Martin Butler's tribute to Ted Codd in IT Week was both appropriate and very much appreciated.  Allow me to quote:

·   "The impact of [Codd's] work should have merited the depth of coverage that seems nowadays to be the preserve of minor celebrities and major politicians."

·   "It is an indication of the greatness of the man that no one has managed to challenge the fundamental truths that he demonstrated so many years ago."

·   "We should all pause and reflect on the greatness of the man, and the benefit that he brought to the world."

I endorse these remarks wholeheartedly.  They're no less than Ted deserved. That said, I'd like to suggest, politely, that Ted's achievement might be even greater than Martin Butler seems to realize.  Here are a few more quotes from his piece:

·   "Alternatives to Codd's relational database have come and gone, but with the arrival of XML there is a sea change on the horizon." 

·   "Vendors ... are now seeking to advance towards a new world of computing--one that can store and manage objects with the same ease that Codd allowed us to store and manage related data elements." 

·   "Perhaps in 10 years' time the younger generation will laugh at our reliance on relational databases ... [A] new database model [will become] generally accepted..."

No, there's no "sea change on the horizon" (at least, not a genuine one--certainly no more than there was a few years back when the object database advocates were touting their wares).  Nor will "a new database model ... become generally accepted," at least if such matters are to be decided on a scientific basis.  The fact is, Codd's relational model already gives us what we need in order to store and manage "objects" in general and XML documents in particular.  This isn't the place to get into the merits or otherwise of objects and XML documents as such; all I want to say here is that there might well be some utility in being able to keep such things in a database, but that database had better be a relational one, or we'll have walked away from one of the best technologies we're likely to see in our lifetime, or in several lifetimes.  One hundred years from now--so long as we base our technical decisions on rational considerations and not crass commercial hype--database systems will still be based on Ted Codd's relational model.

(Ed. Note: See my full critique of Butler’s comments Turning in His Grave? Genuine Respect for Codd’s Work.)

Posted 06/27/03

[ABOUT] [QUOTES] [LINKS]