NOTE: This isn't so much about spitballing about spam-prevention techniques - it's more of a question/plea specifically to Andrew: can we implement something on SSD?

Just a quick question here, I'd love for some feedback and to see what we can do about it.

It's about SPAM.

As SSD has been getting more known throughout the internet and worldwide and more widely received over time (which is fantastic!), it appears lately that the site has been bombarded with an incredible amount of SPAM questions and responses. I'm wondering if we're reaching a tipping point on SSD wherein we need to begin setting all posting of any sort to reCAPTCHA confirmation.

While it won't necessarily kill off all SPAM, it will likely have a huge positive effect and prevent clogging of the Q&A boards.

Personally, I don't mind typing in a little code every time if, in the big picture, it ensure the site stays relatively clean of outside infestation. Is there anything we can do?

I'd love to hear some thoughts on what we can do. I'm finding the SPAM flagging process to be a real nuisance - and furthermore I feel really bad for our admins who have to sort through all the requests, much more than they had to in the past (likely requests they're getting from many of us about the same posts too).

  • 1
    This question appears to be off-topic because it doesn't apply to the new version of the site. – Stavrosound Oct 1 '13 at 8:09

There was a thread about it on the C74 forums who had the same problem.


Some interesting ideas in there. Perhaps it depends on how customisable the forum software is.

How about if a post was marked as spam by 3 users with 1000+ reputation it would be immediately trashed.

Moderating all posts from first time users might be the only watertight way to go. you may then need to appoint more moderators, but I'm sure there are a bunch of us who would be up for chipping in to help, myself included.


Actually, a good picture CAPTCHA is very solid. The only way a robot/program could interpret a good picture CAPTCHA is using character recognition. Now, computer vision simply isn't that advanced and likely never will be (the logic would be extremely complex).

The one used here might be too simple (not messed up enough) or it's not requested every single time.


What @Internet Human said works fine, there is also the very specific question that you could ask, or a simple math problem. Sometimes those captchas can really be brutal.

  • Yeah, although I think that natural language processing is simpler (compared to picture CAPTCHAs and computer vision), in terms of how feasible it's for a program to know how to compute an answer. Math problems that use math terms should be fairly trivial (computers are made for solving math/logic), some sort of general trivia that deals with a very large set of possible answers (from the program's perspective) is perhaps more difficult for a robot. Not that this is particularly relevant, but just interesting. – Internet Human Jul 19 '13 at 20:52

Not the answer you're looking for? Browse other questions tagged or ask your own question.