Moderation criteria?

On October 1st 2024, as I was answering a comment , I was surprised to see my post suspended for moderation. I read it twice to check if there were inadvertently 5-letter words or other offending terms. I explained the subtleties of frame anchors and could not understand why it was automatically sent to moderator’s attention.

Can somebody tell me what triggers moderation? Have I been blacklisted? If so, why? (so that I can apologise).

As far as I know, there is no moderator (except @sophi, probably for technical reason).

Hi, your post was flagged by the system: This post included a Watched Word. See your list of watched words. But I approved it, didn’t I?
You don’t have been blacklisted, on the contrary, thanks a lot for all your work on this forum!
And yes, I’m the moderator here.

Thanks for quick reply.
Yes, I noticed my post was approved. I was confident for that.

Just a small problem: I can’t access the referenced page (because it is in the /admin/ part?)

Yes, I think it’s the reason. But I wanted to download the file and passed it here, however the file is empty. I’ll sort that with Guilhem next week. That is strange that the system flagged a post while the Watched Word file is empty…

I’ll keep you updated, have a nice day!

Lol. Sending the post above ended with this:

In the absence of actual moderation, the work done by volunteers here is hurt by the site’s stupid settings that disallow e.g. me to mark posts like https://ask.libreoffice.org/t/prompt-funds-loan-app-customer-care-helpline-number-91-7047363916-7019785677-now-ts/112235 as spam infinitely:

1 Like

The list of “watched words” isn’t empty (there might be a bug somewhere where it’s shown as empty instead of denying access). Various words trigger various actions; I’m reluctant to publish the full list to avoid helping what appears to be fairly dedicated spammers, but I can tell that it includes “loan app" and variants of such.

Given only TL4 users are allowed to flag posts, I bumped max_flags_per_day from the default 20 to 100. Thanks for helping weeding out spam!

1 Like

Though I may have used “l… app” in another topic (because I am developing such a program for my club), the term does not appear in the post mentioned in the question. So there is another cause.

In your case that seems to be the page range. Posts containing what appear to be a phone number are brought to the attention of an admin for manual review/moderation. It would be nice if that behavior could to be turned off for high-TL users, but that doesn’t seem to be possible unfortunately.

Effectively “eight hundred-one thousand” looks like a North-American toll-free number.

Yes; not here.
But what is the Lounge category for, then? Or there must be another category for level 4 (leader) users?

I have not flagged anyone today. So, why these?


Both were in the Português do Brasil category. Today, at 6 o’clock my post of yesterday was approved.

Not by exceeding max_flags_per_day, which would trigger “You’ve performed this action too many times”.

I looks like it was triggered by Discourse’s own generated UUID which replaces the filename of the image. That UUID has groups of digits which are interpreted as a phone number. Sigh. Fortunately having only digits there is unlikely, but I’ll see if I can patch the code. Or better let posts from TL4 users through.

1 Like

Yes, it could be. Both times it were with a screenshot (using a notebook with Windows 11, not my PC). I will be looking for these numbers sequences next time.
Thanks.