Thank you for the feedback.

Anthony, what do you think? Can we apply this rule only to (P0, P1 & P2) bugs? 

The intent here is to ensure that we are are regularly validating the backlog to ensure that the work (should someone pick it up) is still useful... even if we don't intend to do it in the near future.  The duration, 1 year of complete inactivity on the issue, tends to be a fairly strong signal that validation maybe required.

We want to avoid the scenario, which we've seen, where new engineers start picking up (and working on) old issues that no longer have an utility.

Priority is really an aside, since an Available bug w/ no milestone is necessarily the equivalent of a P3 (i.e. even if it was label a P1, P2, P3 the author is saying do any time).

> We want to avoid the scenario, which we've seen, where new engineers start picking up (and working on) old issues that no longer have an utility.
I think usually new engineers consult the reporter or the owner of the component, on the issue tracker or offline, before actually starting the work.
Even with this auto-archiving system, they still have to consult beforehand as the bug can become outdated in a quarter or two.

Also, if they want to avoid working on old issues, they could sort the issue search result by the last modification date (btw, now this doesn't work because sheriffbot@'s comment counts as the last modification).

> Priority is really an aside
It's more about the type of the task.
It's a shame that there is no issue type for issues where the software is working correctly, but the implementation can be done better.
So people are using an unofficial tag "CleanUp" with type=Bug (e.g.  issue 446115 ), or an unofficial type "CleanUp" (e.g. issue 545474) for this type of issues.

IMHO cleanup tasks are less noisy than bugs.
Because bugs directly affect users, anyone from end-users to web developers file tons of issues against them.
OTOH, cleanup tasks usually affect only Chromium developers, so they are filed almost always by one of them.

I understand there should be some benefit in garbage collecting old bugs filed by external people, but I don't think the same applies to cleanup tasks filed by a Chromium developer.