Software: Success through Sophistry

While differences are expressed, what is more noteworthy is the common viewpoint that emerges, implicitly or explicitly, from all three observers: that competition in the software market is more a matter of deception than of technological progress.

Uriel Wittenberg, June 13, 2000:

I read with interest your article, “What Do You Mean, ‘Post-Relational?’ ”.

First, let me commend you for doing something unusual: naming your target (Joe DeSantis, Director of Software Development at InterSystems Corporation). Critics of all kinds usually don’t have the guts to be specific like this.

What I find unfortunate in your article, however, is that you have avoided the fundamental objection to Mr. DeSantis’s article -- which, as you surely recognize, is his intent to deceive his audience in furtherance of his own pecuniary interests.

I’m sure you also realize that Mr. DeSantis is but one isolated example, and that this problem is absolutely pervasive in the computer field. Indeed, DeSantis and others are acting in accordance with the incentives faced by the industry. Given this situation, responding to Mr. DeSantis is not enough. There is an endless supply of DeSantis’s.

I’d hope you further agree that your response in this case is not workable as a general approach to combat the problem. You have constructed a conscientious rebuttal of Mr. Desantis’s claims, as if the technical issues were actually the point.

One might compare this to a situation where a malfunctioning database application is serving up the wrong phone numbers when names are looked up. Do you turn to alternative sources to find out the right number each time an error arises? Or do you fix the system?

How to fix our “system,” in which the profit incentive leads most players to engage in deception -- generally with impunity?

As with all problems, the first step is to acknowledge it. Then what’s needed is a wake-up call to confront it head-on. Specifically, I’d propose something like the following:

The pitiful state of today’s most popular DBMS products -- [XXX?] years after the foundations of relational database theory were laid out -- is a situation that should be impossible in a world where business enterprises are serious about pursuing profit. That it is nevertheless our reality is largely explained by the fact that one of the software industry’s most important products is not software at all, but illusion -- illusion generated by the army of professional flatterers, seducers, entertainers and deceivers whom software producers depend upon to sell their products.

Why is this army indispensable? In a word, because of the terrible credulousness of corporate America -- or more specifically, the IT managers who uncritically swallow the sales pitches of the industry and spend their employers’ money on shoddy products, with phony features, backed by misleading claims.

Let us take a look at a typical example of the type of deceit that has propelled this terribly “fast-paced” industry -- whose glacial progress year after year continues to cripple the productivity of DBMS users the world over with buggy and ill-designed products.

[Insert brief, non-technical description of DeSantis’s nonsense. You are C.J. Date. You don’t have to prove such a plausible claim (misleading promotion of a software product!) in this article. Others can do that -- or you can point to a webpage for technical details. The people who matter -- the decision-makers -- won’t look at it, but can send their minions to check it out if they want.]

If you happened to be interested in my participation or assistance in developing an article like this, I’d be delighted to discuss it. I have a brief essay along this kind of theme which might interest you, at http://urielw.com/reasonfdn.htm.

Regards,

Uriel Wittenberg

C.J. Date, July 25, 2000:

First of all, Mr. Wittenberg commends me for “naming [my] target (Joe DeSantis, Director of Software Development at InterSystems Corporation).” Well, OK (and thanks for the compliment) -- but I hope it’s clear that my “target” was Mr. DeSantis’s opinions as stated in print, not Mr. DeSantis himself. What’s more, my own article was certainly not meant as any kind of ad hominem attack. I don’t like ad hominem arguments, and I tried to stay within appropriate technical bounds.

Second, Mr. Wittenberg goes on to say that “the fundamental objection to Mr. DeSantis’ [original] article ... is his intent to deceive his audience in furtherance of his own pecuniary interests.” Well, maybe so. However, I’m a charitable soul, and I wouldn’t accuse Mr. DeSantis, or anybody else, of acting in bad faith unless and until the evidence of deliberate deceit was overwhelming. In other words, I would prefer to assume that Mr. DeSantis actually believes what he was saying in his original article. To me, therefore, the “fundamental objection” to that article was not deliberate deceit on the part of the author, but rather the vast lack of understanding it displayed of the very technology it was pontificating about, and the further implications of that lack of understanding (see below).

Anyway: Regardless of whether the problem is one of deliberate intent to deceive or merely one of ignorance, I agree with Mr. Wittenberg that it can’t be solved on an instance-by-instance basis. In other words, I agree that attacking individual writings “is not workable as a general approach to combat the problem”. However, I do think people learn from examples. In part, therefore, what I was trying to do in my critique of Mr. DeSantis’ article was the following:

• To suggest that many if not most articles on this subject in the trade press can be similarly deconstructed;
• To demonstrate the huge confusion, lack of clear thinking, and lack of intellectual rigor to be found in such articles;
• More generally, to encourage critical thinking on the part of readers; and
• Last but not least, to highlight a number of important technical facts from the field of database technology that don’t seem to be nearly as widely understood and appreciated as they need to be.

For purposes of future reference, I’d like to repeat some of those “important technical facts” here:

• The logical and physical levels of the system must be kept clearly apart (though they often aren’t).
• Relations are n-dimensional, not two-dimensional.
• Domains and relations are together both necessary and sufficient to represent absolutely any kind of data whatsoever.
• The one good idea of object technology is support for a proper type system, and the relational model already includes such support.
• Object technology includes several bad ideas (object IDs are one of the worst). Objects and a good model of type inheritance are incompatible.
• Object technology is not a good basis on which to build databases in the classical sense of that term.

And I’d like to add one more, one that I didn’t call out explicitly in my original attack:

• A true “object/relational” DBMS would be nothing more nor less than a true relational DBMS.

Back to Mr. Wittenberg’s letter. He asserts that the real problem is that “the profit incentive leads most players to engage in deception -- generally with impunity”. Clearly, he might be right in this claim (my remarks in the foregoing notwithstanding). He therefore proposes a specific approach (a “wake-up call”) for confronting the problem “head-on”, involving the production of what might not unreasonably be called a polemical essay ... Myself, I don’t really feel capable of writing an essay of the kind he suggests, but I’d very much like to see such a thing, and I’d like to encourage him (strongly!) to write one himself. Soon.

Fabian Pascal, July 25, 2000:

The question of what is the most effective way to meet this need has continuously preoccupied me for more than 15 years. Those old enough to remember know that for most of those years I have been doing online, in writing, and at industry events, precisely what Mr. Wittenberg suggests, namely: naming names and exposing the claptrap irresponsibly disseminated by the industry and trade press, both out of ignorance and disingenuously. And in fact, I had numerous discussions with Chris Date on this very subject, to persuade him that the understated, gentlemanly, “stick to the technical”, rational discourse style that characterizes his work, is ineffective and mostly wasted on the fad-driven, anti-intellectual, cookbook-based mode in which the computing industry in general, and its database sector in particular, operate (our very responses here are an excellent example of the difference between our styles).

Unfortunately, Mr. Wittenberg is new to the game [of criticizing the industry] and has not spent time in the trenches. His approach to “fixing our system”, in which “the profit incentive leads most players to engage in deception” and “credulous IT managers rely on professional flatterers, seducers, entertainers and deceivers,” indicates another type of credulity: He believes that (a) the problem is at the software industry level and that (b) first acknowledging it, then following up with a “wake-up call to confront it head-on,” will produce the fix.

But the problem is social and much more profound. It cannot be addressed at an industry level, let alone by sheer “wake-up calls”. The “system” is extremely effective in simultaneously (a) inducing the public to conform, via propaganda and a regimen of rewards and punishments (of which most are not cognizant), and (b) marginalizing the kind of wake-up calls Mr. Wittenberg suggests. What is more, I can assure him that the price paid for them is enormous relative to the negligible impact -- if any -- that they are likely to have.