I guess the question needs to be asked: is the current method indeed sufficient, or do we hold out for a capture of the undecoded composite video? (or even *gasp* the raw RF on the disc?)
Arguably, the current format is sub-optimal because it uses lossless storage for a signal that has already undergone "lossy" conversion from composite to component format. However, it's the best we can do without exotic, unproven solutions.
I suppose a third option is to use a lossy format with smaller files. This would make a lot of people happy, but goes against the goal of preservation, so I'm not keen on this alternative.
All is not lost, though... Captures are happening that are compatible with the current CHD format, but there's not much interest in releasing them if they're likely to be replaced.
And that's where things have been stuck for quite a while now.