6 Mar
2012
6 Mar
'12
3:17 p.m.
Beman Dawes <bdawes <at> acm.org> writes:
The hangup is variable length data, particularly strings. I still don't think I've got the right solution to that problem.
Is that really a critical shortfall for b-trees? I know it is important for general purpose containers. Also, do you see fixing variable length data storage as an interface changer? Could the library go in using just fixed length records? Thanks for the update! I hope it bubbles up to the top of your queue again some year :-) Joel