Just signed in for the week and I can agree, there was a lot of spam on the forums today. Spammers were seemingly too active in the past days for some reason. While the number of posts that slipped was higher than usual, even the number of posts that actually got blocked were so many that I had to spend about 30 mins today just reviewing those.
With that being said, the plugins and settings recommend by Discourse are already enabled and changing the sensitivity levels to further block spammers might cause more damage to regular users. I see a couple of genuine users stuck in the spam queue on a daily basis, so not sure if that’s a good idea.
As for you being blocked to report, the default limit is 20 posts a day. It doesn’t seem that the limit can be raised per user, so it will have to be global. I personally don’t have an issue with raising the limit to report spam, as there’s not a lot of users who would report something anyways.
I’ll bring this to the team’s attention. Thank you for reporting.
‘Mark as spam’ is still extremely ineffective under the current settings.
It appears to require multiple people to flag an item before hiding it, but with little participation outside of Netlify’s US business hours there is simply nobody else to corroborate or act upon reports.
Thanks for surfacing this Nathan! I’m working with our forums provider (Discourse) to resolve this issue expeditiously! I appreciate your proactivity in keeping our community clean and healthy!
It’s very obviously a Discourse issue and not one that’s specifically targeting Netlify, which you’ll see if you google for text related to the spam posts, as it reveals many Discourse powered communities.
I was just hopeful that if Discourse fob you off there may be an existing setting that could mitigate the noise. In particular the kinds of suggestions made in this thread for the ‘watched words’: https://meta.discourse.org/t/blocking-recent-wave-of-spam/314867
As while not inclusive, adding some Hindi characters to the list may block the bulk of the current spam.
However I did spot that the posts are initially created in gibberish English before being edited and replaced with the spam text in Hindi (which you can also confirm by looking at their ‘edit history’).
Edit:
In the thread I linked, others also identified that the bot ‘editing’ the posts is how it circumvents Discourse’s spam protections, so they’re almost certainly aware of it.