Forum Moderators: rogerd
A few months ago, however, I noticed a marked increase in the amount of comment spam. As a concerned owner, I took steps to fight the spam, including blocking certain countries (via a geoIP database) and blocking individual IPs. That seemed to curb the flow of spam for a time, but in the last few weeks the spam seems to have got worse again.
What can I do to further bolster my defences against comment spam? I'd prefer to implement a solution on my own server than link into central spam databases.
Some options I have pondered:
Block by bad-word in the comment- I should have done this a while ago, but I'll search for certain bad words in the comment (like the V word) and block those comments.
Email confirmation- sending an email out for every comment and requiring commenters to confirm their email address. The only hassle with this is that it's a hassle! Might take a lot of the fun and spontaneity of commenting away.
Somehow harnessing the community spirit of the site to fight spam? Any other ideas would be most welcome!
To convert your blog in a stealth blog, you first need to identify how people find your blog. Many blog spammers find blogs through the search engines. Commonly used packages like WordPress have their own footprint. Specific words and phrases are the same on 99% of all installations and blog spammers search on these phrases in the search engines. This can be words from the normal pages, but for example also the login page to edit the blog.
To put your blog in stealth mode, change those words in something different. Also put a robots "noindex,follow" header on all pages that you don't want to be indexed by the search engines, like login and registration pages, default help pages etc.
I was amazed by the results. On average I have one spammer per two weeks left. And ironically most of those spammers reach my blog when searching on "captcha control" and "comment spam" because I have two posts on my blog about these two issues.
Another thing I find useful is to have a non-standard captcha. Those awful image captchas are everywhere, and they are, for the most part, crackable.
I hadn't thought of doing a captcha code... will look into that.
For the short term, I think I'm just going to check comments content for blacklisted words and automatically add the IP address of the commenter to my ban list if a blacklisted word is found. There might be a little collateral damage, but that would be easily rectified.
I'm also going to disable any HTML code (including links) in comments.
Two weeks after implementing the automatic IP-banning for using blacklisted words, I've caught over 1500 spammer's IP addresses in my little spam trap. I've also prevented users from putting links in comments, although haven't automatically blocked them for doing this.
So far I've had about 2 spam comments in two weeks get through the net, compared to 20 a day prior to implementing this system. And no mistakenly banned IPs at all!
So it seems as if my site is winning the fight against spammers!
I'm actually starting to think that the most effective part of the changes I made was to remove links in comments. Once spammers realise that there is no point in spamming the blogs, they'll soon leave (and indeed they have!).