David -
I would still encourage you to investigate with real code option (1). My
feeling is that the additional overhead of a heirarchy will be low. We
need a real example to show that. I would be sure that you have correctly
rejected (1) before starting on (2).
Otherwise we will end up with probably a NxN set of language<->relational
mappings for sequence orientated data depending on the use case when
I do not honestly believe this is necessary. I don't think you can tell
without trying it.
However - if you are coding it, you do get to decide how to do it ;) I
just get to make comments...
-----------------------------------------------------------------
Ewan Birney. Mobile: +44 (0)7970 151230, Work: +44 1223 494420
<[EMAIL PROTECTED]>.
-----------------------------------------------------------------
_______________________________________________
Biojava-l mailing list - [EMAIL PROTECTED]
http://biojava.org/mailman/listinfo/biojava-l
