Forum:RC Rules Proposal

Forums: Index > Rating Committee > RC Rules Proposal

Proposal

I propose these rules for the ratings committee.

I) The Ratings Committee is made up of 9 members at any one time, and 1 sub-member. These nine members have Favor privileges overall all material except that they have written. The sub-member only has privileges over material that another ratings committee member has written.

II) Ratings Committee members must stay active. If they do not activately participate in favoring, reviewing, or improving articles in a two-week period, they will be replaced.

III) A Ratings Committee member can favor any article that they have not written. Doing so requires a short paragraph or more, made under its own topic-header in the discussion. Each favor gets its own topic header, and will follow a predetermined format (such as "Surgo's Favor" or "Favor 1" - to be decided).

IV) When an article has been favored at least once, it receives a Bronze Star. 3 favors gives an article a Silver Star, 6 a Gold Star, and 7 makes the article become a Favored Article. In order to be a true Favored Article, the author must provide a short description to allow featuring on the front page. A picture, while HIGHLY preferred, is not strictly required.

V) Requests for favoring may be made on a special page. Only the best articles should be submitted, and they should be submitted by someone OTHER than the author. When a request is made, each Ratings Committee member must look at the article, and provide some minor indication that they have read it. If it receives at least one favor, all Ratings Committee members are encouraged to read it before it is removed from the list. If it receives no favors after 4 reviews, it is removed from the list.

VI) When a Ratings Committee member steps down or is inactive for a long period of time, the sub-member becomes the Ratings Committee member. The sub-member inherits the old Ratings Committee favors. He may not re-favor an article that his direct predecessor favored, nor may he remove any favors. However, a replacement Ratings Committee member may still favor articles that were favored by a Ratings Committee member at least two positions back.

VII) When the sub-member steps into the new Ratings Committee spot, a new sub-member is chosen. This likely involves applications and a dart board, and is under the purview of Surgo.

Karrius 16:29, September 14, 2009 (UTC)

Just curious, but at what point in this process would the RC confirm the balance point of the work? Should that sort of thing be worked out and generally accepted before it even gets a favoring request? TarkisFlux 17:14, September 14, 2009 (UTC)

It should require a balance level category in order to be up for review. The raters can then tell whether it is balanced or not at that category.

Regarding "V", I think that people should be able to submit their own stuff. If a user consistently submits stuff that is not favored and discarded, then we might need to change that rule. For now though, I see no problem trusting users to know when their content is ready for review. Maybe in the future we might need to require "X favored articles by a user" in order to allow them to submit their own for review. --Andrew Arnott (talk, email) 17:49, September 14, 2009 (UTC)

Two things:

1. I think there should be some sort of refractory period for rejected articles. For example, once being shot down (if such is the case), an article must wait two weeks (or some other period of time) before it can be re-nominated, regardless of how soon the author makes adjustments.
2. Not quite as related, but if something gets to five or six favors with no picture, I think we as a community should probably be able to find one that is suitable by our combined efforts.

-- Jota 18:11, September 14, 2009 (UTC)

I don't see a problem with either of the two. Furthermore, I think it's pretty easy to see the level of power that an article aims at simply by eyeballing it. The difference between the levels is wide enough that it's not hard to see which category different things fall into. --Ghostwheel 19:06, September 14, 2009 (UTC)

→Reverted indentation to one colon

For replacing the Sub-Member, maybe we shouldn't be sticking more work on the administration? I can see three ways to do it, actually:

1. Democracy: People submit nominations for Sub-Member, accept them, and then get up or down votes of the entire userbase, or everyone with some number of articles. Most approvals wins. You could do a conventional election, but I personally favor approval voting, especially since that allows you to keep voting open while adding new candidates without ruining things.
2. Oligarchy: The Rating Committee selects its new Sub-Member somehow, probably through nomination-acceptance-voting.
3. Inheritance: The previous Sub-Member selects their replacement. This is a pretty bad idea for obvious reasons, but I'm throwing it out so that it's out there; maybe it can be made into something good.

Democracy has the advantage of making the Ratings Committee seen as a more responsive body, while Oligarchy allows you to keep the RC good if it starts off good. --IGTN 00:39, September 20, 2009 (UTC)

At least right now, the difference between democracy and oligarchy won't be so pronounced (since the RC of 9 or so will make the majority (or close to it) of the active user base), but it may come into play later. I think it would be easier to be an oligarchy for now, putting our trust in a safe place, and it can be changed later if something arises in such a manner that forces a reconsideration. -- Jota 01:49, September 20, 2009 (UTC)

Not-So-Final

Since everybody asked me to get a move on and finalize something, here's the next version of the ruleset.

I) The Ratings Committee is made up of ten members at any one time. Each member has Favor privileges over all material except that which they personally have written.

II) Ratings Committee members must stay active. If they do not actively participate in favoring, reviewing, or improving articles in a three-week period, they will be replaced.

III) A Ratings Committee member can favor any article that they have not written. Doing so requires a short paragraph or more, made under its own topic-header in the discussion (either the article's talk page, or the RC forum). Each favor should be under a subsection of a section entitled "Article Favors" (if on the Talk page).

IV) When an article has been favored at least once, it receives a Bronze Star. 3 favors gives an article a Silver Star, 6 a Gold Star, and 7 makes the article become a Favored Article. In order to be a true Favored Article, the author must provide a short description to allow featuring on the front page. A picture, while HIGHLY preferred, is not strictly required.

V) Requests for favoring may be made on the Rating Committee forum. Only good articles should be submitted, and you should only submit an article every other day to prevent overloading the RC members. When a request is made, interested Ratings Committee members should look at the article, and provide some minor indication that they have read it. If it receives no favors after 4 reviews, it is removed from the list.

VI) When a Ratings Committee member steps down or is inactive for a long period of time, a new member is chosen via consultation with a dartboard.

I changed it a bit because I thought the whole sub-member thing was a bit clunky (instead we'll just have 10 members and nobody can favor their own stuff). Also it's my goal to eventually have everything on the wiki rated, and I thought "you can't suggest your own articles for favors" to be counter to this goal so I changed that too. If anyone has a problem with what's here let me know and we'll work it out. Surgo 00:23, September 26, 2009 (UTC)

I also want to note that stuff that is not all that exciting probably shouldn't be getting gold stars (though silver is fine), and I want to avoid this thing where bronze is basically shit because only 1 or 2 people cared about it. I think we need to address that before we go live. Surgo 00:32, September 26, 2009 (UTC)

Maybe if something is deemed appropriate for its listed balance level the first reviewer gives it a star (or something else, perhaps), but anything after that must be earned on the basis of merit (and the first reviewer can say such in their review (i.e. "This article is appropriate for its balance level, but is lacking in... [whatever]."). -- Jota 00:43, September 26, 2009 (UTC)

How's this: While all articles should strive for the distinction of becoming Gold Quality, that won't be the case of course. A Bronze article might be "hey, its a good article, worth using and reading", and if all the RC members look at it, it only takes one member to Favor it. The other members can chime in with "yeah, I agree with that" and not favor it, or they can chime in and say, "hey, this is pretty sweet, I think it deserves more than just bronze" and favor it too, and so on down the line, at least that's how I imagine this working at this point. Does that sound about right? --Ganteka Future 02:25, September 26, 2009 (UTC)

That certainly sounds workable. Surgo 02:33, September 26, 2009 (UTC)

Since I guess I'm in, I support GF's interpretation. - TarkisFlux 02:36, September 26, 2009 (UTC)

I don't quite understand GF's plan. In fact, I don't even understand it well enough to summarize what I understand of it. But, since I was typing this up throughout the three above comments, I'm gonna go ahead and post it:

Perhaps, to keep seven mild favors from making a good (but not great) article into a favored article, we can have degrees of favor: "good" (1) and "excellent" (2). It would work on a point system. 3 points would be needed for a bronze star, 7 for silver, 11 for gold, and 15 for favored (of course these numbers can be tweaked). I chose those numbers so that A) no single person can give an article a star, B) no article can be gold without at least one "excellent" rating, and C) any level can be achieved with only 8 of 10 committee members involved. --DanielDraco 03:05, September 26, 2009 (UTC)

GF's plan is to allow the RC to support the favoring decision of a member without adding additional favor to it. So if someone favored it, and it was a bronze, and you thought that was what it deserved, you chime in as such and don't add on to the favor rating. If you thought it should be a silver or gold, you do the same but do add on to the favor rating. It basically adds the "I like this article and would favor it, but I feel it has an appropriate rating" option to your list of answers, instead of just "favor" for good stuff or "not favor" for poor stuff. - TarkisFlux 03:19, September 26, 2009 (UTC)

That's all well and good, but what if the RC disagrees on what the article deserves? Raters A and B say that Article X should be bronze. Raters C and D scoff at the conservatism of the others, and insist that it should be featured. Raters E and F feel that featured is a status only to be imparted upon the best of the best of the best, and Article X, while excellent, only deserves gold. Raters G and H think it's good, but not amazing, so they say silver. Raters I and J think that it's a very creative and complete class, but overshot even the strongest balance point, and doesn't deserve any star at all. So, what distinction does Article X get under GF's system?

Besides that, this brings us back to the same many-level system that we didn't like back at the old wiki. It ranges from 1 (no distinction) to 5 (featured article). Each Rater would have a different view of the difference between bronze and silver, between gold and featured, etc., so even if two Raters see an article in the exact same way, they might view the distinctions differently. --DanielDraco 22:50, September 27, 2009 (UTC)

Raters A and B would either give it a point, or they would not if it was already bronze and move on. They've said their peace on the subject at that point. Raters C and D would give it a point each, and try to convince any raters who did not to give it one. Raters E and F would probably have given it a point, unless it was gold already, and would have to be convinced by raters C and D to give a point if they haven't already. Same for raters G and H. Raters I and J are obviously on the losing end of this vote, and could try to convince others to withdraw their support, but probably won't have much luck. Which basically makes this function like a drawn out vote, where people are allowed to change their vote after it has been cast.

As to your second point, that seems an issue with the system as a whole rather than the 1 or 0 points you have to give it. As I don't see how your previous suggestion of adding an extra point to the available favoring pool of each RC member sidesteps that problem, I really don't know what you're trying to get at with it. - TarkisFlux 23:31, September 27, 2009 (UTC)

Ah, I misunderstood. GF's model isn't a different point system, it's just a different way for the Raters to use it. Instead of giving a point to every article a Rater supports, the Rater only gives points to articles that (s)he wants to advance to the next step, right? Alright, I like that. Much simpler than my idea. --DanielDraco 23:42, September 27, 2009 (UTC)

Number of Points for Article Ratings

The number of points needed for certain stars/ratings seem a bit high as it is. Perhaps turning them down slightly might be a good idea? For example, if even one member of the FA committee thinks a specific article is "excellent", that might warrant a bronze star all by itself to generate more interest in that article. Perhaps something like Bronze: 2, Silver: 5, Gold: 9, FA: 12? Perhaps push bronze up to 3 and silver to 6 if we don't want a single person with an excellent rating to give an article a bronze star, but the general idea is that our requirements might be too high currently... --Ghostwheel 04:25, December 26, 2009 (UTC)

I don't necessarily agree with the ability of a single rater to confer a bronze star, but I do think the other requirements may be a tad high (thus saying Casper's proposal has some merit), given the difficulty we have had getting everyone to rate everything, especially since some of us are more specialized or specific in what we are willing or able to rate. -- Jota 19:00, December 26, 2009 (UTC)

I agree; we need to downteak the number of points required for various ratings. Last I checked, we have a grand total of one gold-rated article. I also agree that the minimum number for a rating should be at least 3. I'd suggest that we try thresholds at 3, 6, 9, and 12. --DanielDraco 23:16, December 26, 2009 (UTC)