nw.wikipedia wrote:

I just wanted to know; would this mean that articles created by reviewers would just bypass the new pages log? Would there be any way to split this in the future, so that if Flagged Revisions are enabled, people with reviewer don't just automatically bypass the log?

Ruslik00 wrote:

They will not bypass the New pages log. The new pages created by the members of the 'autoreviewer' group will be marked as patrolled (or reviewed) in the log much like the pages created by sysops are now.

nw.wikipedia wrote:

Well yes, that is what I wished to happen when I sparked the discussion. I was actually talking about the proposed 'reviewer' usergroup that would come with Flagged Revisions; I did not think that that group would have a version of 'autopatrolled' that would affect the New Pages log; I thought it would only affect the Special:UnpatrolledPages log.

Ruslik00 wrote:

I forgot to say that users should be added and removed from the new usergroup by sysops.

I can't see consensus for that name, only that there should be a group created.

jake wrote:

It looks like there is consensus for the "Autopatroller" name, but none for the other suggestion. I've changed it from "autoreviewer" to that. That said, this is a really urgent request; newpage patrol is struggling right now on enwiki. It would suck to delay things over something as trivial as the name of the group.

Strongly suggest not implementing this as requested.

With this request you have:

• Yet another user group (leading to more clutter and noise in Special:Log/rights and Special:UserRights);
• More difficulty when a future 'reviewer' group is implemented for FlaggedRevisions and the old group has to be removed;

This would be much better implemented as an implicit group (X edits and Y days, like autoconfirmed). The current practice of one-right user groups is silly and shouldn't be spread further.

(In reply to comment #7)

Strongly suggest not implementing this as requested.

Obviously the community has asked for it, so I don't think we're going to take your advice on not implementing it.

With this request you have:

• Yet another user group (leading to more clutter and noise in

Special:Log/rights and Special:UserRights);

Non-issue.

• More difficulty when a future 'reviewer' group is implemented for

FlaggedRevisions and the old group has to be removed;

If you read the request before commenting on it, you'd see that the suggestion was to create the group that will be used for flagged revisions in advance, and then to assign the appropriate permissions to it when Flagged Revisions is enabled.

(In reply to comment #8)

(In reply to comment #7)

• More difficulty when a future 'reviewer' group is implemented for

FlaggedRevisions and the old group has to be removed;

If you read the request before commenting on it, you'd see that the suggestion
was to create the group that will be used for flagged revisions in advance, and
then to assign the appropriate permissions to it when Flagged Revisions is
enabled.

Well, as long as it's somehow related to FlaggedRevisions, there's a very low likelihood of it happening anytime soon, so I guess I needn't be worried.

You and I differ on the role of one-right user groups (or even "single issue" user groups). But with groups like http://en.wikipedia.org/w/index.php?title=Special:ListUsers&group=uploader it seems pretty clear that the current situation isn't optimal.

You think it's a "non-issue" to go around cluttering rights logs (and watchlists and the UserRights form) with more groups (you implemented the AbuseFilter group, after all), but I think most others would disagree. Oh well.

I'll point out that this group would not be equivalent to an autoreviewer group, please see my comments at VPP and the configuration: [[Wikipedia:Flagged protection and patrolled revisions/Implementation#Implemented in configuration]] (10th point), the 'patrol' (second-level) flag marks a new page patrolled in the npp sense, but it's not reciprocal, so to reply to a previous question: yes, reviewers would be automatically patrolled in the npp sense, like admins. But users with npp-autopatrolled permission would not be patrolled/reviewed in the 'patrolled revisions' sense. The reviewer usergroup would have much more clearance, so the autopatroller group should not be used as a previous group for it, rather try to speed up bug 18334 , for the passive implementation part at least (see latest comment there).

As for one-right groups, I must say I wonder why we still have the uploader group around, beyond cluttering and never being used, it does nothing but confuse new users and makes the documentation page and talk page vandalism targets.

Ruslik00 wrote:

In response to the previous comment:

We are not talking here about 'reviewer' group of FR. We are discussing 'autoreview' permission and group. As I remember many even wanted to give this permission to all autoconfirmed users. So, I do not think that 'autoreview' permission requires "much higher clearance" than 'autopatrol'. In fact, the clearance is exactly the same: if you trust a user to create new pages, you can trust them to make ordinary edits to articles. So, if FR are enabled, it will be reasonable to assign 'autoreview' permission to the same group as 'autopatrol'.

In response to Jake Wartenberg:

I do not care much about the exact name of the group (like majority of the participant of the discussion on VPR). It can be 'autopatroller' after all. I simply thought that 'autoreviewer' was a better name.

nw.wikipedia wrote:

(In reply to part of the previous comment)

if you trust a user to create new pages, you can
trust them to make ordinary edits to articles.

As a matter of fact, I really do not. I trust that if an autoconfirmed user vandalizes a page, their edits will eventually be noticed, possibly within minutes or hours, or possibly even within months. However, I do not trust users who just barely meet the autoconfirmed mark to know what topics are notable for inclusion, and if we were to give out autopatrolled liberally enough, I am concerned that inadvertant BLP vandalism or simply non-notable topics would get through at a much more rapid rate than we wish. So in fact, I believe teh opposite should be thet case; the standards for getting autopatrolled should be higher than for getting reviewer.

Done.

Continued discussion in relation with FlaggedRevs at [[Wikipedia_talk:Reviewers#Autoreviewers]].