Findings, conclusions and recommendations by Piotrus (as of 26 Nov)
Interesting, but (for me) a bit limiting. It seems to jump from a rather obvious statement about joiners,leavers, and +/- reinforcement, to conclude we have now understood what needs doing. The points are fair, but they feel very limited in style. Saying "improve support, survey leavers, nuance and improve sanctions" somehow seems limited. That's not to minimize the points, but it doesn't feel "hands on" enough. What specific changes would have great impact?
Bear in mind the Foundation/Task Force will state "these things need doing". If they are general, like "Improve civility" or "Improve sanctions", then it's an instant fail, because we won't have enough ability to say how, and a week later it'll be yesterday's news. So look for specific changes that, if stated, have a very high likelihood of leading to real direct improvements, and changes that will persist and grow.
- "Survey ex users" probably won't. We can probably guess the main issues and reasons quite accurately right now; it's unlikely to be the best use of our recommendations in terms of "knowledge gained" or "changes made".
- "Improve Arbcom sanctions" probably won't - wait 6 months and watch to see if the Arbcom of 2010 is really going to invent a new approach.
- "Tell people to be nicer to newcomers" probably won't - you have the inertia of many newcomers and "tell people to be nice" has a very low leverage.
By contrast:
- Set up trusted users? Yes - because once set up, that spreads out and influences many things positivelky, without further "pushing".
- Give new users a range of wizards to guide them? Yes - because once set up an ongoing and visible improvement in new user knowhow and understanding (including dealing with slightly advanced matters like edit warriors and processes) becomes likely without further "pushing".
I like the ideas you have, but for my money we need to focus on what would make the most significant change to progress quality, that once set up achieves the most, with littlest communal 'pushing'. That's the key.
I am not disputing your recommendations, in fact I have no problem endorsing them. I do however think that the biggest problem facing us is the erosion to due to hostility/incivility/crude sanctions, and our recommendations need to be very clear on the need to decrease negative reinforcement and increase positive ones.
I also fully agree that specifics are better than generalities. Once we agree on generalities, we can discuss various specific solutions. Note that my analysis above has quite a few of those (recreate PAIN, create 3-strike rule for CIV violations just like for reverting, hire people to specialize in positive reinforcements, have arbcom say "nice" things to prevent editors from feeling crushed by the system, etc.). Here's another idea: community-reviews of editors focusing on positive reinforcement (with constructive critique if needed), targeting stressed editors, those who left or got restricted/banned (with recommendations for more surgical sanctions). This would serve as a community effort to retain/reclaim editors - one could nominate those who departed/are close to leaving, they would see they are recognized and needed, and hopefully they'd stay/come back.--Piotrus 22:35, 26 November 2009 (UTC)
I wasn't denigrating yours either; the sentiments are good, I think as specific suggestions they're at the wrong level. Specifically, anything that relies on telling a large number of users "do this instead" ("this" = some ongoing behavior change) is doomed to failure -- it'll be old news in a week, 99% won't hear in the first place, culture won't be changed by it, new users won't hear it. So I'm thinking in terms of stuff where those sorts of things aren't of concern. Again not as criticism, but so you can see my thinking
3 strike for civility -- what's civil and what isn't is a common hot point as it is. Suggesting that communities form a "clean start" view on acceptable interaction and conduct to other users, gain consensus, and then stick by it from there on, would be better... but if we said that, the odds of it happening are low.
I'd swap all the nice things Arbcom could say, for better tools to address the problems sooner in the first place.
I'm trying to think of users who left, where a "nice things to say" panel or board would do much. At that point it's about their real life (busy, stress, etc), their real life interests (discovered second life or got a girl/boyfriend), etc. A lot isn't under our control. What we need to do is make it more satisfying and something they can develop, not just tell them "we (really do) love you". Especially as a "Oops! We heard you were pissed off and disillusioned so we wanna say we love you". That's not going to help more users on the way out.
I agree that some things are hard to change, due to organizational interia and so such.
Making the ArbCom faster - sure. Would more arbitrators help? You have first hand experience with this, IIRC :)
I guess my recommendations is heavily connected to my experiences. Not only personal, but that of observing WPPOLAND for 5 years. Our recruitment ratio = editors leaving ratio, and with few exceptions, most editors leaving left because they were fed up with unpleasant atmosphere (personal attacks, harassment). Would you disagree that this is a reason most editors leave (mind you, I am talking about leaving, not limiting activity due to RL issues)? And would you disagree that that preventing editors from leaving should be our top priority? --Piotrus 03:27, 27 November 2009 (UTC)
Yes. But (again) I see "fixing Arbcom" (as a conclusion and route) as a complete dead end in the context of this discussion. It's a tiny part of quality intended only for where communities fail. Here, we prioritize "how to make success in these areas much more likely -- without Arbcom".
Assume that anything that relies on persuading a group of users to have different outlooks will fail (see elsewhere for why). Also having worked on Arbcom I can tell you it's the wrong place to be focusing. They're under intense strain, and they cut stuff to the bone and even then it's barely keeping up. We need to focus on the community, and ways to make specific changes that will have a large effect even if this report becomes "yesterdays news" and almost nobody reads it. Instead, engage the mass of people who do edit, and those who don't, to resolve it, by guiding and informing those efforts better.
Assume bulk persuading people "don't do this, do that" is a fail. Instead, make it so that many people doing "this not that" becomes almost inevitable due to simple human nature. Much easier :)
- If you set up newcomer guidance they will take benefit from it, continually, even if the report is ignored.
- If you hammer the basics of quality and highlight improvements on every medium, emphasizing you CAN fix it, more people will, continually, even if the report is ignored.
- If you try and tell Arbcom "be nicer" or "do this not that" it'll last approximately until the next 50 emails arrive; they're firefighting already.
Sorry to harp on it, but it's key to success.
It's not that I disagree with you; it's that your solutions seem to focus on getting new editors - which is fine, but I think we need to figure out how to retain them (and recover lost ones), or we will just keep burning through them until eventually there will be nobody (old, young) left to recruit, and all we will have is a bunch of flaming trolls creating poisonous atmosphere :(
Ah, minor correction perhaps. I'm looking at all aspects, but mostly, what is the most we can do for quality, with 2 - 4 recommendations, that will have strong persistent effects. I think I have said clearly that this includes improving the environment for longer standing users, giving them further skills and help. So we agree there. Where we've differed isn't in what's desirable, but that goals like "Make Arbcom be nicer" or "Set up 3 x CIV as a limit" are in suggestion terms, dead ends. Too narrow, too detailed, unlikely to gain the level of culture change needed, unlikely to have the pervasive effect a task like this is looking for. We agree on goals. I just say those specific approaches are too "slim". Hope that clarifies.
This all makes sense but I fear that the detail level is way too far along for the process. I would like to see more big picture, focused thinking that addresses the infrastructure rather than negotiating exact details at this point. It is good to keep thinking about this and typing it in, however could you focus that same energy and brain power on an assessment of what exists, followed by looking at the whole and then component pieces to assess what works in a Quality framework?
Not dissing anyone, just trying to redirect a bit.
At the very least I suggest we don't discard the more general recommendations; perhaps somebody can make them more practical.