1 May 2010 06:29
Re: Pre-RFC: The Advanced Vocal Tree proposal
On 30. april 2010 11:18, Brian Schweitzer wrote: > On Wed, Apr 28, 2010 at 3:13 AM, Leiv Hellebo wrote: > This would be a simpler discussion if the five releases criteria was > followed, I think :) > > > The negative there, in my eyes, would be that the resulting tree > wouldn't show in order or appear completely. Let's say we do adopt a 5 > release criteria, then apply that to something like fach or choir type. > First, we're likely to more often run into the need on works, rather > than releases - so we'd need to expand it to works as counting for the > 5. Then we get the voice types added in any random order - and to the > best of my knowledge, the ordering of same-level children within a node > of the AR tree is based on order added, and is not controllable by the > relationship editor? Third, we now get incompleteness within the tree - > as I said before, the instruments tree needs this type of basis, as the > list of potential instruments is essentially unlimited. The same is not > true for types of voices or choirs. Those are quite limited, and I just > can't justify why we'd want a situation where, say, soprano is complete, > but all subtypes of bass (or choir-type) are omitted simply because > noone, indiividually, requested them yet. That's why I'd suggest we > simply get it right the first time, then do it, rather than building it > up piecemeal one by one. > > Lastly, when the tree would finally be complete, the 5 release process > would still remain - the potential would exist for redundancy or 'almost > redundancy' in the tree. I'm saying this badly, but what I'm getting at > is this: what happens when someone requests something, gives 5 > releases/works, but the tree is already full, and the thing they're > requesting is either heavily debatable or otherwise problematic?(Continue reading)
RSS Feed