Just one more point to throw in to this discussion about outcomes.
When I was still teaching, I tried a few times to use outcomes to record marks for BTEC qualifications. These work by having a number of Pass (P1, P2, P3, etc.), Merit (M1, M2, etc.) and Distinction (D1, D2, etc.) marks across as many assignments as the college wants to provide for each unit. Each mark can appear in more than one assignmet (theoretically anyway, it rarely happens in practice), although each student can only gain each mark (P1, P2, M1, etc.) once for a unit.
To gain a 'pass' for a unit, the student needs all the pass marks for that unit. All the pass and merit marks are needed for a 'merit'. All the marks are required for a distinction.
I found that using outcomes for this was very unwieldy, as you had a large number of 'outcomes' in each course (especially if each course contained multiple units, in which case they ended up as U5P1, U5P2, etc. for Unit 5 P1). Each outcome also only needed a grade of 'Yes' or 'No'. There was also no sensible way to gather these grades into an overall pass/merit/distinction for a unit.
It may well be that this use case is completely outside of anything that outcomes will ever sensibly address (and maybe there needs to be some different mechanism to support such 'exotic' marking schemes - some sort of 'yes / no' criteria, rather than graded outcomes).