Off topic: Database Unique IDs

Question

Hi folks,
This is a little off-topic, but I figure that there may be other people out
there whose Forte development would benefit from the discussion.
I am currently building a development framework for our new software
product. As part of that framework I'd like to include the facility for
generating unique, user-invisible, integer database IDs.
Now there is some doubt here that this is actually required and that the
primary key should be whatever the programmer wants it to be, including
multiple columns if necessary.
I was wondering if anyone can give us some rules-of-thumb regarding
the use of unique IDs as primary keys. Or if someone can point me to
some on-line resources (or even a good book) that can guide us in this
area.
The arguments I have given for using integer IDs are:
- Single, integer columns should be faster
- User invisible integer ID allows editing/duplicates of all user-visible fields
- Single, integer foreign keys would reduce storage requirements
- Standardising on integer IDs would allow generic functionality built into
framework
- More object-oriented as objects have "built-in" unique identity
I would appreciate any comments people have. We can take this
discussion off-list if that is preferable.
Cheers,
Duncan Kinnear,
McCarthy and Associates, Email: [email protected]
PO Box 764, McLean Towers, Phone: [phone]
Shakespeare Road, Napier, New Zealand. Fax: [phone]
Providing Integrated Software to the Meat Processing Industry for over 10 years
To unsubscribe, email '[email protected]' with
'unsubscribe forte-users' as the body of the message.
Searchable thread archive <URL:http://pinehurst.sageit.com/listarchive/>

Answer 1

Hi folks,
This is a little off-topic, but I figure that there may be other people out
there whose Forte development would benefit from the discussion.
I am currently building a development framework for our new software
product. As part of that framework I'd like to include the facility for
generating unique, user-invisible, integer database IDs.
Now there is some doubt here that this is actually required and that the
primary key should be whatever the programmer wants it to be, including
multiple columns if necessary.
I was wondering if anyone can give us some rules-of-thumb regarding
the use of unique IDs as primary keys. Or if someone can point me to
some on-line resources (or even a good book) that can guide us in this
area.
The arguments I have given for using integer IDs are:
- Single, integer columns should be faster
- User invisible integer ID allows editing/duplicates of all user-visible fields
- Single, integer foreign keys would reduce storage requirements
- Standardising on integer IDs would allow generic functionality built into
framework
- More object-oriented as objects have "built-in" unique identity
I would appreciate any comments people have. We can take this
discussion off-list if that is preferable.
Cheers,
Duncan Kinnear,
McCarthy and Associates, Email: [email protected]
PO Box 764, McLean Towers, Phone: [phone]
Shakespeare Road, Napier, New Zealand. Fax: [phone]
Providing Integrated Software to the Meat Processing Industry for over 10 years
To unsubscribe, email '[email protected]' with
'unsubscribe forte-users' as the body of the message.
Searchable thread archive <URL:http://pinehurst.sageit.com/listarchive/>