Date: Wed, 13 Jun 2001 09:51:30 -0400
Reply-To: "Dorfman, Paul" <Paul.Dorfman@BCBSFL.COM>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: "Dorfman, Paul" <Paul.Dorfman@BCBSFL.COM>
Subject: Re: was "running SAS": is "shame on SAS"
Content-Type: text/plain; charset=iso-8859-1
What you are saying may very well be true, but personally I find
editor-centric discussions a bit silly. It appears as if the editor is a
single most important element of programming, while in reality, the actural
typing of a program and editing it is the last and the least important stage
of programming work. Moreover, SAS programs rarely become exceedingly large.
For that matter, the choice of an editor might be of more importance to a
COBOL programmer (I have seen COBOL programs with upwards of 500,000 lines
of code) - and those are mostly written using ISPF.
I find the native SAS program editor absolutely adequate in all
practicality. On PC, there are other useful features, but they are not
critical for getting the job done. On the Big Iron, almost nobody works with
dumb terminals directly any more, and PC interfaces (Extra, Rumba, etc.)
provide editing and/or cut-and-paste capabilities comparable to those of
Ultraedit. Moreover, one could use any PC editor he wants and then move the
code to the mainframe for execution. But given the average SAS program size,
I find it practical to work directly in ISPF. As far as the latter goes, and
without a slightest intention to start another editor war here, I honestly
find ISPF (and its PC SPF counterpart) to be absolutely the best programming
editor ever created. Again, it is my personal preference, and I do not
believe that the choice of an editor is critical - the code is. It is
further corroborated by the fact that a lot of good programming has been
done using VI, the most maladroit editor I have seen.
Paul M. Dorfman
From: mark.k.moran@CENSUS.GOV [mailto:mark.k.moran@CENSUS.GOV]
Sent: Wednesday, June 13, 2001 8:32 AM
Subject: was "running SAS": is "shame on SAS"
> Another reason I don't use interactive oftenly is that I use Unix SAS,
> Unix SAS don't have a good editor, WIN/SAS enhanced editor is pretty
> good stuff, but not in Unix SAS!
I agree, and shame on SAS! I once spoke with a fellow SAS user about the
fact that SAS on UNIX and VMS was "user-unfriendly" especially in regard to
editing compared to PC SAS. His response to me was that you can't use a
pickup truck to do the kind of work that requires an 18-wheeler. The
problem with this pat response is that the fundamentals of driving a pickup
truck and the fundamentals of driving an 18-wheeler are not, and should not
be, radically different. A pickup truck and an 18-wheeler both have
steering wheels, accelerator pedals, and brake pedals, for example.
Editing SAS code on PC and editing SAS code on many of these other
platforms is so radically different that on one you have the equivalent of
a steering wheel, on the other you have the equivalent of tricycle handle
bars -- only its not the expensive "18 wheeler" that has the nice steering
wheel as you would hope-- no, it's the *pickup truck* that offers the user
what he might expect as the basics, given today's state of the art in
computing. I really find SAS useful and well-conceived in so many ways,
but regarding basic editing, doing simple things like determining what
program created a particular file without having to resort to comments in
programs, or copying and pasting, the behavior of the interactive GUI in
UNIX SAS, etc., I say shame on SAS! It's so backward, and there's a long,
long way to go in some of these areas. Consider one more thing:
> Use a different editor. I do most of my win editing in K-edit, very
little in SAS.
Why on earth should users of SAS be expected to do that? Why should we put
up with having to go somewhere else than SAS itself, once in a blue moon
much less on a routine basis, for something so basic as merely editing
one's code? In my humble but totally accurate opinion, I think it is
totally bizarre that one should have to even consider using an editor not
supplied with SAS/Base for dealing with the most ordinary, day-to-day,
mundane chores in SAS.
> Ronald Smith
> Research Evaluation Coordinator
> Institutional Research
> Portland Community College
Blue Cross Blue Shield of Florida, Inc., and its subsidiary and
affiliate companies are not responsible for errors or omissions in this e-mail message. Any personal comments made in this e-mail do not reflect the views of Blue Cross Blue Shield of Florida, Inc.