Lacewing
For chat, How to store Message Send by any user, and received by user in my server for Cyber Security ?
Printable View
Lacewing
For chat, How to store Message Send by any user, and received by user in my server for Cyber Security ?
What do you mean store message send by any user and received by user?
Do you mean having a text field where everything being sent and received shows up one line after another?
Do you mean the text sent and received being saved as a text file on the hard drive?
What do you mean for Cyber Security? Is it the name for your software or do you mean that you need encryption for the text being sent?
Decal,
thank for response.
yes of your both question.
When one person chat with other person and an illegal message (i.e. abuse, misapplication or sending them vulgar messages to female, or any. iit is under generate crime.).
In this case, how my server safe? and how I help to police or or crime deportment?
You can test for vulgar messages and females sending abuse at males by creating a word censor. Tou would have to make a list of all the words you dont like and add them to a exclusion list of sorts
Scilya151, How ?
How search from list, before sending vulgar messages and block it?
Hi
I have made a simple profanity check for you.
All the bad words a put in the list and it will check if the message contains any of them. If not it will send the message, if it contains any of the bad words then it will not send the message and give the user a reason why.
Thank you Decal
Be advised that multi-language users often get frustrated at words which are innocent in their language being seen as swears and starred out, or messages not delivered. For example, "Sega" has an innocent, but also a coarse meaning in Italian. Everyone else using it, to innocently talk about the makers of Sonic the Hedgehog will be baffled and upset when they get chastised for profanity that they were wholly unaware of. :)
Also, if you motive is really to stomp it out entirely (as opposed to, "being seen to be doing something"), good luck: Trolls are incredibly inventive. Read up on the way Dark Souls players leave humorous and offensive messages in asynchronous multiplayer mode using what is an intentionally incredibly limited vocabulary.
How solve this problem CeriJC??
Pradeep:
The problem in the second paragraph is tricky, unless you go for a "report player" function that effectively uses human beings "after the fact" as the mechanism to detect if offensive behaviour took place.
Or did you mean the first one?
I mean the first one.
It depends on how permissive you want to be. There are two extremes:
1. At one end of the spectrum (short of not doing anything), you *only* censor words that are swears in English (or whatever the 'native' language of the intended market for the application is). You live with the fact that some Spaniards might be using it and swearing every-other word. Conversely, the English speakers might be swearing in Spanish, unwittingly. When Spaniards complain about this (or about the other Spaniards' swearing being unblocked), you highlight that it is an English application and the censorship is only built to accommodate that.
2. You spend huge amounts of development time implementing language settings, with language-specific "blocked words" lists. You employ moderators with native-speaker proficiency in linguistic skills in each supported language, who actively get involved in reviewing complaints lodged against users purported to have sweared.
There's a good article on the general concept of word blocking (and some of the challenges in doing so) at:
http://banbuilder.com/
I hate to sound defeatist, but the conclusion I've reached is that people can say incredibly hateful, genuinely bigoted things (that you may rightly not want to be seen to be the publisher of) without ever uttering a single 'obscenity' and trying to block them by removing particular words doesn't really address the bad behaviour of users which is really what most people are concerned about. Likewise, I get extremely irritated when the meaning of text is obscured by asterisks and I have to read it twice to work out what it is saying, when it is actually very innocent. Ironically, this latter problem actually forces users to think of swears where there were none!
For example:
"S****horpe Software report that Arush ****dik has had difficulty ***embling the ***ets for his new game."
Therefore, at present, unless there is some other driver that dramatically increases the need to censor (EG it's a religious site, or app aimed at children), most developers stray nearer to '1' than '2'. If users/regulators complain, they cite the fact they're "doing something" and life goes on.
Initially, the situation will improve once AI becomes at least as reliable as a human moderator in differentiating between swearing and not. Even then, it will be some time before the AI is smart enough to detect the racist undertones in a seemingly innocuous phrase like, "Mow my lawn, boy." which is wholly dependent on the context.
Fingers crossed for an extension for this in Fusion 7.8! ;)