AMV Content Descriptors

[Phade]

It's not at all that I worry about folks disagreeing with what tags may be on a video. It's that I worry the org would become convoluted and otherwise confusing for newcomers to join or otherwise browse the great amvs hosted on the platform.

Ahh. Yes, balancing a pleasant and inviting user experience with the core mission of the site (catalog, archive, educate) can be a challenge at times. My hope is that if the user mechanics are intuitive and speedy, we can meet the site’s goals while encouraging participation. Right now, we just have a mental concept of the mechanics, which may or may not match reality when it arrives.

What I am suggesting is not necessairly to get rid of the "levels" but instead transform them into much more digestible, and arguably more intuitive/clearer tags.

Take, for example, Blood/Gore "low" and how it encapsulates videos that feature just mere scrapes. I would not intuitively connect the term "blood/gore" (even when associated with the word "low"). In fact, I would be outright uncomfortable attaching such a tag onto my video because I would be afraid of turning away those who would otherwise enjoy my video but avoided it for wanting to filter against blood/gore in it's entirety.

In addition to the classification of the category type, there are also the two modifiers of “frequency” and “intensity”. This should help the potential viewer better know what to expect in case they happen to be sensitive to a particular category type.

Let's say you have a romance AMV that is typical for the genre. The main character is dramatically running to catch their love interest, but they trip and fall, causing a visible arm scrape and a slight bloody nose; nothing graphic, but it’s kinda visible if you’re paying attention. You may decide to put “Blood: low (mild scrape or nose bleed), frequency: low (one or few), intensity: low (hinted or glancing)”. This should alert someone who may be sensitive or phobic of blood that while blood is typically not in a romance AMV at all, this one does have a little bit.

But instead, let's say you have a comedy AMV that shows the typical comedic nose bleed when encountering an attractive person set to the hilarious song “Every time I see you, my nose bleeds”. For this one, you may decide to pick “Blood: low (mild scrapes or nose bleed), frequency: frequent, Intensity: intense / dramatic”. This combination should clearly alert someone who may just be squeamish at the sight of blood that this may not be the AMV for them.

However, to convert each combination of descriptor between just the Type and Intensity value into proper tags, we would need to generate nine specific words or phrases for each category type with intensity. Coming up with nine distinct descriptors for the ten required and twenty-seven total category types would be daunting to create. Viewing and deciding between 90 and 243 words and phrases (or more if we add category types) would be fantastically overwhelming for the video creator to digest and choose from (in my opinion).

With the input mechanic designed as I’ve proposed here, we can provide guidelines that describe the intent of the category type while keeping the view clean, (hopefully) swift understanding of each category type's purpose, and physical input interactions low (prefilled values and radio button interactions).

If frequent further clarification is needed, we could link to individual AMVs as examples of what is meant by each combination of category type and frequency. If the intended use is still unclear even after that, a forum thread or Discord discussion should help with the correct choice.

Now, what if instead of Blood/Gore "low" we have a general fighting tag? Now I am comfortable in adding "fighting" to my video, as there is no visible blood or gore, but for sure there is fighting (its how they got the scrapes afterall!).

General fighting is the Violence type in the content descriptor system. My gut reaction (pun intended) is that a typical Pokémon battle could be “low”, Dragon Ball Z could be “middle”, and [insert highly violent anime here] as a “high” examples.

Combining the Violence and Blood category types (both required) for a particular AMV could tell the viewer that there is lots of bloodless (or mild blood) intense fighting going on.

Now let's I do use a source that showcases blood, well instead of blood medium, I can just use the word blood, and if its a show like naruto, I can also incorporate the previously mentioned fighting tag.

Finally, for "high" lets say this video breaches into including gore, well now they editor can (potentially) use all three tags, fighting, blood, gore.

A bloody but not "guts-flying gory" showcase AMV could have this combination of category types selected:

Blood/gore: Medium, frequent, intense/dramatic
Violence: Medium, frequent, intense/dramatic

Also, I may be inexperienced, but I can’t think of any example of bloodless gore (meaning if there is gore in an AMV, there are copious amounts of blood to go with it). This is why blood and gore are combined into one category type, since gore is the top end of blood (that and that’s how ESRB had it listed). If there are clear examples of low, medium, high gore that show up in AMVs often enough without blood, we can split gore and blood into their own category types.

By finding general, but clear, terms instead of arbitrary levels, you can intutively convey, and discern your want of a "low, medium, and high" for your tags.

Also now that we've separated these tags into clearer terms, it's much more intuitive if we were to mix and match tags with tags under a different "category"

Subjectivity is definitely unavoidable, even under my proposed system. But it's much easier to discern whether a video features blood, as opposed to determining where the blood is on a scale of low, medium, or high.

Again, to duplicate the system I’ve proposed with tags, we would need to provide and review 9 distinct descriptors for blood, 9 descriptors for gore, 9 descriptors for violence, and so on for every category type. That, to me, seems way worse.

I'm coming at this as someone who organizes AMV panels. Just last year I hosted an 18+ panel that featured fanservice, horror, gore, mature topics, and/or raunchy jokes. Those 5 descriptors that I just used are much clearer than, "humor high, language high, sexual themes(medium? high(?)) sexual behavior(medium(?)), alcohol high violence high[...]" and so on.

I, too, am a long-time AMV panel coordinator. While my typical panel is one hour each, sometimes I create 48-hour AMV lounge panels. I often struggle with the “should this be in the show or not” self-doubt. I also sometimes forget that a great AMV actually contained a “high” level in any of the category types (oh yeah, that song has three glancing n-bombs in the lyrics right in the middle of the song; I shouldn’t put that one in the Saturday 11am block).

Requiring a few but clearly related (low, med, high with frequency and intensity) category types will make narrowing a list of 4000 candidate AMVs to a manageable 1000 much easier and accurate than choosing from a list of 243 options to include/exclude. This search feature will be available to everyone.

If the system is too subjective that searching / filtering videos is not straightforward, I (and I assume others) will pursue other means to find the videos I need / want (usually AMVNews, word of mouth, and so on) which leads into my last point.

I want the org redesign to breathe life into the community where we can maybe point those starting off in the hobby (either editors or just casual viewers) to check it out, maybe find a video that they like because god knows finding videos under the current YouTube algorithm is very difficult. Tags aren't at all useful if it will cause people to turn away from either videos they may enjoy, or the website due to confusion or annoyance.

General tags will still be available. The indicators in this “content descriptors” section is meant to be a standardized set of descriptors that provide a level, intensity, and frequency for each category type.

Actually, one thing I literally just now thought of: since we will have general tags, we could map known common tags to a specific combination of category type and intensity within the “content descriptors” section and prefill those values for the editor. If the AMV editor prefers the tags interface, we can allow tag inputs to occur before content descriptors. Once the editor is done with tags, they can confirm “none” for non-prefilled required category types and just select the “frequency” qualifier for the category types that have a non-none value. Once we have a word cloud for each level of intensity, we could automatically add them to the tooltip descriptor.

Would that be a successful approach from your perspective? (I think this approach could work for everyone.)

Thank you again for your continued feedback and input!!

Phade.

[laceproductions]

Overall this tagging system definitely feels more inline with the kind of system I would prefer to use and I hope others would as well.

I would say that the orientation tag very much feels like it shouldn't be considered in the same vein as the idea of "severity" as that would imply that certain types of orientation portrayals are more severe than others. I do however see benefits in including different orientation tags to allow the viewer to filter out or hone in on specific types of portrayals. In that sense then, I would prefer to see a wider definition than just same-sex, non-traditional, and opposite-sex. I would find great use in looking for lesbian depictions when I'm in the mood for such, or if I ever get the inclination to seek out gay depictions being able to hone in on those, or weed them out in my search, very beneficial.

In the end, I believe you are suggesting the current setup in a more informational sense and I see this area being one where severity doesn't make much sense but perhaps further specification can show a benefit.

[Phade]

Overall this tagging system definitely feels more inline with the kind of system I would prefer to use and I hope others would as well.

Thank you!! 🥰

I would say that the orientation tag very much feels like it shouldn't be considered in the same vein as the idea of "severity" as that would imply that certain types of orientation portrayals are more severe than others. I do however see benefits in including different orientation tags to allow the viewer to filter out or hone in on specific types of portrayals. In that sense then, I would prefer to see a wider definition than just same-sex, non-traditional, and opposite-sex. I would find great use in looking for lesbian depictions when I'm in the mood for such, or if I ever get the inclination to seek out gay depictions being able to hone in on those, or weed them out in my search, very beneficial.

In the end, I believe you are suggesting the current setup in a more informational sense and I see this area being one where severity doesn't make much sense but perhaps further specification can show a benefit.

Yes, the orientation tag will be placed in a non-severity-based "informational" spot rather than lumped with the others that do imply severity. It was a last-second add-in and not fully thought out. I also just noticed that "gender" is plain missing.

I think the "intensity" indication could still potentially be a useful modifier for both gender and orientation. "Hinted male same-sex" and "intense female opposite-sex" (along with most other combinations) do kinda make sense.

I could be wrong (since there is no current way to measure it), but I guess that of the AMVs that focus on aspects of gender and orientation, 50% are opposite-sex, 40% are same-sex, and 10% are ambiguous/other while the participants are 45% male, 45% female, and 10% ambiguous/other. (Again, these are numbers pulled out of my butt)

Does this breakdown ballpark number vibe-feel about right to you? If we just had the three options each for both gender and orientation, do you think you could easily find the subject combinations that interest you?

I feel that if we try to implement a full-blown orientation and gender studies-based system here, there may not be enough AMVs to actually find that would improve results over the three-level method, while also increasing the burden on the AMV editor to choose the correct situation. Even for creators who do want to be more specific, there is still the general tags system.

How does that sound?

Thanks again for your feedback! It is greatly appreciated!!!

Phade.

[KeiichiFace]

I think this is a great step in the right direction. I agree with Mango that an "intensity" scale may not work for all categories, and having easily-digestible terms instead would be a better option. I would feel very weird tagging my video with "blood and gore" over a cut or something (especially because my friends know I really can't handle gore). To your nosebleed example, you could have a nosebleed tag.

If you're worried about overwhelming the user with different tags, I suggest simplifying your approach. "90 and 243 words" IS a lot. You don't need that many descriptors. But also, words have different connotations and ultimately different meanings. I think there's a happy medium to be found there. I think you could cover everything you need to with, say, 50 words. Or, do 100 words and have different category headings so that someone isn't looking at all of the words at one time. For example, if I made a straight-up Romance video (no fighting or anything like that), I could just look under the Romance tag category (and you could even put "nosebleed" in there as well, to cover all of your bases) - I know I wouldn't need to look at the "Fighting" category or any of its tags, so I'll just keep that closed.

[laceproductions]

I think the "intensity" indication could still potentially be a useful modifier for both gender and orientation. "Hinted male same-sex" and "intense female opposite-sex" (along with most other combinations) do kinda make sense.

When put that way that does make sense. In this sense severity is really a generalized amount or focus, which I have no issues with.

I could be wrong (since there is no current way to measure it), but I guess that of the AMVs that focus on aspects of gender and orientation, 50% are opposite-sex, 40% are same-sex, and 10% are ambiguous/other while the participants are 45% male, 45% female, and 10% ambiguous/other. (Again, these are numbers pulled out of my butt)

Does this breakdown ballpark number vibe-feel about right to you? If we just had the three options each for both gender and orientation, do you think you could easily find the subject combinations that interest you?

I can't say I'm well versed in the differentiation of different orientations and genders depicted, but from my experience and knowledge being of the first two'ish decades of videos that seems a pretty close number, possibly with a higher percentage on the hetero side of things. Though that ratio might have changed in more recent years. I would hope it has.

I certainly feel as defined and explained I could find what I'm looking for and filter out what I am not.

I feel that if we try to implement a full-blown orientation and gender studies-based system here, there may not be enough AMVs to actually find that would improve results over the three-level method, while also increasing the burden on the AMV editor to choose the correct situation. Even for creators who do want to be more specific, there is still the general tags system.

How does that sound?

We could end up never really finding an end to the number of possible genders and combinations, especially if we bring in topics of plurality, fluidity, and so many others. While some members of my communities might wish for a broader set of definitions, I feel starting out basic and then broadening where requested and appropriate makes sense. My host body's pronouns are she/her/fox/elf, and while we would love to see such levels of representation if we end up being the only ones to use such it isn't exactly a worthwhile feature. Room for expansion later, where warranted, I am fine with.

[Phade]

I think this is a great step in the right direction. I agree with Mango that an "intensity" scale may not work for all categories, and having easily-digestible terms instead would be a better option. I would feel very weird tagging my video with "blood and gore" over a cut or something (especially because my friends know I really can't handle gore). To your nosebleed example, you could have a nosebleed tag.

I'm glad we're moving in the right direction!!

We can change the "Blood/Gore" category label to just "Blood" and instruct people that "Gore" is lumped into the highest blood severity, since I don't think bloodless gore exists (but I could be wrong).

If you're worried about overwhelming the user with different tags, I suggest simplifying your approach. "90 and 243 words" IS a lot. You don't need that many descriptors. But also, words have different connotations and ultimately different meanings. I think there's a happy medium to be found there. I think you could cover everything you need to with, say, 50 words. Or, do 100 words and have different category headings so that someone isn't looking at all of the words at one time. For example, if I made a straight-up Romance video (no fighting or anything like that), I could just look under the Romance tag category (and you could even put "nosebleed" in there as well, to cover all of your bases) - I know I wouldn't need to look at the "Fighting" category or any of its tags, so I'll just keep that closed.

I am not that good at coming up with nuanced descriptor tags (hence my math-based systems). Do you think you could help prime my brain by coming up with some word/phrase tags that would describe the low, med, high severity values crossed with the low, med, high intensity? For example, the Blood/Gore type would look like this:

Hinted or glancing scrapes or nose bleed = [this word/phrase]
Clearly depicted scrapes or nose bleed = [this word/phrase]
Intense or dramatic scrapes or nose bleed = [this word/phrase]
Hinted or glancing blood from injury/wounds = [this word/phrase]
Clearly depicted blood from injury/wounds = [this word/phrase]
Intense or dramatic blood from injury/wounds = [this word/phrase]
Hinted or glancing mutilation or dismemberment = [this word/phrase]
Clearly depicted mutilation or dismemberment = [this word/phrase]
Intense or dramatic mutilation or dismemberment = [this word/phrase]

If the combination doesn't make sense, you can just put "none". Repeating this for all required category type severity intensities (that make sense) would really help get the standard tagging system tags going!

When put that way that does make sense. In this sense severity is really a generalized amount or focus, which I have no issues with.

Yay!! My crazy brain is starting to make sense to someone other than me!!

I certainly feel as defined and explained I could find what I'm looking for and filter out what I am not.

Excellent! That is the most important part about the "Content Descriptor" section. Standardization and ease of inclusion should make entering the descriptors consistent, resulting in quickly finding the exact video you are looking for.

We could end up never really finding an end to the number of possible genders and combinations, especially if we bring in topics of plurality, fluidity, and so many others. While some members of my communities might wish for a broader set of definitions, I feel starting out basic and then broadening where requested and appropriate makes sense. My host body's pronouns are she/her/fox/elf, and while we would love to see such levels of representation if we end up being the only ones to use such it isn't exactly a worthwhile feature. Room for expansion later, where warranted, I am fine with.

Cool! If the combinations of the three listed types of gender and orientation quickly find what people need, then it is a good start. If we find that more nuances are needed at a later time, we can add more or come up with a better system for gender, orientation, and other related aspects later after the initial launch.

What I can do is create a "word/phrase suggestion" tool that people can submit their ideas for combinations of category type severity intensities. This should come in VERY handy when trying to gather tags that mean "hinting male" or "intense female", as many synonyms clearly exist for different gender intensities.

[/me adds new feature to list...]

[Phade]

I am not that good at coming up with nuanced descriptor tags (hence my math-based systems). Do you think you could help prime my brain by coming up with some word/phrase tags that would describe the low, med, high severity values crossed with the low, med, high intensity?

Actually, never mind. I found a word cloud generator that created categorized lists for each of the content descriptor category type severities. I have a spreadsheet with 3,460 words and phrases that link to the categorized list. This can be used in a type-ahead input box or viewed for picking and choosing. It seems to have done a good job for a first pass rough draft, but the list is certainly not complete and I have not reviewed every single result (many of the words and phrases on the severe unpleasant ends of the scales were triggering). If the tag you are looking for is not available, you can add it to the list and the admins will review it for possible use and public view.

Hooray for the internet!

[Ileia]

Maybe I'm misunderstanding this but over three thousand words/phrases seems super overwhelming as part of the upload process. I generally like the idea of being able to tag/flag videos, but I feel like this is way overcomplicating it.

The main reason I heard from people who don't catalog their videos here (whether it was when I was a mod ten years ago and was PMing them by the hundreds to walk them through the process or the editors who I've chatted with IRL or online in the time since then) is that it was too intimidating. That it was hard to understand, a lot of steps, too technical, etc.

Maybe this could be rolled out as a couple of initial descriptors and expanded later? Especially if the idea is to have admin approval for additional ones.

Just want to re-state that I'm not against flags/tags and I think they are useful and I totally understand what you're trying to provide for viewers BUT I also know if we can't make cataloging simpler and easier for editors, we will struggle to get them to do it at all. Anything that we can shave off the process is going to go a long way during the initial stages of the new site.

[Phade]

Maybe I'm misunderstanding this but over three thousand words/phrases seems super overwhelming as part of the upload process. I generally like the idea of being able to tag/flag videos, but I feel like this is way overcomplicating it.

Yes, I completely agree, the way you describe it would be unusable. But, the vision I have is for the user's tags to automatically match the content descriptor to make things far more automated, thus reducing complexity.

My brain works in a way that I envision the definition of the content that matches the category type, severity, intensity, and frequency. This is why my initial proposal revolved around degrees of a concept. However, other people see things differently (which is perfectly fine!).

The use of tags is common and anticipated (and was in the original plan). One goal here was that when a user starts typing a tag in the input box, the site can use type-ahead to help autocomplete the word or phrase, much like Google does within the search bar. Having a library of thousands of common tags will allow a rich variety of type-ahead options.

For those creators who may be unsure of what tags to include, the site can also suggest common standard and related tags for a given anime or song. This would make entering tags easier for the creator if they naturally struggle to choose tags (I am one of those persons).

Once entered, the tags will map to the content descriptor values to auto-fill the likely choice there. For example, if someone tags their video with “consentacles”, the site already knows to autoselect the highest levels and intensities of body nudity, intimacy, and theme. Given common tags associated with the specific anime, the site may also preselect levels of blood, physical conflict, drugs, and alcohol. The song choice will allow autoselections for language and humor expression. The site could then analyze brightness changes over time within the video file itself to help preselect the “flashing” value. The same would happen with the optional category types. All that is left now is for the creator to confirm the values preselected by the tag content descriptor associations.

Let’s say the video creator put in only one tag, “consentacles”, and the content descriptor association system did not exist. Someone wanting to filter out that severity of AMV from their search results would either need to add “and not ‘concentacles’” to their search (assuming they even knew such a term existed) or they would need to know to filter a specific anime from their results.

But with the content descriptor system, anyone who wants to filter out that type of material would be able to select an upper-bound below the body nudity, intimacy, and theme values in the search without having to know all possible combinations of tags that revolve around those aspects of content descriptions. On the flipside, searching for these specific severities of content may introduce you to new terms for things that interest you.

(NOTE: I am not judging or kink-shaming anyone totally into consentacles material; I’m just using that as an example since the associated content descriptor correlations should be obvious to those who are familiar with the term.)

The main reason I heard from people who don't catalog their videos here (whether it was when I was a mod ten years ago and was PMing them by the hundreds to walk them through the process or the editors who I've chatted with IRL or online in the time since then) is that it was too intimidating. That it was hard to understand, a lot of steps, too technical, etc.

Again, I agree entirely. The site, in its current form, is written on technology available 20-25 years ago. Compared to modern methods, the input and upload processes blow big donkey... well, you get it. I wrote the site, the site is my baby, and I’ll be the first to tell you my baby is ugly, bigtime.

One major goal with the redesign is to make interactions minimal while maximizing information, and this will be in mind during every step of the UI/UX redesign.

Maybe this could be rolled out as a couple of initial descriptors and expanded later? Especially if the idea is to have admin approval for additional ones.

Hopefully, the above explanation clears up the purpose of seeding the tags with thousands of entries. The only time someone would see even portions of the list would be when using type-ahead, viewing common associated tags lists for their anime and music, or throwing some tags up as tooltip examples for the meaning of different content descriptors. However, pre-mapping thousands of words and phrases to a handful of standardized descriptors should ultimately make things easier for everyone involved, both entering tags and searching for content.

Yes, new tags can be entered by anyone, but they won't be standardized for reuse until an admin reviews them.

Just want to re-state that I'm not against flags/tags and I think they are useful and I totally understand what you're trying to provide for viewers BUT I also know if we can't make cataloging simpler and easier for editors, we will struggle to get them to do it at all. Anything that we can shave off the process is going to go a long way during the initial stages of the new site.

This is a valid concern, especially since nothing is visible right now and we're all depending on our own separate individual imaginations to envision the new system. So far, I’ve only been describing focused technical portions of the process and user interface in threads. I’ll create a post next describing the vision for the new entry method to show how the process should go from the current “it blows” to “it's beautiful”.