Earlier I published a couple of critiques of Simon Williams’ and Lazy Software’s so-called “Associative Model of Data”, one in response to a reader who insisted it was a new and useful data model. I recently received a message from a practitioner who put my critique to good use in practice, demonstrating that, contrary to the common fallacy, there are practical implications from theory and data fundamentals, that users can rely on to see thru questionable vendor claims (to put it politely).

The reader is invited to ponder the significance of the author’s preference to withhold his name and that of his employer in a context in which the latter actually benefited!!! What does this say about an industry in which practitioners need to remain anonymous when they are right, and about the chance for eliminating ignorance and regression?

From: SC

To: Editor

Date: Oct 25, 2003

Incidentally, you may be interested to know that the way I came about your website was that a few weeks ago, we had the sales guys from Lazy Software come in and try and sell us their product based on William's Associative "Model" of Data. And ashamedly I have to admit that we were all rather impressed, because, well on the surface it seems like a great product. It generates all your data entry forms for you etc. and takes away a lot of the hassle of building an application around their database. Of course I now know that such a tool based around a TRDBMS would be a much better option and just as feasible (if such a thing existed).

While I was impressed by their tool, I came away with some nagging concerns about their "model" despite my lack of data management theory. So I typed "criticism associative model data" into Google and found your site via your exchange on the "model". And thank heavens I did! I certainly won’t be recommending that we buy their product, so you may be happy to know that your site proved effective in combating the FUD of AMD.

After reading a few of your articles, I downloaded and read William's paper on his "associative data model". It was a truly disturbing read. I felt like publishing a criticism of it on the web (and may still do so yet), until I realised that to do it proper justice I'd have to attack almost every single statement in the paper! It really surprised me that someone could build a company based on such a poor foundation, but I guess I didn’t realise the true sad state the IT industry is in. I know it doesn't surprise you. I fear it can only get worse as our universities keep churning out more "tools-focused" Comp Sci graduates into an industry that is happy to accept them. And that's why I'm more than happy to help spread the word.

Ed. Comment: Generating screens does not a DBMS make. SC has also discovered why it is neither feasible, nor worth it to debunk so much of the nonsense being published these days. Too often we must invoke Date’s Incoherence Principle: It is not possible to treat coherently that which is incoherent.

Posted 1/30/04