Re: Stupid Database Tricks
Date: Fri, 25 May 2007 12:51:00 GMT
Message-ID: <UYA5i.20646$Qz.9388_at_trndny09>
"Roy Hann" <specially_at_processed.almost.meat> wrote in message
news:hdGdnbvOwKYTD8vbRVnytQA_at_pipex.net...
> "David Cressey" <cressey73_at_verizon.net> wrote in message
> news:FDt5i.8082$hw.4094_at_trndny08...
> > Good thought. A lot of people who propound this stupid trick as a "best
> > practice" are thinking in terms of chains of pointers. It's the revenge
> > of
> > the CODASYL Zombies.
>
> I don't buy it. *I* was a network DB coding zombie in my youth and it was
> precisely because I endured that pain for several years that I fell
> instantly and hopelessly in love with the relational model when I first
> encountered it. I don't think I am likely to be very different from
anyone
> else from that era. It is highly implausible to me that anyone who worked
> with that shit would want to perpetuate it after being exposed to a better
> way.
>
I am sure you speak for many, if not most CODASYL DBMS programmers. without contradicting you, I'm going to say for sure that the opposite camp existed, and has existed continuously to this day.
In the mid 1980s, I was fortunate enough to be called to teach DEC Rdb/VMS programming and design. Some of the people who knew VAX DBMS were the hardest to reach in the course of a one or two week course. They simply saw the relational model as missing the pointers without which they couldn't see the sets.
But the real problem was with the VAX DBMS instructors (of whom I was never one). These people kept getting bad reviews when teaching Rdb. I finally got to observe one of them at work. She was teaching Rdb design, but she was still selling network design, at an almost subverbal level.
In the 1990s, I got to do short term contracts working on lots of Rdb databases that were in sad shape. I can't tell you the number of them that had been designed by reinventing plex structures. They ran slow, and they inflicted all the pain on Rdb programmers that CODASYL had inflicted on the previous generation.
> It is just the new, naive manpower flooding into the workforce who keep
> rediscovering and re-entrenching the obvious approach. Who is going to
> think of looking for plywood if they can plainly see a pile of logs and
they
> need a cabin quick?
>
I'm going to suggest that, at a certain point of discovery, a seasoned CODASYL Zombie *is* new, naive manpower. He's been born again, but he's still naive. Kudos to you for not staying in that stage of development very long. From what you write, I'm guessing you passed through that phase in anywhere from 15 minutes to two weeks. Received on Fri May 25 2007 - 14:51:00 CEST