Quote:
Hmm...
I can appreciate the logic of what JP is trying to do here. I even seem to remember a thread where the suggestion was made to have a thread point aggregate be listed alongside the thread link, so that we could see if the thread was about to waste out time and bandwidth before we clicked on it.
If you continue thinking along these lines then an auto thread close feature is the next logical development. In the case of some recent threads, such as those by a certain Dark[Zeus], that got flamed to bits, this would enable the system to automatically close the offending thread.
I can also appreciate why such a development would require additional fine tuning of the antipoint system to avoid the accidental discontinuation of threads. I can further appreciate why this system has to be a black box, or 'magic' if you prefer, to avoid misuse of the Anti Points system.
Lastly, I appreciate, as a developer myself, how annoying it can be to have someone, especially someone who doesn't understand how the underlying decision model works, second guess what you're doing. So let me be clear that I'm _not_ second guessing here, but I do have some observations.
I'm fairly careful about giving out anti-points. I only assign them under two conditions. The first condition is that I finish reading the post and think 'Wow - that was a really good/useful/worthwhile post. You desvere some greens'. The second condition is when I finish reading a post and think 'hanging is too good for this offensive moron. Eat my reds'. And without any sort of balancing at all my ratio of assigned red to green has stayed at around the 80% green 20% red mark.
In truth I think that this is a credit to the members of Anti Online that, on good day, 4 out of every 5 articles are worthwhile and only 1 in 5 are worthless.
So observation number one. Is the 70% positive threshold truly representative of the good post / bad post ratio on this site? Are there enough /really/ poor posts (as opposed to just not very useful posts) to go round? 70% negative seems fair, but maybe it should be slightly higher for positive.
Observation number two. I have been looking at other peoples assignment ratios since I got the 'Miyagi' message. Thre are a lot of people with a whole ruck load of negative anti-points they have to give out to get even. This is where I feel the biggest stumbling block is. I myself had to try dumping negatives onto already banned users to try and balance myself up and /I/ wasn't /that/ out of balance to begin with :(. Some of the seniors are going to have real problems.
I have even noticed that someone dumped me with a red one (a drive-by), on a post that was otherwise extremely positive this morning. As this was my /first/ red one (I'm _not_ whining about this btw) the timing seems a bit of a conincidence. Surely this dumping of red points will skew the sytem even more.
This change is designed to have an effect on the way users assign anti-points from here on in, but it takes into account those anti-points assigned prior to the implimentation of this system. Maybe the implimentation of this system requires that the amount of points assigned by users is returned to zero so everyone starts from a 50% positive 50% negative position again.
Observation number three. Strategic re-evaluation. Is the goal of auto-closing threads sound? Have we fully explored the possiblility of having more human moderators? Community is a two way thing (which I know you know ;)) - I'm sure people would fall over themselves to get involved. I would. Just a thought.
As I said before I'm not second guessing. I have faith that 'things will turn out form the best' - they usually do IME. I am fully aware that JP will have already given /considerably/ more thought than I to these any of us and I have total faith that AO will eventually be a better site for the changes.
My only real concern is that the short term effects of this could be negative, especially on newer users who might get a lot more reds in this dumping process, unless the appropriate actions are taken to negate these effects during the implimentation phase.
