[ art / civ / cult / cyb / diy / drg / feels / layer / lit / λ / q / r / sci / sec / tech / w / zzz ] archive provided by lainchan.jp

lainchan archive - /q/ - 12613

File: 1483194777812.png (278.63 KB, 82x300, 1464682300784.jpg)


/sec/ is being spammed with some CP soykaf.

Yes, it is very ironic.


Yes, and it is already being taken care off.


Is it just /sec/ being pedo spammed?
It always seems to be /sec/.


Not it is not just /sec/ , /r/ and other boards get occasional spam but not by the same botnet.


Any estimate on when this issue will be solved? It might be just me, but I feel as though the images being posted are becoming more and more explicit.


The images are continuing to be posted. Can some sort of security feature be added to prevent or at least deter this? Like captchas of some sort? It's getting a little frustrating when I pull up what's supposed to be a sfw work board in public and there's CP on it. The images aren't being pulled fast enough.


File: 1487438898119.png (132.43 KB, 200x112, xae65ypfqycy.jpg)

Also, I would highly suggest disabling the recent image feature on the front page or something until this is dealt with. I don't think anybody wants to see CP there.


For captcha something like https://2captcha.com/ can be used. (1$ for 1000 captchas)

Images can be completely disabled on laincahn. This is most effective solution. Or all posting must be premoderated.


This is because users were complaining about the post filters, which were there to stop this from happening. It's hard to be on the lookout 24/7, which is why they stay there for a bit, most of the time. Captchas probably aren't the best solution due the nature of this chan, so there's a decision between having this happen once in a while or users getting their posts blocked in a way that's often hard to get around.


>having this happen once in a while
It feels like i'm reporting one cp spam thread per hour.


Most of it's handled within minutes of being reported.

It usually only stays up for a few to twenty minutes and it's not frequent.


They have gotten significantly more frequent with the removal of the filter, the frequency can be seen on the public ban page. There have been around 14 in the last 12 hours, which is the highest frequency I am aware of, but I have no idea what to take from it.


The issue will never be solved, spam on imageboards is a perpetual arms race.

In terms of what I have done to combat this in the last 16 hours, I have attempted to enabled tesseract OCR in order to see if OCR can detect the advertisting on the image and add it to the post body in order to make post filters based on that.

So far that doesn't look like it is working correctly.

I have also added specific name and subject filters based on the current dictionary word the botnet is upto, which is only a temporary measure that causes it to jump to the next word.

I have also added a specific filename filter which is also only temporarily until the botnet gets to the next non matching image.

Recent Image for frontpage has been set to 0, and will remain that way forever, unless requested otherwise, see my initial comment.

Other post filters I had in place were changed or removed because lainons complained or attempted to ignore them (minimum post length filter).


Maybe antispam filters must fail silently? I don't fink what bots recheck post result. Of course this is only for filters what created specially for that spam.


File: 1488469212387.png (149.88 KB, 200x113, shitposting-agencies.jpg)

>Recent Image for frontpage has been set to 0, and will remain that way forever, unless requested otherwise, see my initial comment.
Was wondering why this was the case. Assumed it was a bug.

Hope it goes without saying but it seems evident that /sec/ is being spammed by political adversaries in an effort to encourage Lain to adopt bo‍tn‍et friendly user identification practices such as forcing captchas or cookies. The only individuals who would feel "secure" doing this are politically motivated agencies who are legally protected from the consequences of their actions.

Despite the issues with spam, I think having no images on the frontpage of an imageboard is a mistake. Frontpage looks like soykaf right now and would be greatly improved if it featured more than just text-based content.

Perhaps by selectively allowing content that has existed beyond a certain time threshold and/or viewed a certain number of times without being reported to appear on the frontpage, you could prevent having the vast majority of b& or otherwise problematic content show up on the frontpage. Content flagged for moderation wouldn't show up on the frontpage. Whatever you do, for what its worth I think the frontpage should be more dynamic.

Making a simple, obvious reporting interface on the frontpage to report b& content if it appears would allow the images to be dealt with more expediently.