Re: Do this. Don't do that. Can't you read the signs? #10

The only reason I can imagine for this limitation, other than wanting to be
compatible with ... v6?, which would be silly at this point, is
compatibility with other software on export.  I think SPSS had trouble
reading >8 length character variable names from SAS at one point (though
this is no longer true), and undoubtedly other software might.  Therefore,
not knowing what other software might be used with your data, keeping to <=8
might be best (in certain environments).

In general I'd hope we were beyond that entirely, of course, but who knows
....  I certainly don't keep to 8 character names entirely, but when I don't
have any reason to go over 8 characters, I do.


On Mon, Aug 17, 2009 at 8:08 AM, Paige Miller <paige.miller@kodak.com>wrote:

> On Aug 13, 11:23 am, iebup...@GMAIL.COM ("Data _null_;") wrote:
> > The clinical trails SAS programming community is asking for input on
> > this document.
> >
> >
> http://tinyurl.com/q3gwmyhttp://www.sascommunity.org/wiki/Good_Programming_Practice_for_Clinic.
> ..
> >
> Can anyone explain why this "best practices" document currently reads
> that data set names and variable names must be both "meaningful" and
> limited to 8 characters? I thought that restriction went out with SAS
> 5 or SAS 6.
> So now SAS gives us the ability to write more meaningful variable
> names and data set names (up to 32 characters) and the best practices
> document wants us not to use this useful feature in SAS, as if it was
> still 1988. Maybe they are writing a best practices document for 1988.
> This is just one of many infuriating recommendations in this
> document.
> --
> Paige Miller
> paige\dot\miller \at\ kodak\dot\com
snoopy369 (1724)
8/17/2009 4:19:19 PM
comp.soft-sys.sas 142828 articles. 3 followers. Post Follow

0 Replies

Similar Articles

[PageSpeed] 29