[
Lists Home |
Date Index |
Thread Index
]
bob.ducharme@lexisnexis.com (DuCharme, Bob (LNG-CHO)) writes:
>>2) Create a mapping between formats A, B, C, etc., or possibly from
>>those formats to a new format _if_ a hub seems more useful that
>>connecting all the pieces individually.
>
>If the hub is more useful, it will be because it's a form of linking,
>generalized from formats A, B, and C, that can be shared among users
>of A, B, and, C. I was going to say "Aha! So you admit that
>generalized linking can work!" but of course what works for A, B, and
>C, won't necessarily work for X, Y, and Z. I suppose the dream of
>something generalized enough to work for A - Z is unrealistic, but I
>like to hope that there might be something that works for A - T (80/20
>rule and all that). That's why any such efforts attract my attention.
I think the hub - and likely multiple hubs - has to be allowed to
evolve, not derived by committee and imposed.
The 80/20 rule, while a hopeful indicator, seems to produce different
result for everyone looking at a problem set, especially one as complex
as linking.
|
