For consideration in the context of the discussion had the other day with @methecooldude:
Combined with https://github.com/cluebotng/reviewer/blob/main/cbng_reviewer/management/commands/add_reviews_from_huggle.py (would have been skipped by the bot, if in the whitelist at the time) 1 action can result in a completed classification, which might be un-clear to report admins x-ref https://en.wikipedia.org/wiki/User_talk:ClueBot_Commons#ClueBot_NG_Needs_You!
This was intended to reduce (half) the work for maintainers/people who are both report admin and reviewer, that list is now larger due to auto-rights.
Additional consideration: Should report get merged with review with a different status, based on specific group (we have this already today) etc, that would reduce the maintenance overhead.
Currently report is tied to the same database as the bot uses & loading report can affect the bot, review is entirely stand-alone... only some extra meta would need to be added (e.g. bug/invalid reports) and a user interface (simple, with feedback of status).
On the flip side, what we have has worked for 12-15 years so why change it (prior to blowing up and being forced to, see review).
On the AI will take over the world side, the likely hood of re-training the bot is low... so the value of reports/reviewers is unknown or low at this point.
For consideration in the context of the discussion had the other day with @methecooldude:
Combined with https://github.com/cluebotng/reviewer/blob/main/cbng_reviewer/management/commands/add_reviews_from_huggle.py (would have been skipped by the bot, if in the whitelist at the time) 1 action can result in a completed classification, which might be un-clear to report admins x-ref https://en.wikipedia.org/wiki/User_talk:ClueBot_Commons#ClueBot_NG_Needs_You!
This was intended to reduce (half) the work for maintainers/people who are both report admin and reviewer, that list is now larger due to auto-rights.
Additional consideration: Should report get merged with review with a different status, based on specific group (we have this already today) etc, that would reduce the maintenance overhead.
Currently report is tied to the same database as the bot uses & loading report can affect the bot, review is entirely stand-alone... only some extra meta would need to be added (e.g. bug/invalid reports) and a user interface (simple, with feedback of status).
On the flip side, what we have has worked for 12-15 years so why change it (prior to blowing up and being forced to, see review).
On the AI will take over the world side, the likely hood of re-training the bot is low... so the value of reports/reviewers is unknown or low at this point.