Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : IFMX - Investment Discussion -- Ignore unavailable to you. Want to Upgrade?

To: surfinSteve who wrote (9842)	3/8/1998 2:09:00 PM
From: Charles Hughes	Respond to of 14631

>>>Chris Date, along with Ted Codd, is widely credited with pioneering the relational database model. But Date is highly critical of how vendors have warped his creation over the last twenty years and is issuing dire warnings of potential catastrophic failures. In addition, Date says the SQL language is much too complex and the time has come to replace it with something more viable. <<<< I think this is mostly right. Date was Codd's student, I believe. He did not however, invent RDBMS. He did help spread the word and wrote important texts on relational databases that are still standard works. He did other important work on theory of structure and use. Codd, while having invented relational databases, has always been critical of implementations, including SQL. SQL was done the way it was partly against his will. As he tells the story, at first IBM only wanted something you could embed in COBOL and PL/1. (So as not to cannibalize other product lines, etc - the usual B.S.) Thus the first variants of SQL were of the embedded type. So it never developed all the full features you expect in a language. SQL itself is dead simple of course. Simpler than any other computer language I have used. However, what makes it complex is those missing features that irked Codd from the beginning. Because SQL doesn't include all the basic features a language must have, everybody has invented their own. Therefore you can't move code easily from one system to another, because everything becomes tool dependent. As when you embed SQL in Java or markup scripts, for instance. This wheel has been reinvented thousands of times now since the 60's. It is long past time to do a better job of this. Perhaps via objects that are 'closer to the metal' than the wrappers typically used, and more appropriate to relational underpinnings than the object strategies out there. Anyone for standardizing some of the work Informix has done lately? Java JDBC? Or something completely new? Chaz