Date:         Mon, 21 Dec 1998 09:33:01 -0700
Reply-To:     "Jeff J. Voeller" <atrocity@MY-DEJANEWS.COM>
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From:         "Jeff J. Voeller" <atrocity@MY-DEJANEWS.COM>
Organization: Deja News Mail  (http://www.my-dejanews.com:80)
Subject:      Re: hashing, big formats,etc.
Content-Type: text/plain; charset=us-ascii

On Sun, 20 Dec 1998 15:50:19   david pider wrote:
 
> I wonder if somebody else on the list is annoyed by certain academic
> types polluting the list with their homegrown "routines".
 
[saracasm mode on]
Yes, I'm completely disgusted with it.  If I wanted creative solutions
to real-world programming problems, I'd read the alt.sex.* hierarchy.  I
subscribed to SAS-L for the flaming and to watch people get upset at
each other for not all treating the same problems identically.  I'm glad
to see that _you_ understand the true purpose of this list.
[sarcasm mode off]
 
> For instance, what is this 'hahsing' BS?? To code something like that
> one must've never worked in the industry. I've been programing for
> years and never even heard the term!
 
Or maybe they've simply worked in a different industry from you.  I've
been coding SAS for over eight years now and I've never used any of the
obscure-to-me statistical procedures that get talked about here, but
that doesn't mean I arrogantly assume that they don't have real-world
value.  The fact that you or I have never used/needed something hardly
suggests that it has no value.
 
Maybe these crazy-to-you suggestions have never come up because you have
either unlimited resources or relatively small datasets.  Some of
us--and I've been in this position many times--have had situations where
the vanilla SAS solutions made perfect sense conceptually but failed due
to resources.  Live with it.
 
> No sane manager will allow this kind of stuff in production. I'd fire
> anyone on the team who'd have the audacity to code a monstrosity like
> that and claim it works better than merge or sql.
 
Ah, so obviously you tested it and found that it _didn't_ work better.
I mean, I'm sure you're not just venting because someone solved a
problem that you didn't need to...or couldn't figure out how to.
 
> Or what about this 'big format' thing?
 
For the record, Karsten and I quite independently of each other have
recently found large formats to be a solution to real-world problems we
were experiencing.  In my case, I needed to select tens of thousands of
rows from millions in a flat file.  Creating a format that simply
converted policy number to "YES" and using that on my INPUT statement
was a major time and resource saver.  I'd post the syntax here, but it
appears that it might upset you.
 
> If simple merge was used nobedy would have no problem in the first
> place.
 
Merging millions into tens of thousands and then discarding most of
those millions?  No thanks.  Not in an MVS shop with tight resources.
 
> BTW, I've noticed that those posting those extravagant "methods" are
> never on the money trying to answer a normal question about standard
> SAS coding. Anybody thinks it's a coincidence?
 
Perhaps mundane problems don't interest them, just as real-world ones
don't interest you.  By the way, what solutions have you posted here
lately?
 
 
-----== Sent via Deja News, The Discussion Network ==-----
http://www.dejanews.com/  Easy access to 50,000+ discussion forums