Staff and Pardons

This thread needed to be made.

There are currently 23 staff members, out of those 23, 19 usually, or are expected to post on a pardon in the first stage.

Out of those 19, 14 are inactive when it comes to moderator duties, or posting on pardons. This means 73% of staff don't actively post on pardons (support or not).

You need 5 supports for your pardon to proceed to the next stage, but with the inactivity, that aspiration is 35% likely to happen (it's higher if you're well known), and the likelihood of it passing that second stage is 70%, meaning there is a 65% chance that your pardon will fail (the number is most likely higher in actuality).

Let's focus on that. There needs to be measures taken to assure people who pardon are given a fair chance, especially for those with minor offenses.

Either give modscore for pardons, cut the number to 4 for the first stage, increase the waiting times before dismissal, and tell staff to do their job because this is ridiculous.

 

agreed, but most of staff are inactive cause they're on vacations.

 

Lack of pardon participation has been going on for years. I'm aware that some staff are busy with real life, and because of that it makes the process much more problematic. There are still plenty of staff who log in on a day to day basis who don't post on pardons.

 

Where are you getting these statistics? Your opinion?

 

Fraction percentages. Is it not obvious enough?

Example:

14 out of 19 staff members who aren't actively participating = 73%

How did I get 14 staff members? I went through about 30 or so pardons in the archives for the estimation.

How did I get 35% for pardon success rate to the next stage?

You take the 14 out of the 19 staff members who don't actively participate for first round pardons, and you make it a fraction.

5/14 = 35%

I'd go on, but I think you get the point.

 

Well the math is sure right.

 

It's a cyclical thing, there are quite a few people right now who are on some sort of vacation or otherwise unable to use sythe to the fullest and we don't want mods half-reading pardons and making verdicts. These days we're not so hardass about the 7 day rule, if a pardon is receiving supports and it looks like they would receive 5 supports if they had more time, we'll extend the timeline and make that pardon in particular a priority

 

So when does the 7 day period start for pardons that don't match your description?

An example of lack of pardon participation before the inactive spikes:

http://www.sythe.org/dispute-forum-archive/1846435-pardon-scam-1-5-a.html

This isn't a spur of the moment type of ordeal. It's been going on for quite a while now.

Roary even promoted more people for the sake of balancing the system to no avail.

http://www.sythe.org/suggestions/1810551-changes-high-pardon-system.html

Something needs to change.

 

I always always vote on every pardon that enters the upper staff lounge. Typically I read all the pardons and only cast a vote if I support but I could post no support if needed. Unfortunately often I'm pestered on Skype after every no vote.

 

I agree that there's a problem with this, probably the best way to solve it would be having extra support mods.

Btw shit maths there m8. What the fuck did you even do there. Then say your first percentages were right (35% and 70% to pass rounds) the overall fail rate would be 75.5% not 65%

Even if you did the calculations correctly, this percentages are insignificant and nonsense.

 

I think posting does more good than not posting. It wouldn't make much sense to enforce it though.

What is 19-14? 5.

5 supports by the 14 inactive staff members is 65% of participation not given at all or on a consistent basis.

100-65 = 35

Therefore making a success for first stage pardons 35%. I stated the % increases for well known members, and in actuality may often be lower for unfamiliar members.

This was an estimation pulled from 30 pardons in the archives. It would be hard to give a perfect calculation if outside variables are being considered. The calculations based on the figures presented are correct and sound. The significance of those calculations are up for interpretation. I think it greatly reflects pardons nowadays.

More mods aren't necessary, and will probably bloat modscore duties. There are more than enough mods, the problem is the activity.

There are still underlying problems like not reading pardons all the way through, bias, and sheep influence which can't be challenged.

I was being too broad. I meant failure for the first stage, not counting the pardon in the second stage. The way it's worded implies otherwise, my bad.

The calculations were just to exaggerate the reality of the situation.

 

I try to post on every pardon, after reading/researching it. I'll generally post on every pardon prior to the 7 day rule. May miss a few here and there if I'm away for a few days and the time overlaps, etc. However, normally I'll always post my opinion on said pardons daily.

 

Giving a short explanation why you didn't support would probably stop the pestering on Skype.

I guess this is all that's needed.

 

You're mixing up two different concepts, participation rate and failure rate, which aren't related at all.

 

Hi, +1

I've been trying to fix the issue score-wise to incentivise the process. Originally I wanted to give a modscore point for each post on a pardon by staff members. Unfortunately; with how the modscore system works we can only give modscore for a moderator action like locking. We have started to give a modscore for locking until we can come up with something better - but of course this only incentivises locking expired/denied pardons.

One of two ideas I ran past Matthew were;
- Adding a column on the modscore table for posts in the pardon section but not adding any score for them. This way we see who is active and who is not and praise based on this.
- Award modscores for votes on polls in pardons.. even though this isn't a mod action I was hoping it could be coded in if we only gave permission for staff members to participate in polls in that section. But I'm yet to hear back on this proposal and it's ability to be coded in or work.

I'm open to any other incentivising ideas for the section. And, as you can imagine.. having something implemented which is beyond my control is a timely process, so it won't suddenly happen overnight but I am actively chasing these things up without being too annoying at the same time.

 

Or spend the literal 10 minutes it would take to count the posts yourself?

 

They're relative. You need support from staff to proceed, if not your pardon fails. How do you base failure from a pardon if you don't take into consideration the factor needed for a pardon to pass? The calculation from the figures are correct, although it doesn't consider other variables like if mods support or won't support. That's impossible to measure. Like I said before, it's to exaggerate the magnitude of the situation. Probing these calculations does nothing for matter behind it.

Sounds good. What do you think of decreasing 5 supports to 4 and possibly lengthening the 7 day period after a staff member gives their take?

 

It would take more than 10 minutes for sure!

I think 7 days after the first response works well. As for lessening supports.. I'm indifferent. Since we have a second round I don't think lowering really hurts.. but whether it's 5 supports or 4 supports the inactivity problem still remains.

 

I didn't read any of the posts above because tl;dr, but I like how the pardon system is right now. I really don't even believe in pardons in general so might as well make them super hard to go through. If you're not a well known member who has been banned, why should you get the benefit of the doubt of having a bunch of the active staff members even consider you. Just don't break rules and we won't have to worry about a pardon.

 

one thing that people often forget is that no-votes don't need to be posted. so there are a good number of threads that people read, decide they don't support, and accordingly don't post. i'm not a fan of this system, but that's how it is.

in the past, we talked about a "+/-5" system, where the pardon would move to USL if it got five supports and be denied if it got five no-supports regardless of time spent. i liked this idea and forget why we didn't implement it, maybe someone else remembers?