Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
Home -> Community -> Usenet -> c.d.o.server -> Re: Naming Convention for Columns
Jason McElroy wrote:
>
> I always use types as well. My convention is:
> ID = RDBMS generated unique identifier.
> NUM = number that is not generated by the RDBMS. Could be generated
> by external systems or programs
> COUNT = contains a count of some entity
> TYPE = a category of some sort
> CODE = a short identifier for some entity with a longer name. E.g.
> company_code = AMEX
> NAME = the proper name for an entity. E.g. company_name = 'American
> Express'
> DESC = a description of an entity or code. E.g. error_desc = 'list
> index out of bounds'
How 'bout adding these:
date - where only the days/months and years count. If your database has
only datetime, set the time to 12:AM or whatever the default is when you
stuff 3/27/97 into one of these beasts.
time- where only the time matters.
datetime- where both matter.
For those who insist on abbreviations:
DT, TM and DTTM
Now, for financial considerations:
trade for trade
setl for settlement
local for local
cstmr for customer
cstdn for custodian
account for account. I've seen a show where they use accnt. It saves two
letters. Why bother?
instit for institution
broker for broker
My general rule is 7 is the max number of letters you want to type. Which jives with a number of psychological studies that have concluded the human brain has trouble with more than seven. That's why your phone number is seven, your Social Security id is 3-2-4 (3+2=5<7 or 2+4=6 <7), etc. Received on Tue Mar 10 1998 - 00:00:00 CST