I’m having an issue with a discount kitchen spammer. They keep changing the ip and email address however the content of the message is always the same. Is there some way to just instant ban them if they link to the spam website?
1 Like
This would be useful. Something like the “white listed spam host domains” setting, except blacklisted.
Are you 100% sure they are using the same URL? Or that there is a URL in there consistently and it is the same? Usually the url is the most fluid part of the spam in my experience, they change them super fast as they get blacklisted.
Yes, I am 100% sure. It’s always www.exdisplaykitchens1.co.uk
I don’t know why they target appium because discuss.appium.io doesn’t have anything to do with kitchens.
I need some better way to deal with the spam because it’s becoming more frequent.
3 Likes
SEO manipulation. More links to their site == more prominence in search results for “kitchens”.
You’re right. I found this on the website:
We are a SEO + Marketing company based in indonesia
I want to auto ban all the keywords/phone numbers/urls they’re using.
Edit: On google groups, there’s an option to allow a users first post to be manually screened before it goes live. Maybe that’d be a more general solution. They register an account and only post once, and it’s always based on whatever they’re spamming about.
Although their SEO tricks actually won’t work with the default Discourse setting of “add rel nofollow to user content”.
I double checked and I have that setting enabled. Unfortunately it doesn’t demotivate them. 
Neither does SitePoint, yet
exdisplaykitchens1.co.uk do_nothing 13 2015-01-05 20:49:51 2014-09-17 12:19:15
That’s 13 matches from Sept 17 to Jan 5
persistent buggers
Having System Flag posts containing the word “kitchen” would be a godsend for us.
Looks like the spam made it here: https://meta.discourse.org/t/cut-price-kitchens-uk/23772
It has been flagged (I flagged it at least).
I noticed this bullshit kitchen spam on Ember, Docker and Here. This strongly indicates this is somehow automated and there is a new bot of sorts targeting us.
It’s not just happening on Discourse it’s popped up on a phpbb board I moderate on as well on Tuesday. This ‘kitchen spam’ has been happening on other UK boards I’ve moderated professionally for a few years now, it’s popped up all over.
Yes, it has been going on for so long I’ve forgotten when it started.
At vB, being a web dev forum, the likelihood of a legit use of the word “kitchen” was slim enough that a word filter block worked great with no false positives that I can recall.
Of course if a forum does discuss kitchens it would cause problems.
1 Like
I had forgotten about that plugin! Thanks for the reminder.
1 Like
Akismet sounds like the best solution, thanks for the link!
1 Like
Yeah better to bring the Akismet plugin into the fold as @eviltrout is currently doing (after tagger) rather than building out a lot of redundant text detection stuff.
I can see a use case for blocking arbitrary text with a custom message, this would fix your problem with the inserted default text links, too, @cpradio. Kind of a dangerous feature in some ways but if we say you can’t block text unless it is 6+ chars in length, less so.
In general spammers, as dumb as many of them are, are still a bit smarter than using the same URL over and over. We banned spam URLs for years at Stack Exchange and the repeat hit rate was tiny. Once the URLs are burned they tend to switch really fast.
2 Likes
Curious with the akismet plugin, if we are to start using it now, what will happen when it is bundled by default?
No idea, the main difference from my perspective is you have a stable of official plugins that are guaranteed to work with each new release and probably the betas too.
Agreed. Low priority for me though (as I think you are referring to the “link description here” that @Mittineague actually posted), but if you have something else in mind… please do say so (I’m just not connecting the dots).
I like the idea of keyword based auto spam, but I like it more of: Hide the post automatically if said keyword is in the post. That way false positives can be dealt with and the word removed if it does too much harm.