Separate "watchers" from "reviewers"

[bryandamon]

In my experience, there are generally 2 types of people that should "watch" a page.

1. Experts on the subject that should be notified when the page is changed to ensure the change is correct
2. People who are reliant on the content of the page to do their job

WatchAnalytics scrutiny and reviews scores are aimed at the former, giving people a good feeling that the "right" people have reviewed and approved the page, but the later can confuse these numbers. Would a delineation between "watcher" vs. "reviewer" or "defender" improve the system? I could imagine the page watch icon (star) have a drop-down that shows a different icon for reviewer/defender (better name?) and only the this type would count towards scrutiny and reviews.

[jamesmontalvo3]

There's been some discussion on that. It's a not-insignificant amount of work to create a separate method for tracking reviewers. At present the concept of reviewers just piggy-backs on MediaWiki's built in watchers.

I'm also not sure it's something we want. I think the vast majority of people would choose watcher over reviewer simply because it's less responsibility. The current setup sort of forces people to take responsibility.

[bryandamon]

Understood. Maybe it's something I could try to tinker with in the future.

I understand the resistance employees may have to assigning themselves more work and/or responsibility for blame.

On the other hand, using the new tracking system as a way to capture and share company expertise could provide an incentive if their reputation in the wiki meant something.

[jamesmontalvo3]

We've talked this at length in the past, and I'm not certain on the consensus...but again, I'm not sure that splitting watchers and reviewers would have the desired result.

However, expanding upon the concept you gave in #96 may be what we want. If people could be categorized as in the sales or engineering departments (for example), then you could say a page already has 3 reviews from engineering department but zero from sales, and thus push people in sales to review the page. Other engineers who haven't seen the page, but are considered reviewers, could then see the page as deprioritized.

We've also talked about providing a "I'm not going to review this" option, which is sort of like making them temporarily just a watcher (versus a reviewer) but with the friction of having to click what is essentially a "I'm not doing my job well" button.