of that child post.
Yeah, I don't see the point of textless. It's just noise.
S1eth said:
Wouldn't the orignal image without text just become the parent post of the version with text?
Why do we need a textless tag for it? The fact that image has no text is not a feature of the image. The fact that the child post (scan or whatever) has text added on top is a feature of that child post.
I think you're looking at things from the wrong perspective here, which is why it comes off as completely useless to you, but that also just shows that you're not at all thinking of the situations where the text version would be the parent and the textless version would be the child. Authors releases things like post #1028796 and with images like post #1573275, I think the image with the text (post #1573274) comes off as "more complete" than with its removal. Don't think there is example under the images currently tagged, but if you have images like post #797463 that are pooled together because they tell a story, if they included a version without text it wouldn't be the parent, because the one with text is kind of the completed intended version of the image.
S1eth said:
Why do we need a textless tag for it? The fact that image has no text is not a feature of the image. The fact that the child post (scan or whatever) has text added on top is a feature of that child post.
Question, how does a user find images that have both a version with text and a version that is textless? If it is easy with our current setup, then I guess we don't need the tag, but if it isn't then I think this tag has a lot more worth than you're giving it credit for. For common use textless should normally be used only when there is both a version with text and version without (some exceptions, such as that manga page example).
Updated
Authors releases things like post #1028796 and with images like post #1573275
Those are 2 completely different situations (which shouldn't be handled with the same tag).
I thought this was about magazine scans, and the removal of text from these scans vs. the original image without the text.
Either way, we have a tag for post #1028796: blank_speech_bubble.
This tag actually describes a visible (tag what you see) feature of the post.
Would anyone even want to search for "post x that has child or parent post y where x has text and y has no text"? (you need a parent/child relationship (or a link in any form) to be able to find the other post, or it's useless)
And what if only the version with with text is uploaded and the version without is also available on pixiv (because the uploader thought it would be redundant to upload both versions)? How do these tags help people find the textless version?
It currently doesn't matter which is the parent and which is the child post, since if you find one, you become interested in finding the other. I don't think people will want to find one without knowing the other exists.
NWF_Renim said:
At minimum detexted rightfully should implicate photoshop, while textless wouldn't.
Kikimaru said:
The problem with that is that detexted images are not necessarily photoshops.
E.g. Text post #1292007 and the detexted post #1386170 which came from an issue of Megami Magazine Deluxe.
It's not "detexted" if it's from the original artist.
S1eth said:
Those are 2 completely different situations (which shouldn't be handled with the same tag).
I thought this was about magazine scans, and the removal of text from these scans vs. the original image without the text.
I don't really agree that they are that different. They're both essentially (for lack of a better word) "incomplete" and that a more "complete" version exists with text.
If textless was limited to just scans and magazine covers, then I would think textless is too narrowly defined. That would be because I think we'd find its occurrence much higher with manga, image sets, and the like.
S1eth said:
Either way, we have a tag for post #1028796: blank_speech_bubble.
This tag actually describes a visible (tag what you see) feature of the post.
I have a problem with that tag in its current form. How do you distinguish between essentially incomplete manga pages (post #1250477) from complete manga pages that utilize a few blank speech bubbles (post #1131120)? Including a blank speech bubble in comic pages that are complete have a very different meaning from comic pages being completely textless due to it being incomplete. I will admit I don't know what I'd do with things like post #1203287, but at minimum as far as manga pages go I think there is a huge flaw if there isn't a distinction between the incomplete textless manga pages and the completed ones.
S1eth said:
Would anyone even want to search for "post x that has child or parent post y where x has text and y has no text"? (you need a parent/child relationship (or a link in any form) to be able to find the other post, or it's useless)
Well I think it would be useful to have something that did indicate that, though I can only speak for myself. As for the relationship matter, I think that's kind of a silly thing to question, as we already have tags that rely on needing such relationships: cover_image, revised, reference_work, reference_photo, derivative_work. All of them require a relationship with another image to have the tag used.
S1eth said:
And what if only the version with with text is uploaded and the version without is also available on pixiv (because the uploader thought it would be redundant to upload both versions)? How do these tags help people find the textless version?
I think you're being unreasonable with making that request of the tag. You're asking for a tag which would have a scope that has to reach beyond our own website, when a tag that is more grounded to our own website would make much more sense. Having it based somewhat on the revised tag wouldn't be a bad approach, thus having Textless used only for images in which we have both a version with text and a version without text on danbooru, with the exception imo of comic pages where all the speech bubbles are empty, as its very nature would normally indicate that text was supposed to be there. Additionally if the poster links the other version then using the tag would be fine, though if that makes things too open then we don't have to do that.
S1eth said:
It currently doesn't matter which is the parent and which is the child post, since if you find one, you become interested in finding the other. I don't think people will want to find one without knowing the other exists.
I can agree that which is the parent and the child shouldn't matter, the only thing that should matter is the linkage between the two. If textless requires a linkage being present (exception again on comics, which should be easily filterable from the search), then by looking up the tag you're easily able to find both images, which should be a good thing. The way you talk makes it seem like it is easy to find any of these images that have such a linkage, when as far as I'm aware we don't have something that would actually aid in finding images that have such a relationship. Outside of coming across by chance an image that is parent-child to the alternate version, I'm not sure how you're expecting any user to readily find images with such a relationship.
I have a problem with that tag in its current form. How do you distinguish between essentially incomplete manga pages (post #1250477) from complete manga pages that utilize a few blank speech bubbles (post #1131120)?
All bubbles should be empty. Wasn't the tag created in the first place for "template posts" (aka fill in the bubbles yourself)?
S1eth said:
All bubbles should be empty. Wasn't the tag created in the first place for "template posts" (aka fill in the bubbles yourself)?
Then isn't the name of the tag a poor choice? Blank speech bubbles can be a feature of even completed manga pages with text, if the tag is intended for incomplete textless or template images, then it's not tagging "blank speech bubbles" as a feature, it's tagging images that are left deliberately incomplete.
Updated
We actually have a blank_speech_bubble tag to cover that.
Kikimaru said:
We actually have a blank_speech_bubble tag to cover that.
That's the tag they were already talking about.
bump. other opinions? there's still no clear-cut definition to distinguish both.
Couldn't we just simplify it?
• If an image had text and now it doesn't, detexted
• If the text removal is from the original artist, no further action needed
• If the text removal is from a third party, manually add photoshop
Yes you could have an artist-detexted image that has been photoshopped in some other way by someone else but isn't defining it more specifically a bit unnecessary?
If we really do want to keep a strict the distinction, then keep it as textless = artist text removal, detexted = third party photoshop to remove text
jxh2154 said:
Couldn't we just simplify it?
• If an image had text and now it doesn't, detexted
• If the text removal is from the original artist, no further action needed
• If the text removal is from a third party, manually add photoshop
I guess just to have something done instead of it remaining in limbo and nothing done, go ahead with that, but the name should be "textless" and not "detexted." "Detexted" inherently implies its been photoshopped to remove text, while "textless" just means without text (whether it was before text was added or that text was removed).
Given the detexted wiki states its use for those that were photoshopped, before they get merged the images under that tag will need to have the the photoshop tag applied to them before mixing them with the images under textless.
Also I still stand by that manga pages with no text should also fall under the tag.