Recent incidents have made me think that we need a strengthening of the security protocols. While something can be made at the software level, something needs to be made at the social level. So, I will summarize what we have up to now, and what we can do. I would like to receive your feedback, because it is important to see where Jaderune is heading to.
So, everybody knows that we have been under attack since uncountable months. In December, a huge work was done, purging something like 59,500 accounts, most of which were just bots (non human). Then, we observed some crazy people trying to re-register new accounts, because they found out they had been banned (probably the bot's commanders). In the registration process, we added a security code that changes every time. So, bots cannot pass through. Therefore, now we only have real human beings who have to register manually. However, many register, try to post spam, links, etc., and get automatically deleted. Others are smarter and create "dormant accounts", ready to be used when they need it. A lot of those dormant account come from China, where lot of young people are paid to hack into forums and websites. They cost less than writing a robot, and that's why they come to Jaderune. This led to a stricter enforcement of the software rules. All dormant accounts (without posts) for 15 days are automatically purged. This was a hard decision with the site owner, because the purged "false positives" (casualities, genuine accounts) skyrocketed. It's rare, but sometimes we get a complaint from a genuine user whose account was terminated because of "profile abuse" - which simply means no posts in 15 days since subscription. But it is clearly stated in the registration form in red, and it kept the forum very clean and safe until now.
Then, we had one or a few trolls who come with a hidden agenda and try to discourage the participants usually with flagging an entire people with some preconceived ideas. Mostly they talk about money, poverty, etc. Members become disgusted, and we lose users, because instead of a nice place the forum becomes a place of war. That is an abusive behavior, because it offends the soul of a people. One is entitled to such thoughts and can share them privately, but in public it is a violation of the forum rules. So, we have software that flags certain words and check the identity of the offender. If it is found that he uses proxies, fake names, etc., he is asked to be verified by sending us a valid ID card scan (passport, ID card, driving license, etc.). If they don't comply, their accounts gets disabled. Here we can make some improvements. We can make a module that automatically checks every user IP for proxies, hidden servers, etc., and purge their account automatically. Also, we can state more clearly in the registration rules that it is forbidden to talk about money issues and that at any time, at simple request, for verification purposes they are obliged to show their ID card.
Recently, we spotted out DannyJones. I have no doubt he was a scammer, and probably BigBlastGuy. But I still have a doubt about Gary2310. One reason for all is that, if one is a cheater, he does not use his home DSL connection straight away, knowing that he will get caught. But... he got resented of the verification check and does not collaborate showing us a valid ID card. His main objection is that he is afraid of identity theft. I would like to develop some protocols that ensure us a proper security check without making the people feel uncomfortable.
One of my question is, is it so unusual in the US to ask for a document? I know about the identity theft in the US, indeed, but any office asks for a document anyways (from banks to anything). In my country there is no identity theft, because protocols are very strict. And it is very normal to show one's ID. So, I would like to know everybody's opinion and proposals.
The night brought me another idea. We could ask the complete generalities and phone number of the person as they appear in whitepages.com, and somebody of us could call to ensure that the person is real. Google and many sites work with SMS verification and operator calling. Obviously it is a cost, but we would do it only in extreme cases such as the recent ones.
What do you think?