I have no problem whatsoever with removing offending content from Snipt. I regularly remove and ban users who post illegal or otherwise inappropriate content on the site.
If they contact me, I'll happily remove anything even remotely "pro terrorist group ISIS", but as mentioned, I've received no such requests (and I don't proactively review private snippets for offensive content unless it's reported to me).
All terrorists suck, and they haven't any right to use services like yours to coordinate hurting other people. Nor does anyone have the right to incite violence. But people do have a right to political opinions, and some such might be interpreted as "remotely pro terrorist", without actually being coordination or incitement.
Finding the right lines for these determinations is hard, and I don't envy anyone the task of drawing them. And again, things may be moving fast now, and I understand that if there are real concerns about safety that a broad brush might be the first starting point for you. Even so, I hope that whatever policy you adopt iterates over the fullness of time to something that respects people's rights to express their opinions.
As a tiny (open-source) side project, this stuff is no fun, but I realize it's now a requirement if I want to continue leaving the site up.
Being the descendant in a family that was host of several terrorists, I have to disagree with you. See once the war was over and nazi germany got booted out, all those terrorists in my family instantly turned into resistance heroes.
Turns out those terrorists in my family were just ordinary people who got into this for reason as simple as having a bike allowing them to move around and relay messages, sort of an internet over bike and some of them got caught and executed for that.
"Insurgent" is used by the media quite frequently, but pundits have twisted this word into little more than an alias for "terrorist". My brain subconsciously makes this mapping despite my knowledge of a contrary meaning; I think maybe language is too pliable in politics.
Maybe we need yet new language to escape the difficulty of communicating without biases inherent in the vocabulary (our minds already do enough to short-circuit logic already without the English language making the matter worse). "Terrorism" and "unrest" are at such a forefront in the global consciousness that the whole subject is deserving of the few extra mental bytes that would be needed to disambiguate the mess.
If I were in charge of such a task, I would try to borrow words associated with certain well-known historical anecdotes in a hope that the pundits wouldn't simply redefine the words from under us. Hopefully such vocabulary would have enough of a semantic anchor in our generation to dispel such attempts, conscious or not.
To press on the issue further, it seems like it would be wise to engineer our language in a way that leads to better thinking. I suppose that would be too artificial to ever be accepted though.
Terrorists _target_ innocent people to gain advantage through intimidation. That's wrong, everywhere and always.
Not muslim/chiite/sunnite ? Not innocent.
I suspect that THAT will cause you issues.
:P
(1) a service provider monitoring its service or affirmatively seeking facts indicating infringing activity, except to the extent consistent with a standard technical measure complying with the provisions of subsection (i); or
(2) a service provider gaining access to, removing, or disabling access to material in cases in which such conduct is prohibited by law."
Dude, you desperately need a designated agent. You just aren't under safe harbor at the moment.
The only way you break safe harbor is if you know of offending material, eg by a filter or having been alerted to it, and leave that material in place. If you run content filters (such as at the point of upload), then you have to be more careful.
The difference being between after-the-fact moderation and pre-screening. If you pre-screen you take on a greater burden of risk, in terms of copyright holders being able to argue you knew about something and didn't remove it, or should have known.
The safe harbors provided by S512 and S230 are absolute and are not impeded by proactive (or reactive) administration. This is part of their reason for their existence.