Making website with function for content uploading, how do I catch out pedos and terrorism? Any word - Developers

Hey, I am making a website where users can enter chat, and share content like images, it's not a porn site however, it's meant for peaceful purposes. But I know some creeps will realize they can share their stuff and junk on website. I want to implement message filter but I don't know where to find decent words list nor I want to write it myself. I want it to filter for words like: pedo, underage, 1yo - 17yo, rape, behead etc.. basically a words that average person wouldn't use every day. Once filter matches one of words - I will get email and sort it out asap. Also if matched content accompanies photo - blacklist that user until I unblacklist that person. So is there any words lists available? P.S - the reason I don't want to write words list myself because I am pretty sure there are some kind of secret phrases that are used by malicious people, there must be some kind of API out there available that is constantly updated with newly discovered bad words?

What about putting up a fake warning when the user uploads content? Something like "We are working closely with [some notorious security agency] to ensure no illegal content is uploaded. Failing to keep files family-friendly will result in possible legal persecution." and if the user decides to stop the upload you get notified :v:

[QUOTE=TheDrunkenOne;47597195]What about putting up a fake warning when the user uploads content? Something like "We are working closely with [some notorious security agency] to ensure no illegal content is uploaded. Failing to keep files family-friendly will result in possible legal persecution." and if the user decides to stop the upload you get notified :v:[/QUOTE] Would that actually stop anyone? That's kinda based on assumption that some freak will act like I want him to act, but that doesn't guarantee he/she will?

[QUOTE=arleitiss;47597198]Would that actually stop anyone? That's kinda based on assumption that some freak will act like I want him to act, but that doesn't guarantee he/she will?[/QUOTE] Hey some people are dumb enough to fall for it. No website can really stop people from uploading illegal content though.

[QUOTE=TheDrunkenOne;47597213]Hey some people are dumb enough to fall for it. No website can really stop people from uploading illegal content though.[/QUOTE] Well I suppose that could work, might give that idea a try + will also add quick report button that can be clicked by other users which will instantly trigger this report to email. Thank you.

[QUOTE=Yahnich;47597230]do you really expect them to tag obvious shit like pedo and behead and stuff though? i'm pretty sure most of these people use codewords[/QUOTE] well I don't really know how they behave or what they write, I was assuming they do?

[QUOTE=Yahnich;47597271]i'm pretty sure pedophiles and terrorists are a little more clever in their tagging, i saw a documentary where they mentioned barbie being used for example to look for little blonde girls and snow white for pale black haired girls so i don't think you'll really be able to use tags to filterblock images unless you look up every single codeword they use[/QUOTE] well shit, I was kinda expecting that but was hoping there is a way to filter them. Guess report button and warning will stop some but not all. Still better than nothing.

These kinds of problems are generally easier to deal with as you encounter them because then you can specifically target the issue you're having.

Try this: Are you uploading child pornography or other such illegal materials? If so, please tick the box below so we can ensure your data is encrypted with a password: Yes I am [x] Password:[********* ]

[QUOTE=Rocket;47598489]There is very little chance that someone will use your site for child porn or terrorism (especially terrorism). Deal with it when it comes up.[/QUOTE] You mean AFTER someone has uploaded something through a proxy then reported your site via a different IP/ID to the police / authorities?

[QUOTE=nuttyboffin;47598595]You mean AFTER someone has uploaded something through a proxy then reported your site via a different IP/ID to the police / authorities?[/QUOTE] Because uploading such stuff to facebook and then reporting facebook itself is gonna work.

[QUOTE=cartman300;47598663]Because uploading such stuff to facebook and then reporting facebook itself is gonna work.[/QUOTE] Facepunch has a big list of terms and conditions to say what can and can't be uploaded to keep themselves safe, along with an entire room full of legal staff. Not to mention that it would be pretty easy to upload something from one computer (public computer or remote access) and then report from a different identity. (calling up the police or whatever, I don't know how it's done.)

Yeah my main concern with not dealing with this now is that - if somebody uploads something illegal. Then someone else sees it and reports my site as some kind of extremist/pedo site - I will get fucked by law, isn't there like a point saying - you are responsible for your own site and what's hosted on it? I mean I assume even if I give police the details of uploader - they will let me go free without any problem?

[QUOTE=TheDrunkenOne;47597195]What about putting up a fake warning when the user uploads content? Something like "We are working closely with [some notorious security agency] to ensure no illegal content is uploaded. Failing to keep files family-friendly will result in possible legal persecution." and if the user decides to stop the upload you get notified :v:[/QUOTE] so basically [t]http://blog.yoocare.com/wp-content/uploads/2012/06/new-fbi-greendot-moneypak-virus.jpg[/t] only less bribing :v:

[QUOTE=Rocket;47599050]That's not how it works. There's a whole bunch of legal stuff here, but basically they would need to prove that OP had the intent to keep child porn on their website. Stating that child porn is not allowed on the website and showing that it's deleted whenever it comes up is enough for that. Sort of like how someone uploading pirated content to your website is OK as long as you delete it when you see it and specifically state that it's not allowed (DMCA safe harbor). If you are found to be assisting in illegal activity in any way, that's when things get bad. Compare 4chan to 8chan - 4chan deletes child porn whenever it comes up, while 8chan kind of just sweeps it under the rug. 8chan are going to get themselves into some serious legal trouble if they keep this up, but 4chan's pretty safe (at least from child porn related offenses). [editline]25th April 2015[/editline] Disclaimer: I am not a lawyer. Talk to a real lawyer if you are worried (which you should be doing anyways when you write your terms of service).[/QUOTE] that does sound more logical... however you still have the risk of dealing with legal butthurt.

[QUOTE=Rocket;47599050]Disclaimer: I am not a lawyer. Talk to a real lawyer if you are worried (which you should be doing anyways when you write your terms of service).[/QUOTE] So copy & pasting an existing ToS isn't a good idea (with find & replace of course)?

[QUOTE=Rocket;47599050]That's not how it works. There's a whole bunch of legal stuff here, but basically they would need to prove that OP had the intent to keep child porn on their website. Stating that child porn is not allowed on the website and showing that it's deleted whenever it comes up is enough for that. Sort of like how someone uploading pirated content to your website is OK as long as you delete it when you see it and specifically state that it's not allowed (DMCA safe harbor). If you are found to be assisting in illegal activity in any way, that's when things get bad. Compare 4chan to 8chan - 4chan deletes child porn whenever it comes up, while 8chan kind of just sweeps it under the rug. 8chan are going to get themselves into some serious legal trouble if they keep this up, but 4chan's pretty safe (at least from child porn related offenses). [editline]25th April 2015[/editline] Disclaimer: I am not a lawyer. Talk to a real lawyer if you are worried (which you should be doing anyways when you write your terms of service).[/QUOTE] 8chan, still abides to every US law, even if their response time is not nearly as fast as other places. The stuff they do have, are pedo communities, but as long as they don't post post or link any CP(which they generally dont) apparently its legal.

[QUOTE=DaMastez;47602802]So copy & pasting an existing ToS isn't a good idea (with find & replace of course)?[/QUOTE] That's copyright infringement in itself, ironically enough.

Damn this whole legal side around these topics seem to be very complicated and winded up. Too bad I am not a lawyer, I can't even write my own terms and conditions or even warnings decently.

Well perhaps I should've mentioned it: it's a college final year project, and our lecturers tend to smoke us with questions about project. During early prototype presentation I got at least 5 questions from lecturer related to security and content controlling basically.

to put it plain and simple just because you have tight security measures it won't stop 100% of the violation's as long as you have a good T&S condition agreement that should take the liability off of you but you do need to take action against it as soon as you are aware of it.

[QUOTE=arleitiss;47661223]Damn this whole legal side around these topics seem to be very complicated and winded up. Too bad I am not a lawyer, I can't even write my own terms and conditions or even warnings decently.[/QUOTE] i dont believe a terms of service needs to be in lawyerspeak, as long as its clear and clearly states that you agree to it when using the site. just look at imgurs terms, its literally just typed out that you cant upload that kind of stuff: [url]http://imgur.com/tos[/url], could just type something like that but more specific with a list of banned things. like already mentioned because its something that someone else uploaded to your site against your own terms of service, you wouldnt be liable unless you were made aware of it and didnt take action or i guess didnt have the measure in place to be able to take action on it. as for prevention, the only real way is a report system with moderators to manage them, as well as blocking certain search terms and blocking certain tags. another thing used by big companies is building databases of hashes from illegal images, although im pretty sure you could get around that by resaving the image with 1% less quality or changing a single pixel, but it still probably catches out a fair bit and is something you could talk about.

[QUOTE=Rocket;47666787]You can use perceptual hashes: [url]http://www.hackerfactor.com/blog/index.php?/archives/432-Looks-Like-It.html[/url] The thing with Imgur's TOS is that, while the summary is clearly written, the details section below (the actual contract you agree to when using their service) is definitely written by a lawyer.[/QUOTE] the section below makes no mention of illicit content and is about copyright and such, the top part is still a part of their agreed to terms of service rather than a summary. even so there are plenty of templates online and it wouldn't be massively hard to come up with something by looking at other sites tos.