Photo credit: Chris Noble at The Wildwood Trust
More Close Votes — and Double-Badger Superpowers
You make good points. I worry that unless this is mitigated according to track record, it risks people just going crazy with it and tossing bad CVs all over the place. Some sort of dampening effect that fed back into the determination might be enough to address that, but maybe not.
Imagine if everyone had 100 close votes the way we all get 100 flags (well, if we're on Santa’s nice list). Would everybody having 100 close votes help improve the site, or would it just make for more mess needing clean up?
If the answer is that it would just make more work, then what about only some users having more close votes and having that based on some metric? Simply basing it on review queues completely might not be good enough. People can just game that too easily. Here are two ways to mitigate that:
- This bonus awarding only happens if hold the gold Steward badge in the close vote review queue. Or give a little for silver and more for gold. Or make it require n times Steward there, like 2 or 3 or 4 or 5 or 10 or over 9000.
- Tie it in with actual success metrics, not just simple counts. Did the close votes actually do any good, or did they later evaporate or had the question reöpened? You would need to have a good track record.
Close and reöpen votes are in a way the same thing, even though they have their own respective review queues. Should domain expertise in one count for expertise in the other? Should the extra votes only count if you have say a silver tag badge in one of that question’s original (not edited in or out) tags?
I avoid running the Close Vote Review Queue because of how quickly it exhausts my close votes, and I feel like I need to hold those back until the end of the day in case they’re more needed. Come 23:55 UTC, if I remember and am around, I finish off the rest of them. But plenty of days what I’ve put off never happens because I’m doing something else.
That means the Close Vote Review Queue is counter-productive to its own purpose. The SE model works best when privileges are exercised, not when they are husbanded for a rainy day.
What’s more, the moment you max out, you cannot even flag for closure. This just motivates people into creating low-rep socks whose whole purpose in life is to close-flag because their owner has run out of proper close votes. This is the wrong sort of motivation.
I like your idea how it is not one-for-one, that when you hit the 10 CVs through the review mark you get those +10 CVs back. I’m not completely sure stopping there to help avoid burning out expert close-voters is necessary or sufficient: it is too little, too much, or is Goldilocks still sleeping in my bed?
Another way of going at it might be a point-of-dimishing returns model. That way when you your Review Queue Close Vote count hits an even multiple of 10, you get back not 10 but 5, but it is not just at the first 10 only.
Other options exist, and these are not mutually exclusive with your own suggestion or my modified version of yours. Here are just a few that come quickly to the top of my head:
I run out of close votes all the time. All. The Time. Perhaps it’s time to recognize that there should be something between the 50 CVs you get on SO and the 24 CVs you get on other sites by increase the higher-traffic non-SO sites from 24 to 36 CVs.
Delete votes scale but close votes do not; how in the the world does this make any sense? Fix this bug by awarding +1 CV for each 3k over the 3k minimum needed until a max of 100 CVs is reached on non-SO sites, or 200 CVs on SO.
Grant “very” high rep users with a good or excellent track record of being right about closure "higher weighted" CVs, so:
- Grant double-strength (counts as 2x) CVs for those with good rep+record. This way only 3 such power-votes are needed to close an open question or reöpen one.
- Grant triple-strength (counts as 3x) CVs for those with excellent rep+record metric. This way it takes 3 votes from the enhanced power-closers and just 2 votes from the topmost ones.
This is of course for both close and reöpen. It is not Mjölnir, but it helps. By tying it to other criteria than merely reputation alone, it ties the privilege to actual performance not merely to popularity gained by answering joke questions with joke answers the way almost all other privileges are granted. You have to have proven yourself good at CVing for your votes to be back a double- or triple-whammy.
Unlike option #0 from the OP or even in my modified form tying it to success rate at CVs, option #3 addresses this concern:
Very, very few people hit the current limits on a regular basis; not sure this is much of a privilege. Also, unlimited non-binding votes potentially just creates more work for other users. I'm pretty sympathetic to the desire to make close votes more effective for trusted users, but more weak opinions isn't likely to accomplish much. — Shog9♦ 2015-04-04 16:05:07Z
One possible way to address this would be to make the double-strength CVs a new 30k privilege and the triple-strengh CVs a new 40k, 50k, or 60k privilege.
The problem with that is that it is wholly reputation based instead of based on some combination metric that factors in not just reputation but track record as well.
I thought about tying option #3 to a new silver and gold CV-related badge respectively, but that might be problematic if once the privilege were somehow earned, subsequent “bad” behavior (unrequited close or reöpen votes) eventually made that user a poor candidate for having double- or triple-punch CVs.
So here’s a better idea.
You must hold a gold Steward badge in the CV queue. Then you get more effective CVs for those questions whose original tags you hold a tag badge for as follows:
If after CV Steward, you also hold a silver tag badge for the question’s original tags, then your CV counts for double-strength.
If after CV Steward, you also hold a gold tag badge for the question’s original tags, then your non-dupe CVs now count as triple-strength there on those questions.
An actual dupe CV of course remains plenipotentiary for gold tag badge holders under the Mjölnir exemption and would not be subject to Steward. This is just to up the weight of the non-dupe close votes for badge holders.
This two tiered double-badger approach is attractive because it actually adds new badge privileges, which the SE Team has said they would like to find more ways of doing. It is difficult to game because it requires proven domain expertise.