Mixed CaSe Identifiers

From: Paul Kent <kent_at_manzi.unx.sas.com>
Date: Tue, 16 Nov 1993 02:09:33 GMT
Message-ID: <CGKBBy.603_at_unx.sas.com>


Hello

We are modernising a product that dates back to the 70's when all you needed for a name was 8 characters and case was not an issue because most printers used uppercase only.

I have some questions that you might have an opinion on. This questionaire has been posted to comp.databases.{many} and a summary will be posted to comp.databases (given suitable interest)

We are wrestling with the issue of mixed case names. Some contemporary DBMS software allows them and treats "Aname" different from "aname" which seems contrary to SQL2 -- it has a syntax rule that says names are folded to uppercase.

My Questions:

 do you find it useful to have "Aname" be a different  identifier than "aname"

 do you have applications whose tables/columns have names  that differ in case only  

 have you found "middleware" software that handles this  distinction gracefully

While i have your attention, allow me one last one...  

 how long should identifiers be?

Mail me yer answers, and i'll summarise if there is enough interest

Thanks

Paul Kent
SQL R+D, SAS Institute Inc.

-- 

Paul Kent (Base R&D)                  " nothing ventured, nothing disclaimed "
kent_at_unx.sas.com         SAS Institute Inc, SAS Campus Dr, Cary NC 27513-2414.
Received on Tue Nov 16 1993 - 03:09:33 CET

Original text of this message