Hello! The name's starbyt, and I've been encountering a problem.

This isn't my first account on scratch, so it isn't the first time I've experienced this.

Filterbot.

Filterbot is perfectly amazing for the majority of one's scratch career. It prevents many accidental and purposeful inappropriate comments. However, for one's first two weeks on an account, filterbot is a nightmare. It says your comment is inappropriate even if you can't figure out why then eventually mutes you and you still don't understand why you were muted.

I feel this can be easily fixed if filterbot is slightly redesigned to highlight the inappropriate section of one's comment or say why a comment is inappropriate. It would help new scratchers better understand the guidelines and expectations of scratch.


Concise descr:

Filterbot needs to be more specific about what, specifically, is offensive about a user's comment.

Oooh, yess. All errors in either forums or comments should be highlighted so you know what to delete/change.

Concise descr:

Filterbot needs to be more specific about what, specifically, is offensive about a user's comment.

Support 101%!!! This would help stop the drama over unexplained mutings.

girlsruless wrote:

Support 101%!!! This would help stop the drama over unexplained mutings.

Yeah. I got muted once upon a time for a comment that was only one line and it took a whole 15 hours just for me to figure out what went wrong.

Support.

I do believe this is a dupe, but the link has alluded me. Furthermore, I do NOT support. If one accidentally typed a bad word and the filter said “NO! No say BLANK!” Then bam, the child learned a new bad word. Now he/she/they cannot use Scratch because of the very reason the filter exists. Telling a user what is inappropriate with a post could lead to users trying to break past the filterbot by making a list of all words blocked. The main point, telling children inappropriate words is bad.

CatGuyAnimates wrote:

I do NOT support. If one accidentally typed a bad word and the filter said “NO! No say BLANK!” Then bam, the child learned a new bad word. Now he/she/they cannot use Scratch because of the very reason the filter exists.

Learning that something is inappropriate ≠ using something inappropriate. In fact, if you knew that something is forbidden and you could get punished for doing that, you would avoid it, wouldn't you? And if you didn't know that there are rules, wouldn't the chance of you breaking them increase?

The main point, telling children inappropriate words is bad.

At least they would know what not to say. Sure, they can misuse the warning and speak rudely, but that's why the report button was created - to tell the Scratch Team that they broke the rules, and let them stop the user.

WindOctahedron wrote:

CatGuyAnimates wrote:

I do NOT support. If one accidentally typed a bad word and the filter said “NO! No say BLANK!” Then bam, the child learned a new bad word. Now he/she/they cannot use Scratch because of the very reason the filter exists.

Learning that something is inappropriate ≠ using something inappropriate. In fact, if you knew that something is forbidden and you could get punished for doing that, you would avoid it, wouldn't you? And if you didn't know that there are rules, wouldn't the chance of you breaking them increase?

The main point, telling children inappropriate words is bad.

At least they would know what not to say. Sure, they can misuse the warning and speak rudely, but that's why the report button was created - to tell the Scratch Team that they broke the rules, and let them stop the user.

Think of it from the point of view of a parent. Would you want your child learning bad words? Learning its inappropriate may cause some children to want to use it even more. Teaching children bad words when they accidentally type them may cause some parents to think Scratch is not friendly for children.

w

CatGuyAnimates wrote:

I do believe this is a dupe, but the link has alluded me. Furthermore, I do NOT support. If one accidentally typed a bad word and the filter said “NO! No say BLANK!” Then bam, the child learned a new bad word. Now he/she/they cannot use Scratch because of the very reason the filter exists.

That's a pretty big “if” right there. It's pretty hard to type bad words on accident, there's only one example I can think of (a contraction of ‘who are’), in which the child would just be confused and not realize it was a bad word. The way I imagine it, the filterbot would just highlight the offending portion without further explanation of why that portion is inappropriate. Also, the filter exists to prevent people from saying rude things, cussing each other out, etc.
Besides, if someone types something bad on accident once, you'd REALLY not want them to type that in a different, unfiltered setting. It's good for kids to learn what they shouldn't say, they don't have to know what it means and the way filterbot would hypothetically work they may not even realize it's a bad word.
And if the kid is the type to use bad words they find… there's no point trying to keep filterbot vague.

I would say this problem is outweighed by the frustration of people who already know what's inappropriate and might have mistyped something. It's really hard to piece through a comment and figure out what's wrong because of how vague it is.

CatGuyAnimates wrote:

Telling a user what is inappropriate with a post could lead to users trying to break past the filterbot by making a list of all words blocked. The main point, telling children inappropriate words is bad.

This might actually be a good thing, because scratch team can block/ban people who deliberately try to get past filterbot in order to insult people. wouldn't it be nice, though, to know what you can't say? Filterbot doesn't have to explain why something is bad, just tell the commenter what is inappropriate in their comments.

TLDR, Children do need to know what's not okay to say. If they're immature enough to go and use those words in real life, that's out of the domain of scratch and ST. And if they want to try to get past filterbot, that's not okay and scratch has other measures in place.

Filterbot's only purpose is to prevent people from saying certain rude things. If people knew what rude things they couldn't say, it would be more effective.

That being said, I do get where you're coming from and it is a valid concern parents could have! I'm glad you brought it up, I didn't really think about that.

starbyt wrote:

The way I imagine it, the filterbot would just highlight the offending portion without further explanation of why that portion is inappropriate.

That's a nice idea!
However, there is one problem: how long should the highlighted portion be?

I would support this too, or at least making the system less sensitive. I cannot say “click Milkywillow” on my game because it's “inappropriate” which makes no sense because Milkywillow is a name and I see nothing inappropriate about a “Milkywillow”.

Blue
Unicorns
Munch
Pickles

CatGuyAnimates wrote:

I do NOT support. If one accidentally typed a bad word and the filter said “NO! No say BLANK!” Then bam, the child learned a new bad word.

I imagine that if there was an inappropriate word in there, it would say something along the lines of “please do not use inappropriate language in your comments. Remember to be respectful when you post.”

I think this is a good idea, especially when it comes to links to unmoderated chat. The project/studio description filter will just say “Hmm… the bad word detector thinks there is a problem with your text. Please change it and remember to be respectful.” This doesn't really make sense in the case of a link to an umoderated chat site. Here's what I think it should say:
Link to unmoderated chat site: “The filterbot thinks there is a link to an unmoderated chat site in your text. Please do not give links to such sites on Scratch.”
Giving out personal information: “The filterbot thinks you are trying to give out personal information. For safety reasons, please keep such information private.”
Any other ideas?

Or maybe it could use the asterisk system, like on the forums?

girlsruless wrote:

Or maybe it could use the asterisk system, like on the forums?

That wouldn't be good- because someone could figure out what the bad word was if it was asterisked.

mtech22 wrote:

girlsruless wrote:

Or maybe it could use the asterisk system, like on the forums?

That wouldn't be good- because someone could figure out what the bad word was if it was asterisked.

Then they won't go around saying a bad word by accident, if they go around saying it knowingly, they could have said any bad word they know, if they know it's bad, they wont say it, and if they do, they're knowingly doing something wrong.

Support.

We want to know what happened.

Scenario:

some random person wrote:

*Says filterbot triggering word*

Filterbot wrote:

I saw you say something blacklisted!

some random person wrote:

IDK what happened to make my comment so bad? *Says filterbot triggering word again because they didn't know what triggered the bot*

filterbot wrote:

*Mutes person*

some random person wrote:

WHAT DID I DO WRONG FILTERBOT?!

The same person hates the filterbot and/or quits Scratch

If this suggesstion were to come true:

some random person wrote:

*Says filterbot triggering word*

Filterbot wrote:

Something is blacklisted, here, lemme show you: (INSERT BLACKLISTED WORD)

some random person wrote:

*Fixes writing to not include the filterbot triggering word.*

See? Better.

For this reason, I SUPPORT!

Boop bump.

warriorcats2155 wrote:

If this suggestion were to come true:

some random person wrote:

*Says filterbot triggering word*

Filterbot wrote:

Something is blacklisted, here, lemme show you: (INSERT BLACKLISTED WORD)

some random person wrote:

*Fixes writing to not include the filterbot triggering word.*

See? Better.

For this reason, I SUPPORT!

Perfection.

warriorcats2155 wrote:

[View post]

If this suggesstion were to come true:

some random person wrote:

*Says filterbot triggering word*

Filterbot wrote:

Something is blacklisted, here, lemme show you: (INSERT BLACKLISTED WORD)

some random person wrote:

*Fixes writing to not include the filterbot triggering word.*

Not quite.
What would happen is:

some random person wrote:

*Says filterbot triggering word*

Filterbot wrote:

Something is blacklisted, here, lemme show you: (INSERT BLACKLISTED WORD)

some random person wrote:

*Asks their parents what that word means, and tells them they learned it from scratch*

No support.

mybearworld wrote:

warriorcats2155 wrote:

[View post]

If this suggesstion were to come true:

some random person wrote:

*Says filterbot triggering word*

Filterbot wrote:

Something is blacklisted, here, lemme show you: (INSERT BLACKLISTED WORD)

some random person wrote:

*Fixes writing to not include the filterbot triggering word.*

Not quite.
What would happen is:

some random person wrote:

*Says filterbot triggering word*

Filterbot wrote:

Something is blacklisted, here, lemme show you: (INSERT BLACKLISTED WORD)

some random person wrote:

*Asks their parents what that word means, and tells them they learned it from scratch*

No support.

It would censor it, but only leave the first two letters so they can figure out what they should remove.