@PaulWhite An ASN is a number that represents a routing group on the internet. It effectively indicates that IP addresses with that ASN have a logical association. ASNs are usually allocated in blocks on a regional level, and then to individual ISPs. Effectively, it, more-or-less, identifies an ISP. ASN numbers are watched due to a significant number of TP posts originating from IP addresses associated with the ASN.
tbf, that website doesn't work if you don't give it your actual first name (this is the Internet, you never know which Rand al'Thor is real and which one is not :P)
Noun: light at the end of the tunnel (uncountable)
(idiomatic) A better situation after long hardship.
1999, Abraham J Twerski, It's Not As Tough As You Think: How to Smooth Out Life's Bumps
If there is no light at the end of the tunnel, chronic pain can render a person dysfunctional. In desperation, people often turn to potent pain-killing medication for relief.
2008, CIO magazine (volume 21, number 7, 15 January 2008)
But by fall 2005, there was light at the end of the tunnel. The team could see the new business processes and financial data governance mechanisms actually being used by Nationwide employees.
Why am i again banned from chat? Even if for just a week.
I'm one of the good guys, from outskirts helping this operation go while trying to foster cheer across forgotten barriers.
I curtailed my language long ago. Please reinstate me and delete this post.
but more than sure, he's been politely told, impatiently told, and seriously warned for, say I can count, 7+ times, for dumping streams of consciousness here, once every few weeks
the argument that "userX is harmless" isn't quite true, IMO, as his incomprehensible thoughts are in fact disruptive, to some extents. not that harmless - only harmless to an extent that it doesn't emit flame, but still being actively counterproductive
@Randal'Thor they left on their own shortly after being warned in the rest of the occasions, making kickmuting not really necessary
@iBug There's another bad argument. Most of what everyone says in chat would be flagged and closed on any main site as "unclear what you're asking" or "not an answer".
@iBug Have fun clearing up tens of thousands of messages of idle chatter then.
@iBug Tbf, there was a recently conversation (rough starting point) that technically had nothing to do with spam-fighting, moderation, or any of the projects (directly anyways). Does that count as disruptive?
guess I'm not conscious enough now to figure out some convincing arguments
should like, chit-chatting isn't directly or inherently disruptive, but it could be when no one else understands, or something so-so
chat is just chat, it's as casual as it inherently is, but when many people don't think you're being constructive - then it's bad - the majority decides
as said, in the first few cases, humn's been politely asked "we do X, do you have something related to X to talk about? if not then this isn't the right place"
I don't know what happened exactly, but IMO the appropriate solution would be to just keep them out of this room, if that wasn't the exact thing happened
you know, I only logged in after seeing that big meta post, and didn't realize it's specific to CHQ before reading the comments
@iBug Yeah, that would be a good solution. Unfortunately there's no way to ban someone from a specific room ... apart from, as mentioned, kick-mutes, which are very temporary.
just said, I'm sorta unclear now, so if you want some convincing arguments, maybe ask Art or tripleee or Mithrandir - whoever with more experience in moderation and administration could make better arguments than I can do now
a bad, yet valid argument could be as short as "we define what's welcome here", and a longer version could possibly be a definition of "what's welcome here" - surely you don't want that gory thing, cuz chat is meant to be casual
@iBug If the post is deleted while red flags are still pending on it (which is one way of manually validating them), then IIRC the contents will be hidden behind a link to the revision history even for 10k+ users.
What are the differences when a post is
deleted by Community ♦ for having 6 spam flags from regular users or one spam flag from a ♦ moderator (and consequently, locked by the Community user)
deleted by methods other than red-flag-nuking (by ♦, 20k+ users or by Community ♦ through user deletion/...
@bertieb There's currently no way, but there's certainly a desire. Doing so would result in considerably fewer FP. It'd be good to have, even as a static list in findspam.py.
@bertieb Currently, there are only a few that need to be on the list. I really don't see a need to have it be something that is added/removed using SD commands. We can just make changes by submitting GitHub PRs. If it becomes something that needs, is desired to be dynamic, then yes, we can add SD commands. But, just a static list would solve 95–99% of the problem.
I agree that a full implementation of whitelisting would be beneficial, but then we're talking about affecting lots of different detection reasons. There are also times when we want different whitelists for different detections, and to not share the list, or at least not share some entries between detections. A full implementation gets complex.
Yes. While there are possibilities that could be beneficial for a more generalized implementation, what we're looking at, at the moment, is something that's really limited and easy to implement, and that gets us a good amount of benefit for a small amount of work. Let's not expand it into a large project, at least not right now.
@bertieb And even that implementation is bad, as it just searches for those in a regex that is delimited by \b, which means that it will match a considerable number of spoof sites.
Even worse is that it's doing that detection on the entire URL, so the detection includes the URL path, which makes it even more prone to matching spoofs, or even something that's not really putting any effort into trying to spoof the domain, but just happens to have one of those pieces of text in the URL path.
otherwise, "generally think system-level block, self-promotion per SE policy, vandalism tp-, garbage k, foreign language n unless spam in which case k, plagiarism k but don't report it, repairable offensive posts f, and n is for things you'd flag NAA"
@Randal'Thor Could be important if we were trying to train machine learning on it. But since we're not, it's more useful to do the easiest/most sensible thing and make life easier rather than requiring people to look at things in even more detail to be able to classify them