context: (vb6 / ado / .mdb / jet4.0 / not using access)
hypothetical problem
say I need to track the properties of boxes.
There are three kinds, cardboard, wood, and sheet metal.
Each box has some common properties, width, depth, height
So one table makes sense to store that....tblBoxe s
but...each box type also has properties that only apply to that type...
cardboard...fib er content, IsRecycledBool, etc
wood...species, etc
metal....gageSi ze, IsGalvanizedBoo l, etc
so if I have one table, each entry will have lots of nulls for the
properties that don't apply,
and the calling code will have some kind of branching code to decide what
properties to query???
or should I have 3 tables tblCardboardBox , tblWoodBox, tblMetalBox
then the calling code still needs branching code to decide what table to
query but there aren't a bunch of nulls...
is there any argument *against* lots of null values for "not applicable"
fields?
is it easier to make branching code to access different fields versus
different tables?
any suggestions?
Thanks
Mark
hypothetical problem
say I need to track the properties of boxes.
There are three kinds, cardboard, wood, and sheet metal.
Each box has some common properties, width, depth, height
So one table makes sense to store that....tblBoxe s
but...each box type also has properties that only apply to that type...
cardboard...fib er content, IsRecycledBool, etc
wood...species, etc
metal....gageSi ze, IsGalvanizedBoo l, etc
so if I have one table, each entry will have lots of nulls for the
properties that don't apply,
and the calling code will have some kind of branching code to decide what
properties to query???
or should I have 3 tables tblCardboardBox , tblWoodBox, tblMetalBox
then the calling code still needs branching code to decide what table to
query but there aren't a bunch of nulls...
is there any argument *against* lots of null values for "not applicable"
fields?
is it easier to make branching code to access different fields versus
different tables?
any suggestions?
Thanks
Mark
Comment