At 10:24 14/09/00 -0400, Ian Whitlock wrote (in part):

>In SAS, macro provides the ability to provide flexible and complex >programs that could probably not be otherwise produced in SAS.

Ian, let me first say that I agree with most of what you wrote. In particular, with SAS 'as it is', one has little option but to 'master', at least to some extent, SAS Macro in order to be able to use SAS for many purposes.

However, is it not the case that many of those programs which "...could not otherwise produced in SAS" SHOULD be able to be 'otherwise produced in SAS'? In other words, could it not be said that a lot (but, obviously, by no means all) of the need for SAS Macro exists because of deficiencies in the SAS Language, and the implementation of SAS PROCs? In particular, one could cite the traditional absence of user-defined functions and subroutines in SAS as being a very common cause of the need to resort to SAS Macro, particularly in relation to DATA step programming - and perhaps also the traditional inability to use 'DATA step code' within most PROCs.

... that's how it seems to me, anyway.

Kind Regards,

John

---------------------------------------------------------------- Dr John Whittington, Voice: +44 (0) 1296 730225 Mediscience Services Fax: +44 (0) 1296 738893 Twyford Manor, Twyford, E-mail: John.W@mediscience.co.uk Buckingham MK18 4EL, UK mediscience@compuserve.com ----------------------------------------------------------------