Xref: utzoo comp.sys.dec:967 comp.arch:7947
Path: utzoo!attcan!uunet!husc6!hscfvax!pavlov
From: pavlov@hscfvax.harvard.edu (G.Pavlov)
Newsgroups: comp.sys.dec,comp.arch
Subject: Re: DECstation 3100 info.
Summary: glass houses
Message-ID: <709@hscfvax.harvard.edu>
Date: 19 Jan 89 20:55:58 GMT
References: <979@isieng.UUCP> <85330@sun.uucp> <558@oracle.UUCP>
Organization: Health Sciences Computing Facility, Harvard University
Lines: 26

In article <558@oracle.UUCP>, rbradbur@hqpyr1.oracle.UUCP (Robert Bradbury) writes:
>  [ a long diatribe on the lack of absolute standardization among Unix-based
     hardware vendors ]

   As a user, I have not found the differences between the systems we have used
   particularly onerous.  Maybe it is the nature of our particular applications,
   but the majority of our intra-Unix ports have been relatively painless, some-
   times surprisingly trivial.  Nothing compared to the "bad old days" when we
   tried to move between hardware vendor-specific operating systems.

   But what has been a royal pain is moving data between different brands of 
   dbms's.  Each vendor insists on storing data differently, using proprietary
   interfaces between front ends and back ends, etc.

   The DBMS industry has overwhelmingly declared support for the "SQL standard".
   Too bad that it is so far away from reality.  Differences in syntax, exten-
   sions, and host-language call implementations make this a hollow marketing
   claim.     

   I appreciate the problems caused by Unix variants.  But the Unix industry
   is way ahead of the dbms industry in offering inter-vendor operability and
   portability.

   greg pavlov, fstrf, amherst, ny