Mark as spoiler / sensitive content #944

Open
opened 5 months ago by ullarah · 8 comments
Owner

Just a thought that perhaps we could add a "sensitive content" option for posts. Not sure it could be done. Perhaps adding a (%nsfw) or some such in the beginning of a post?

For Yarn.social this would automatically collapse the twt content with a "sensitive content click to open" message.

Just a thought that perhaps we could add a "sensitive content" option for posts. Not sure it could be done. Perhaps adding a (%nsfw) or some such in the beginning of a post? For Yarn.social this would automatically collapse the twt content with a "sensitive content click to open" message.
ullarah added the
area/ui
discussion
enhancement
labels 5 months ago
ullarah added this to the 0.15 milestone 5 months ago
Poster
Owner

Thinking about this more. Would this be one of those extensions on top of the twtxt specification?

Since the 'spoiler' tag could be used for other clients as well however they want to deal with them.

Thinking about this more. Would this be one of those extensions on top of the twtxt specification? Since the 'spoiler' tag could be used for other clients as well however they want to deal with them.
Owner

What does "sensitive content" mean here in this case?

What does "sensitive content" mean here in this case?
Poster
Owner

What does "sensitive content" mean here in this case?

Anything that could potentially either have emotional triggers, perhaps a nsfw/nsfl imagery and text, or something like a movie spoiler (which has been discussed before).

I'm not quite sure how CLI clients would tackle this kind of content though.

> What does "sensitive content" mean here in this case? Anything that could potentially either have emotional triggers, perhaps a nsfw/nsfl imagery and text, or something like a movie spoiler (which has been discussed before). I'm not quite sure how CLI clients would tackle this kind of content though.
Owner

I see... Hmmm 🤔 I'm not so sure on this one per se, let's ask the community? 😅

I see... Hmmm 🤔 I'm not so sure on this one per se, let's ask the community? 😅

Since most of the spoiler/NSFW stuff would be images on way could be to add NSFW or SPOILER to the alt file in the markdown of an image: ![NSFW](URL/file.jpg]

Since most of the spoiler/NSFW stuff would be images on way could be to add `NSFW` or `SPOILER` to the alt file in the markdown of an image: `![NSFW](URL/file.jpg]`
Owner

In the weekly call, I floated the idea of emulating Discord's ||double pipe syntax|| for spoilers. Text enclosed in double pipes would be an opaque box until the user clicked or hovered over it.

On Discord, double piping an image URL embeds the image as a spoiler. When uploading an image, there's a checkbox that does that for you. Images marked as spoiler are blurred until clicked by the user. ||![That method would be clunky](https://example.com/image.png)|| using full Markdown syntax, though. @sorenpeter 's alt text idea is a much more elegant solution.

"Inline spoilers", as I called them, raise concerns for the JavaScriptless experience on the Web client. I can't think of a way to implement them without a checkbox hack.

CommonMark's Deployed Extensions list mentions >! as an alternate method of handling spoilers. This is how spoilers work on StackExchange. If a >! marks a blockquote as a spoiler, maybe >!nsfw could mark it as NSFW.

StackExchange's implementation does require JavaScript, but it doesn't need to. It could be rendered in the web client as a <details> tag, as was mentioned in the call. No JavaScript required.

In my opinion, from the perspective of a user, >! is the way to go. But I'm not the one implementing it. What do you all think?

In the weekly call, I floated the idea of emulating Discord's `||double pipe syntax||` for spoilers. Text enclosed in double pipes would be an opaque box until the user clicked or hovered over it. On Discord, double piping an image URL embeds the image as a spoiler. When uploading an image, there's a checkbox that does that for you. Images marked as spoiler are blurred until clicked by the user. `||![That method would be clunky](https://example.com/image.png)||` using full Markdown syntax, though. @sorenpeter 's alt text idea is a much more elegant solution. "Inline spoilers", as I called them, raise concerns for the JavaScriptless experience on the Web client. I can't think of a way to implement them without a checkbox hack. CommonMark's [Deployed Extensions](https://github.com/commonmark/commonmark-spec/wiki/Deployed-Extensions#paragraph=) list mentions `>!` as an alternate method of handling spoilers. This is how spoilers work on [StackExchange](https://meta.stackexchange.com/questions/72877/whats-the-exact-syntax-for-spoiler-markup). If a `>!` marks a blockquote as a spoiler, maybe `>!nsfw` could mark it as NSFW. StackExchange's implementation does require JavaScript, but it doesn't need to. It could be rendered in the web client as a [\<details\>](https://developer.mozilla.org/en-US/docs/Web/HTML/Element/details) tag, as was mentioned in the call. No JavaScript required. In my opinion, from the perspective of a user, `>!` is the way to go. But I'm not the one implementing it. What do you all think?
Owner

My first reaction is: I don't want this. Somehow reminds me on censorship.

I'm wondering if this is really worth the effort. Can it be even done in a meaningful manner to work kind of reliably? Getting this right requires a lot of things I imagine. First, users need to flag their own twts with the appropriate tags. Remembering to do that is hard enough alone.

And then there's the question of what is considered "sensitive" and what not. We already saw that there are several categories of sensitive material. For some the community™ might get some common agreement, but other topics are probably sharing too diverse opinions. Take movie spoilers and strong language, I don't give a fuck for that, SCNR. ]:->

So then, let's say you know all the sensitive content categories your post belongs to, you need to find out which tags to apply (okay, will get easier the more often you do it). If you flagged it correctly, the filtering system will work and all is nice and dandy. But then, if you forgot, it's displayed anyways. Realistically, the editing feature is rudimentary for ages (or was it fixed in the meantime?), so repairing tags afterwards is tricky these days.

If you encounter a spoiler, just don't read on. It's easy enough. That's something I do all the time. Oh, that topic doesn't interest me, let me skip and move on. You need to open up the collapsed twt anyways to see which movie it is about. Maybe you've seen it already and then it's not a spoiler for you. But maybe you haven't, than you need to stop reading on, exactly like above. By collapsing it by default, you don't get any benefit, you've just made more work for yourself to expand it manually.

Regarding NSWF content, simply just don't browse your private stuff at work. :-) Solved. (Separating work and private things works perfectly for me.)

Anyways, I won't add support for this in my client. But of course, others might.

My first reaction is: I don't want this. Somehow reminds me on censorship. I'm wondering if this is really worth the effort. Can it be even done in a meaningful manner to work kind of reliably? Getting this right requires a lot of things I imagine. First, users need to flag their own twts with the appropriate tags. Remembering to do that is hard enough alone. And then there's the question of what is considered "sensitive" and what not. We already saw that there are several categories of sensitive material. For some the community™ might get some common agreement, but other topics are probably sharing too diverse opinions. Take movie spoilers and strong language, I don't give a fuck for that, SCNR. ]:-> So then, let's say you know all the sensitive content categories your post belongs to, you need to find out which tags to apply (okay, will get easier the more often you do it). If you flagged it correctly, the filtering system will work and all is nice and dandy. But then, if you forgot, it's displayed anyways. Realistically, the editing feature is rudimentary for ages (or was it fixed in the meantime?), so repairing tags afterwards is tricky these days. If you encounter a spoiler, just don't read on. It's easy enough. That's something I do all the time. Oh, that topic doesn't interest me, let me skip and move on. You need to open up the collapsed twt anyways to see which movie it is about. Maybe you've seen it already and then it's not a spoiler for you. But maybe you haven't, than you need to stop reading on, exactly like above. By collapsing it by default, you don't get any benefit, you've just made more work for yourself to expand it manually. Regarding NSWF content, simply just don't browse your private stuff at work. :-) Solved. (Separating work and private things works perfectly for me.) Anyways, I won't add support for this in my client. But of course, others might.
Owner

Maybe spoiler tags aren't necessary, but there should absolutely be a function to mark a twt as NSFW. It isn't taken literally most of the time. Many people, myself included, just don't want to see porn or gore on their feed. Right now, nobody's posting that kind of thing. That will change as the network grows and it's something that should be addressed.

Maybe spoiler tags aren't necessary, but there should absolutely be a function to mark a twt as NSFW. It isn't taken literally most of the time. Many people, myself included, just don't want to see porn or gore on their feed. Right now, nobody's posting that kind of thing. That **will** change as the network grows and it's something that should be addressed.
Sign in to join this conversation.
No Milestone
No Assignees
5 Participants
Notifications
Due Date

No due date set.

Dependencies

No dependencies set.

Reference: yarnsocial/yarn#944
Loading…
There is no content yet.