It would be reasonably trivial to set up a bot that mass-imports metadata from Spotify to MusicBrainz (note that MB rules do not allow this, community cleanup from a single user doing this with another source, years ago, is still ongoing).

The value that MusicBrainz adds is the community editor who spent a few hours going through YouTube videos and wayback machine social links to figure out that Fog (Wellington, NZ, punk/post-punk) and Fog (Auckland, NZ, Post-Punk) are different bands - even if they share a Spotify profile. The editor that hunted down and listened to 5 compilations that have mixed up a radio edit and an original mix of a track, to find out which is which, and separate them in MB and make notes. [these are made up examples]

That's not to imply that these two projects are 'competing', or that the ISRC figure comparison isn't useful and correct. But community database + scraped data is apples and oranges. And a mixed fruit bowl is wonderful.

they also offer a bunch of stuff like

- searchable website

- incredible well thought out postgres database that differentiates, for example, a recording from a release (and much, much more)

- ability to replicate changes to said database hourly to your own environment

- workable system for schema updates

- cover art archive

- refined interface for submitting/moderating listings

- etc.

I was wondering if MB had any rules on such things. I get the motivation, but I hope they'd be willing to work with some trusted editors to figure out if this data would be useful/could be imported without risking quality.

But MB is one of the best resources out there - precisely because of what you said - so I'm not complaining too much :)