Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!wuarchive!uunet!ogicse!cs.uoregon.edu!oregon!joe
From: JOE@oregon.uoregon.edu
Newsgroups: comp.sys.next
Subject: Re: Alpha version of SAS for the NeXT shown at SUGI 15
Message-ID: <19029@oregon.uoregon.edu>
Date: 13 Apr 90 15:40:12 GMT
References: <8408@tank.uchicago.edu> <36205@brunix.UUCP>
Distribution: usa
Organization: University of Oregon
Lines: 43

In article <36205@brunix.UUCP>, rca@cs.brown.edu (Ronald C.F. Antony) writes:
> This kind of policy by SAS is really a bu....it! If they want to wait
> until the cube is a success before they sell their software, they they
> rather should sell no software at all. How should the cube ever become
> a success if developers have this attitude? Do they really think that
> there will be any users other than a few crazy ppl like us, shelfing
> out a couple of thousand $$ just to make the cube a success and to
> wait until finally get some software? It really has to work the other
> way round...
> 
> Ronald

I'm not sure this is really fair -- SAS has clearly already put a LOT of time
into this product. Remember, porting the SAS System to a new platform isn't 
exactly like porting 200 lines of vanilla C...

The version I saw included ALL of the SAS add-on products (such as SAS ETS,
SAS/Graph, etc)... All told, this is some 200 MEG of code. While some of the
SAS people have stated that they've reduced the system dependent kernal to
less than 1% of the total code, this still represents a couple of MEG of
very tricky code used to do dynamic linking, exploit operating system features,
and do other non-standard things.

Frankly, I think you *should* say, "Thanks, SAS Institute, for demonstrating a
product many of us would dearly like to have, even if it isn't a full 
production release. We're glad to know you're at least working on it, and 
it is nice to be able to SEE FOR OURSELVES how it is coming!"

I'm as critical as anyone else of the NeXT's limited software base right now,
but give SAS Institute a break! They do EXTENSIVE quality control on their 
products before they release them, and I, for one, am willing to wait for a
solid product that I can RELY on to provide numbers which are RIGHT.

The vendors you should jump on, in my opinion, are the ones who've either 
refused to port their products (or, even worse, the ones who've SAID they'll 
port their products but have accomplished absolutely nothing in that direction 
in so far as anyone can tell).

Just my two cents,

Joe St Sauver

Disclaimer: I don't work for SAS Institute; all opinions are strictly my own.