Closed LoopRethink the Universal Data ModelThe universal data model is unattainable, as Joshua Greenbaum argues in "The Quest for the Universal Data Model" (May 31), but the idea of integration of enterprise-level systems, shared data, and a high level of universality is attainable. The reason a universal data model is unattainable is that the fundamental methods and constructs — the way that data is conceptually formed and represented — is very flawed. The best industry practices in data modeling and data structure are barely adequate for a single application. The resulting data models and database structures have none of the properties necessary. With these flaws, the phrase "universal data model" is an oxymoron. The first issue is quality, not conformance. Being big, powerful, and able to force conformance to a declared industry standard isn't going to change what is fundamentally not "universal" in its properties. Data experts need to step back and rethink what an entity is, what an attribute is, and what a relationship is, and how each is conceptually and physically represented. Are they even the best concepts to begin with? We must also rethink the concepts of activity and protocol. Natural patterns are universal. They are consistent and support a variety of events and circumstances. Why not look for the "holy grail" in something that's at least amenable to and has the properties of universality? There's no reason on earth to accept current best practices in data modeling as very good at all when it comes to supporting the universality we seek for enterprise systems integration and information integrity. I'm certain what we seek is there, if we just rid ourselves of the current fundamentals of data models and rethink, reengineer, and retool. James Pennington Virtue Has No PriceI'm not sure if Don Tapscott just had a kind of "senior moment" or if he suffers from the common American illness of moral relativism (in "They'll Find Out Anyway," June 17). Either way, he's just plain wrong on this one. Don Carty's downfall was, in fact, due completely and utterly to his executive-suite gluttony (and that of his board and upper managers). Carty's "sin" wasn't in failing to be "up front" about his greed, it was in his willingness to screw everyone else for the benefit of himself and a select few. Tapscott suggests that had Carty simply told everyone they would have to weather layoffs, paycuts, longer work hours, and what-have-you while he and his little band of thieves would take what they were already getting and more, everything would have been fine. Sorry, but I'm not buying it. To begin with, even if he could have pulled off his little stunt by doing so, it doesn't give him a free pass. Carty was wrong to do it, and the attempt to pull it off alone (regardless of its success or failure) shows him to be unfit for the mantle of leadership. Beyond that, can Tapscott actually say he believes that had the union known in advance about this that they would have simply said, "That's great. Thanks for keeping us in the loop on these types of matters. Oh, and by the way, we and our membership will be happy to forfeit our hard-earned dollars so that upper management personnel can continue to live in the style to which they've become accustomed"? I don't think so. It's time for us to stop believing that we must rationalize good behavior. We shouldn't try to hide our greed because it might hurt morale and that might hurt employee productivity. We should reign it in because it's wrong! We shouldn't treat employees with respect because that helps retain them, thereby reducing training costs and improving the bottom line. We should respect them because it's proper! Marc Thornsbury
|
 |
Most Popular This Week
IE Weekly Newsletter
Subscribe to the newsletter
| |
|
|