Moderation statistics :)

No you wouldn't - in the cases where the default "keep tsr friendly" has been removed, you put in the filler instead into the explanation. Think of it as a global, nameless report history done monthly into a lovely table with pretty graphs and the like.

So clearly you're problem is with the notion of showing the cards overturned. And yes, it IS unquestionably a sign of poor moderation if contested cards is high for ambiguous topics like offensiveness (something TSR has a problematic focus on) but with a low reversal rate. Moderation is subjective, therefore this should be reflected. If the stats don't show this, then clearly there's something amiss with attitudes towards moderation and being humble.

What filler? Either you put in an explanation of why it is actually offensive, which requires a lot of legwork, or you leave the default comments, which have no benefit.

Imagine you get a report, and you see a load of cards for "offensiveness". If those offensiveness cards just have the default warning comment, then how do you know if we're being too harsh in judging what is offensive, or whether people are being genuinely offensive?

So clearly you're problem is with the notion of showing the cards overturned. And yes, it IS unquestionably a sign of poor moderation if contested cards is high for ambiguous topics like offensiveness (something TSR has a problematic focus on) but with a low reversal rate. Moderation is subjective, therefore this should be reflected. If the stats don't show this, then clearly there's something amiss with attitudes towards moderation and being humble.

And yes, it IS unquestionably a sign of poor moderation

What filler? Either you put in an explanation of why it is actually offensive, which requires a lot of legwork, or you leave the default comments, which have no benefit.

Imagine you get a report, and you see a load of cards for "offensiveness". If those offensiveness cards just have the default warning comment, then how do you know if we're being too harsh in judging what is offensive, or whether people are being genuinely offensive?

Er that's not how the current system works ()

I'm talking about what people see on their report page, the thing that a moderator fills out when they actually give the warning:

Example 1 - generic explanation done by the system:

Offensiveness/trolling
Please don't make offensive posts. The Student Room is intended to be a fun and helpful place for everyone - let's keep it friendly.

Example 2 - this is an example of what is literally on my report page, albeit amended:

Offensiveness/trolling (<- reason)
Please do not fist fight with other users or bring their mothers into this (<- explanation)

All of that currently exists. I'm just saying collaborate it for a community report.

No, it's not just an issue with that. But that is certainly an example of a flaw with your notion.

It's not unquestionable though, not without context. You don't know why the card was contested, nor what the post was - and I think it's fairly apparent that people like to test their luck if they think they have even the remotest chance of getting it reversed, regardless of whether it actually should be.

Modding is subjective, yes. But that doesn't mean that contested and overturned figures should be equal, or close - that wouldn't be a sign of being humble, that would be a sign of not actually doing our jobs, and letting offensive posts off the hook purely because they protested.

Why would it mean that? If the number of contested cards is high, and the reversal rate is low, it could also mean that a lot of people contest their cards and yet their posts did happen to cross the line, which leads to the cards being upheld.

In other words, once more, those conclusions cannot be drawn from stats. Context would be required.

I'm pretty sure you're the only one who thinks that.

Talking as someone who has to deal with the current system on a daily basis - Yes it is actually.



The former is typical, custom comments like the latter tend to occur when a full explanation is needed for various reasons (multiple rules broken: e.g. do not avoid the swear filter, and refrain from making personal attacks; previous warnings about the same topic going unheeded, etc.) - the former is of no use in these reports without having a proper explanation added.



Yes, it currently exists, but a lot of it only exists in the moderators heads, with what's actually in the database being the generic "don't be offensive" comment: which as I've explained, isn't of any use for drawing meaningful insights into moderation standards.

With the moderation on TSR being focused on offense, which is subjective, there is going to be a load of grey area. Something needs to be done to highlight and address this, and showing the hilariously near-perfect conviction rate of the moderation team would surely get the wheels turning.

Everyone being geoking? The only benefit given so far seems to be transparency, or variations on that. The point isn't whether it would take long first and foremost, it's whether it would actually be worth doing - and no-one, as far as I can see, has addressed the issue that the stats would be completely meaningless unless you could see AAM and/or everyone's community record.

The changeover to cards doesn't mean there was a problem before, it just means this new system offered some benefit or improvement that the old one didn't.

You want to give feedback/input? Great, go ahead, no-one is saying you shouldn't. Hey, there's even a section for it - the one this thread is in, in fact.

If you wanted useful stats, yes it would be - cards issued per month is meaningless without context. I remember a night, not long after I started, where there was an invasion from 4chan, resulting in a good couple of dozen new members being banned, but not before some had managed to get a few bits of illegal material posted on here. If you look at that months cards, you see 24 bans issued on one night: without the context given above, what does that actually tell you?

You'd need the reasons for each card - and I'm not talking about the generic "please don't make offensive posts" card comments, because that could cover anything from inciting hatred/calling for violence against black people to just calling another member a bit of a pillock. To do that, you either need to make card comments specific to each warning, or have the mods write up a reason for each card for the reports.
Likewise overturning cards - there's no reason for it being reversed saved - it could be anything from a card being unjust to a new mod learning how to use the warning system on a set of dupe accounts.

Without context, the stats mean nothing, and the context isn't saved in the database, that requires the mods, SLs and CT to sit down and produce it - which requires a lot of time.

I thought it was near-perfectly phrased, and thoroughly apposite.

I shall repeat myself. They treated OP's arguments with exactly the degree of respect they deserve.

I don't think I can express myself any more clearly without being carded for offensiveness.

Why isn't the standard rule not good enough? If it's good enough to slam on someone's report history why is it then not good enough for other people??

And you said if more explanation is needed, then it is given...

Either you've found a problem with the existing moderation or are making holes where there are none.

Point in case. You mean they dismissed his points, ergo they didn't give a counter argument What a surprise.
It's also hilarious that all those defending mods who are not mods are subscribers....

Then add that detail in, or you could simply not assume people will start some witch hunt and actually possess common sense.

Well I think that sums up how the system is broken if you lump racists and casual insults (which could be true) together.

As for writing up - make a database It should be automatic and require no input after setting it up.

Then your database needs improving, and no, it wouldn't take much time at all.

Why not give it a shot? I have never seen so many mods jump to defend something so avidly. It just looks incredibly suspicious and that you are desperate to keep something hidden.

So how many bans did you have with warnings compared to cards? What benefits has there been? Again, it just looks bad.

Well I have given enough feedback over the years to mods and it always gets ignored, plus I'm not ignorant enough to believe that you would take on feedback if you won't post stats.

That doesn't actually address any of the points.

And do you actually know what the current rate of overturn is, or is this all speculative?

The stats would be pointless to mods, and even more pointless to users. It's a good idea in theory, but as mentioned above, stats would be skewed depending on section they mod, how long they have been a mod, and influx of trolls on certain dates.

Since the mods don't even have access to this, I highly doubt users will for a long time, if ever.

What? The standard warning comment is one of the generic ones in example one, the custom one in example 2 is rarely used. Meaning more explanation is required - which comes from us, it's not saved anywhere: if I warn someone for being offensive, using the standard comment "don't make offensive posts", I don't record what was offensive about it anywhere - that is purely in my head, to add it I'd need to write it down. Which brings us back to the issue of it not being a simple thing to put together.

So stats you are unaware of would be useless. Well if you can't see stats how else can you see if there is a problem or not? How is showing trouble hotspots in the forum a problem?! Surely it shows something isn't quite right with that part of the forum?!

Yes it does. You are human. You are not perfect, running a system based on subjective rules, yet are right how often? Don't you think the error margin should be higher or have you been all given months of 9-5 training?

So stats you are unaware of would be useless. Well if you can't see stats how else can you see if there is a problem or not? How is showing trouble hotspots in the forum a problem?! Surely it shows something isn't quite right with that part of the forum?!

It would most likely show the most active parts of the forum. We know this information already.

Again, do you know the current figure, or are you just speculating?

You seem to be equating you not having the stats to them not existing - clearly the mod team are held accountable to the SLs and CT, who actually will have access to the relevant data. The question here is how would the general TSR users having access to them have any benefits, not whether statistics in general can highlight issues.