Kraznaya
Princeps
Why is there not a poll in this thread asking whether members prefer this system or the old system?
The new infraction system explained
- Thread starter Mathilda
- Start date
Because we're not asking you which system you'd prefer.
Kraznaya
Princeps
So why go through all the motions of an OT survey? What a waste of time for everybody.
I thought this was a forum for the users, not a playground for moderators.
Mathilda
Queen
The OT survey helped us when we were rewriting the rules and redoing the infraction system. It highlighted the areas we needed to focus on and gave us direction.
Now we are asking for feedback on what we've come up with.
There isn't an A/B switch that would force us to go with old or new, we can tweak either.
I appreciate wanting to get serious about spam or the proverbial problem posters...that should have a long term positive impact on the forum.
That being said, there are some obvious problems with the way the math works on the new infraction list, but that's being discussed in the discussion group, and I have confidence that those incentive structures will be redone.
I'm not really sure if the problem of moderator trust, or any of the real day to day problems are going to change....it isn't like the rules have been altered in any substantial way, just that the stakes have been raised. I guess I'll have to wait and see how it goes, but I'm not super enthusiastic.
That being said, there are some obvious problems with the way the math works on the new infraction list, but that's being discussed in the discussion group, and I have confidence that those incentive structures will be redone.
I'm not really sure if the problem of moderator trust, or any of the real day to day problems are going to change....it isn't like the rules have been altered in any substantial way, just that the stakes have been raised. I guess I'll have to wait and see how it goes, but I'm not super enthusiastic.
It’s not like there hasn’t been adequate discussion or options for input/feedback on the rules changes, from either the OT survey or the Rules social group. Heck, this thread is even meant for feedback on the new proposed rules prior to them being implemented.
Would you rather have new rules being imposed on us with no opportunity for discussion?
I'll say again that basing the infraction system on these two principles, particularly in conjuction, make the infraction system fundamentally less effective.
What is this? Points for 'minor support for piracy'? So you are neutering anyone who even dares express a contrary opinion or who supports piracy in some form? And what's this, two points for minor flaming?
So basically, I was aware that some people had trouble handling penalties on both sides of the argument, but these new rules are just.. You are meddling with the free debate, you are imposing poorly worded rules, and even the slightest misstep will now be punishable by the mercy of the moderators. I might not have been the most active poster lately, mostly due to lack of time, but this is certainly the ultimate nail in the coffin. If you have trouble moderating the forums, perhaps you should try to improve your own approach rather than slap new and improved strict guidelines and hope behaviour you personally dislike goes away.
Azash
I'm also iffy on this one. Obviously, there's no reason to be discussing piracy of civ (or other) games, but I should be free to discuss the reasons idiotic laws like the DMCA are failures.
So do I end up with 2 points every time you see my avatar?
Gogf
Indescribable
I strongly agree with Mr. Town. This does nothing to address the greater structural issues with the system, and instead looks like an attempt at "broken window theory," when we all know the problems with OT aren't a general disregard for the rules, but rather a few posters who make things miserable for almost everyone else.
Here is an explanation from Perfection as to why this system creates perverse incentive structures:
Source: http://forums.civfanatics.com/group.php?do=discuss&discussionid=1865&pp=20
Here is an explanation from Perfection as to why this system creates perverse incentive structures:
Perfection said:
Source: http://forums.civfanatics.com/group.php?do=discuss&discussionid=1865&pp=20
Gogf
Indescribable
It doesn't seem like these new rules really address any of the major concerns that were raised. Perhaps it would be helpful if the staff more clearly articulated how the changes are meant to improve upon specific concerns that users raised.
Mathilda said:
(This format is easier to deal with)
Well let's look at what the problems with the proposed system seem to be.
- It punishes non-problem posters more than it punishes problem posters.
- The system can be quite easily gamed so as to allow for a very high number of infractions without any ban.
- It doesn't effectively deal with problem posters.
So the first point. Say you have two posters, A and B. A has 0 points, and has never been infracted. B has 41 points, and is consistently a problem. Both get hit with a 4 point trolling infraction and a 1 point signature/avatar rule violation infraction. Poster A gets banned, and poster B does not. The only difference that has effected the fact that A gets a ban whilst B does not is that A has been a better behaved poster in the past. So you're effectively punishing good behaviour here, which is really counterproductive. Someone who has 0 points is more at risk of a ban than someone who has 11-15, 21-25, 31-35, 41-45, 51-55 points. I have to say, this frankly worries me and makes me far more insecure in what I post. I am at greater risk of a ban under this system than most problem posters will be. It isn't sufficient to say "then don't get infracted", because I doubt it would be too difficult for someone to try ever so hard to avoid infractions and then just slip up once in an entire year, which for someone like me is about 5000 posts. What's more, if I know that people can be so easily banned, I'm not going to report rule breaking posts unless they are pretty damn serious (I don't know, maybe you'd prefer it if I personally stopped reporting posts, but you get my point).
Secondly. I'm really not good with technical mumbo-jumbo, but I get the impression the way the vBulletin ban/infraction system works is by defining a series of usergroups, with people moving into a new usergroup (and incurring a ban) when they reach a certain number of points. So what people have to attempt to avoid is moving usergroups. If they stay in the same usergroup, they suffer no consequences. This problem has been pointed out by many, but I'll just give an example to make it perhaps a little clearer. Poster C has 52 points. They've accumulated these at a rate of one per week, or four every four weeks (as is perhaps more likely given the four point infraction for trolling (which, btw, is an excellent change)). If they continue to gain infractions at that rate, they will never face a ban. They will never suffer any consequences, even though they accumulate 52 points every single year. That is obviously a broken system. Each week a point will expire, but if they make sure they get another point, they will remain in the same usergroup, which will mean they won't get any punishment.
Now, the third point is mostly related to the second. If someone who gets 52 points a year doesn't face any bans, then obviously problem posters have not been dealt with. The system is too easy to game for them.
So, my solutions? I'm glad you asked.
Get rid of the five point ban for starters. That prevents the first problem. A simple solution. Making the expiration on infractions 6 months rather than 12 would also help with this, as it would essentially halve the risk of non-problem posters getting banned even when they're making attempts to be well behaved (note that I do definitely agree with longer expiry; I believe I advocated for such in the OT survey, but 12 months is simply excessive).
Now, the solutions to points two and three are more difficult, and this is where BSmith's idea comes in. The system cannot be gamed (at least in the way specified in this post) if each extra point puts you at a higher risk of being banned. The key here is to constrict the gaps between the levels as they get higher. That way, each new usergroup a poster gets put into makes it harder and harder for them to avoid a ban. That's what BSmith's idea does. I would suggest levels of 12 points, 22, 30, 36, 40, and then every 4 points after that (if the levels are 4 or less points apart, the system should not be able to be gamed, given that 4 points would be a standard trolling infraction). If think 12 points is a good starting point because that's three trolls over a 12 month period (or 6 months, if you change the expiration). That doesn't catch non-problem posters out for one time lapses. Once you get to that 12-22 usergroup, you are better off in the long term returning to a lower usergroup (0-12 points) because there is more room for error there. And if you are in the 22-30 usergroup, there is incentive to move back to the 12-22 usergroup, and so on. You cannot be secure in the 36-40 usergroup, because any trolling infraction will result in a ban. An alternate solution that I suggested was constricting the levels through the use of more custom infractions. If you get 6 points for trolling if you are at a higher level of points, but only 4 if you are at a lower level, then you are less secure in the higher group. Higher points in an infraction for repeat offenders would have a similar effect. However, as this relies upon a consistent inconsistency in moderation, I think the former idea is better.
That would help deal with problem posters, but still wouldn't get rid of them. You can get an indefinite number of bans and never get permabanned. Maybe if instead of having increasing bans for increasing point levels, have increasing bans determined by the number of times you've been banned (is this how "the Road" worked?). So the first ban could be a couple of days. And if you've banned once before, you get a week. Twice, you get a fortnight. Three times, a month. And so on until a permaban. This means that if you do get continually banned, you suffer the consequences. This shouldn't be too difficult to combine with the system proposed. Instead of having specific bans for points levels, have general bans when you reach those point levels, with the length of the ban simply determined by how many bans that person has had previously (IDK if this might have technical limitations, however).
Camikaze - one quick note - warnings will still be used, so while the first ban level may seem low, most new users get a couple of warnings before ever getting infracted (unless of course the post is completely out of line).
Second, I like the thought you put into your post but in the later stages, do you think that the possibility of only one infraction between bans is too harsh, or no?
Second, I like the thought you put into your post but in the later stages, do you think that the possibility of only one infraction between bans is too harsh, or no?
Gogf
Indescribable
It seems like an awful lot of effort is being put into devising an ad hoc system to get rid of certain specific posters. If everyone really dislikes these people, why can't we just vote them off the island (have, say, a 90% ostracism vote)?
I'm hardly a new poster, so I doubt I'd get a warning. TBH I'm more concerned about how this would relate to me (or how it would relate to regulars who are by no means trouble makers, but occasionally (that is, every one or two thousand posts) make a post that is out of line) than how it would relate to new posters. I'm arguing here out of personal interest. But even in that case, I'd think it'd have to be a fair few warnings before an infraction to make it fair in this regard.
Perhaps. I'm not sure. It would of course be situation dependent, and maybe it would be harsh on some people, but if someone has got to that stage, then perhaps they're one of the problem posters that you're trying to get rid of. No need to give them that much leeway. Tying this in with a six month expiry rather than twelve months would make it much fairer in this regard too (the problem posters would probably be hit by it more than non-problem posters who happen to get infracted a lot).
I guess the idea is to come up with a system that continually identifies and deals with problem posters, rather than simply getting rid of the current crop and waiting for the next batch to come along.
GhostWriter16
Deity
I'm not Camikaze obviously, but for people who intentionally game the system? No I don't think its too much. For people who slip up way too many times, but are posting in good faith? Yes.
Another problem is even though the rules are more clear, they will still be vague, and people are going to slip up on vague rules. Frankly, if you only have to deal with an average of 5 infractions per year by most active posters, and that's still too much, then the problem is with the rules, not the posters. I understand (And agree with) the desire to make this place desirable to post in, but borderline posts are going to be made, and people are going to get angry and troll or flame every now and then. Its not a good thing, but I don't think its causing serious harm if its only every now and again. Now, such things can obviously cause great harm if too excessive, and so should be infracted, but my point is, you shouldn't strive for a system with 0 infractions. It won't happen no matter how many people you permaban, unless you bring it down to a small group of elite posters. Heck, even KD admitted to being warned once...
In fact, from what very little I've read on here, its seems like some people who are now mods (Valka D'Ur comes to mind), may have been banned under this system.
So, while our goal should be to reduce the number of infractions, I say we are taking it too far and too seriously. IMO even 6 months is a bit much (Though it may be OK with a higher threshold for bans like BSsmith said.
So first of all, the thing I think you need to recognize is that some infractions aren't really a big deal. These include, in my opinion: double post infractions, accidentally quoting a post that tripped the autocensor, very minor spam, quote-altering, exc. (I put these in order from least subjective to most subjective.) For these types of infractions, I see the current system of one point and 10-20 day expiration as good. If you don't agree, think about it, do you moderators REALLY want to make someone remember for a year that they accidentally double posted in some thread? I know it only gets one point, but with 5 points in 12 months for an infraction, it would be easy for someone to be banned for the "Crime" of accidentally double posting a few times! Now, if you want to infract things like that, fine, but it should be done lightly to avoid annoying us to death. Obviously double posting all the time isn't a great thing so it should be dealt with, but we shouldn't be targeting to "Ban people who double post too often."
For more major infractions, such as trolling or flaming, I believe you to be on the right track, infractions need to be longer. I hate flaming comments that people know they can get away with, especially if it is veiled. But still, should we have to remember it for a year? I think Camikaze's six months is close to the mark for these posts, though I'd personally advocate for four months. That's still a long time.
As for ban lengths, Camikaze is right, but I should say another thing, weakening bans doesn't make them less annoying. Its still annoying to have to not post for three days, and for 5 minor infractions in a year, seems a bit much to me. The lowest ban should be somewhere between 10-15 with the changes I suggested. This still makes it easier to get banned then before, but leaves leeway for those who genuinely made mistakes in good faith. But as you get higher up the "Ban level", the gap should close. If the first, one week ban is 15, the second, 2 week ban should be 25, the third 33, the fourth one 40, exc. (These numbers should not be taken as the gospel, I'm just demonstrating.)
@Camikaze – Yeah, my second idea (dual ban scales) was an attempt to address the problem posters that would skate at a fairly static point level. It doesn’t sound like that will be technically possible though, so some other method is definitely needed. Perhaps increasing the length of each consecutive ban would be the way to go up to the permaban.
I also think aligning point thresholds more closely with likely point values based on the most common infractions (4 points) is a good idea. I was more trying to explain the concept I had for the system of easier to get bans the higher you get in points than actual point thresholds.
For someone with a ton of points already? No. They have already shown that they do not know how to behave appropriately.
I also think aligning point thresholds more closely with likely point values based on the most common infractions (4 points) is a good idea. I was more trying to explain the concept I had for the system of easier to get bans the higher you get in points than actual point thresholds.
For someone with a ton of points already? No. They have already shown that they do not know how to behave appropriately.
GhostWriter16
Deity
I don't recall to it saying this. It seems to imply this, but neither quote altering nor nation-bashing are on that list either, and I can almost guarantee you the latter at least will be infracted.
I do see your point, but still, being banned EVERY infraction for a 6 month period seems a bit much. I'd make the maximum ban window 5 points. So then 2 moderate infractions or 1 major one.
Also, why is PDMA 4 points? Unlike other infractions, PDMA, quite frankly, IS infracted because of moderator's being offended. While I don't blame them, 4 points seems much for most cases of PDMA (I know there are exceptions) that I have seen.
My apologies, it's not in the rules specifically. I was thinking of what Mathilda posted on page one:
Similar threads
