Can some people try it out.
I'm trying to test an online list:

https://scratch.mit.edu/projects/89415193/

I'm willing to help!

Works fine for me.

The chat room is not safe, sorry. People are able to cuss on it.

1004587 wrote:

The chat room is not safe, sorry. People are able to cuss on it.

He has it set up so people can be banned from the project, but I would agree he needs to blacklist swear words.

BKFighter wrote:

1004587 wrote:

The chat room is not safe, sorry. People are able to cuss on it.

He has it set up so people can be banned from the project, but I would agree he needs to blacklist swear words.

No, he needs to whitelist what is acceptable, otherwise The Powers That Be will ban it. Only whitelist-style chats are allowed.

gtoal wrote:

BKFighter wrote:

1004587 wrote:

The chat room is not safe, sorry. People are able to cuss on it.

He has it set up so people can be banned from the project, but I would agree he needs to blacklist swear words.

No, he needs to whitelist what is acceptable, otherwise The Powers That Be will ban it. Only whitelist-style chats are allowed.

because, otherwise people could see inside and view the list and might use those words in real life.

dinogirl112 wrote:

gtoal wrote:

BKFighter wrote:

1004587 wrote:

The chat room is not safe, sorry. People are able to cuss on it.

He has it set up so people can be banned from the project, but I would agree he needs to blacklist swear words.

No, he needs to whitelist what is acceptable, otherwise The Powers That Be will ban it. Only whitelist-style chats are allowed.

because, otherwise people could see inside and view the list and might use those words in real life.

But it's also impossible to include all versions in a blacklist.

Imagine ‘boat’ was a rude word - you'd need to chack for boat, and bo4t and b04t and so on.

Even a whitelist isn't perfect though if it's just word-based. Words can be fine in one context and rude in conjunction - ‘dim wit’ for example. Using phrases rather than words would be my preference.

I believe that Blacklisted chat should be allowed, as long as it has these requirements:

> You can ban people completely from using the project and/or mute them so that they can still see what people are saying but THEY cannot say anything to other people using the project itself.
> To not just be able to detect obvious cuss words, but also to detect slight variations, as what deck26 mentioned with variants of the word: boat (b0at, b04t, bo4t etcetera)
> A method of storing all the blacklisted words/phrases inside the project so that if a young child were to visit the project and click see inside, then they could not just simply open up a list and bam, he can see all the rudest words out there. Instead you would have to encrypt the words into numbers or symbols or just jumbled up words rather than plain text and then store the encrypted versions of the words into the list. When you enter a message, the decrypting engine deciphers the meaningless string of characters, store the decrypted cuss word into a temporary variable, and then scans the message to see if the sender has used that word. If so, it stops him from sending it, if not then it moves on to the next encrypted word in the list. As soon as the engine is done with finding rude words it deletes the contents of the temporary variable.

An alternative for the third requirement, which I believe is even better than the one above, is to encrypt the message with the same style that is used to encrypt the swear words. Then the engine compares the string of characters formed by the encryption to see if it contains a swear word. This way, the swear word is NEVER actually decrypted, meaning that you can never see it in plain text, even if you click see inside. This is better than the previous requirement in two ways: The last one makes it so the project shows the swear word, in which if the child somehow stops the project with the red octagon while the engine that detects swear words is currently working and has assigned a swear word to a variable in plain text he/she will be able to see it. The second way this is better than the other alternative is because even though the engine may have been placed in a custom block with “run without screen refresh” enabled to make it run as fast as possible, scratch still has limitations on the speed it runs scripts. The message could be really long, or the user's computer may be very slow, both possibly resulting in the engine running for long periods of time, meaning that it could take, for example, three seconds for the engine to check for just one word, meaning that word is visible in the variable as plain text for three seconds; same applies to all the other words.

Ultimately, you can still have a safe chat using the blacklisted method, as long as it has a way of banning/muting people, and for detecting for swear words WITHOUT actually displaying them inside the project in plain text. I have used whitelisted chats before and to be honest, they kind of seem slightly boring, and sometimes I like to use phrases like omg and lol and xD because that's what i'm like, and with whitelisted chat you can't do that. Also it can take a hectic amount of time adding every single word of the dictionary into one single list. In my opinion, i'm all in for black listed chat.

SonicCreeper7 wrote:

I believe that Blacklisted chat should be allowed, as long as it has these requirements:

> You can ban people completely from using the project and/or mute them so that they can still see what people are saying but THEY cannot say anything to other people using the project itself.
> To not just be able to detect obvious cuss words, but also to detect slight variations, as what deck26 mentioned with variants of the word: boat (b0at, b04t, bo4t etcetera)
> A method of storing all the blacklisted words/phrases inside the project so that if a young child were to visit the project and click see inside, then they could not just simply open up a list and bam, he can see all the rudest words out there. Instead you would have to encrypt the words into numbers or symbols or just jumbled up words rather than plain text and then store the encrypted versions of the words into the list. When you enter a message, the decrypting engine deciphers the meaningless string of characters, store the decrypted cuss word into a temporary variable, and then scans the message to see if the sender has used that word. If so, it stops him from sending it, if not then it moves on to the next encrypted word in the list. As soon as the engine is done with finding rude words it deletes the contents of the temporary variable.

An alternative for the third requirement, which I believe is even better than the one above, is to encrypt the message with the same style that is used to encrypt the swear words. Then the engine compares the string of characters formed by the encryption to see if it contains a swear word. This way, the swear word is NEVER actually decrypted, meaning that you can never see it in plain text, even if you click see inside. This is better than the previous requirement in two ways: The last one makes it so the project shows the swear word, in which if the child somehow stops the project with the red octagon while the engine that detects swear words is currently working and has assigned a swear word to a variable in plain text he/she will be able to see it. The second way this is better than the other alternative is because even though the engine may have been placed in a custom block with “run without screen refresh” enabled to make it run as fast as possible, scratch still has limitations on the speed it runs scripts. The message could be really long, or the user's computer may be very slow, both possibly resulting in the engine running for long periods of time, meaning that it could take, for example, three seconds for the engine to check for just one word, meaning that word is visible in the variable as plain text for three seconds; same applies to all the other words.

Ultimately, you can still have a safe chat using the blacklisted method, as long as it has a way of banning/muting people, and for detecting for swear words WITHOUT actually displaying them inside the project in plain text. I have used whitelisted chats before and to be honest, they kind of seem slightly boring, and sometimes I like to use phrases like omg and lol and xD because that's what i'm like, and with whitelisted chat you can't do that. Also it can take a hectic amount of time adding every single word of the dictionary into one single list. In my opinion, i'm all in for black listed chat.

The decision has been made and I would think it is highly unlikely to be reversed. You cannot possibly check for all possible versions of swear words. The whitelist method is the only safe one and even that isn't perfect as I stated earlier.

This really belongs in Suggestions but I'm sure if you search has already been dealt with extensively there.

I see no objection to having ‘lol’ in a whitelist. I guess ‘omg’ is trickier for some people so I don't know what the official view on that would be. If you can use them here and no one object it should generally be OK.

Why not hide the blacklist under a comment in the scripts so no one can see the list?

NathanCoded wrote:

Why not hide the blacklist under a comment in the scripts so no one can see the list?

Because no one would ever think of right clicking and cleaning up the display of scripts!

NathanCoded wrote:

Why not hide the blacklist under a comment in the scripts so no one can see the list?

What if somebody moves the comment. Also lists are displayed on the project, not in the scripts
edit: lol, ninja'd by @deck26