CTO Articles
Published in IT World
September 27, 2005
From middleware to metroware
By Sean Mc Grath
I have an idea for the worst geek B Movie plot of all time. My working title is 'Attack of the Killer Market Segment'. Picture the scene ... The movie opens at the gates of an unkempt graveyard, late on a full-mooned, breath-fogged night. A shadowy figure can be seen brushing away the undergrowth to peer at the inscriptions on the headstones. We zoom in as a shaky but well manicured hand, clears away an inscription which is caught in a ray of moonlight. The Latin inscription reads simply:
Tertium non datur.
In sub-titles we display the English translation:
The law of excluded middle.
The shadowy figure drops to his knees, sobbing uncontrollably. We zoom out and can see that the soles of his shoes have also caught a ray of moonlight. The shoes look expensive. On one shoe we can see the letter 'V'. On the other, the letter 'C'.
The plan is for the next 3.5 hours of the movie to be a series of plot-revealing flashbacks. First we learn that the VC acronym on the man's shoes stands for 'Venture Capitalist'. Later we learn that Acme Tertium Middleware Inc. was an application middleware company the VC had invested heavily in.
As the thrilling plot reaches its climax, a brilliant young mathematician disguised as a shoeshine boy, gives stock tips to the VC. He explains that the fundamental laws of logic, such as the law of the excluded middle (Tertium non datur - flashback to the gravestone) has doomed the concept of middleware to failure from the very start. It is only a matter of time, he predicts, before Acme Tertium Middleware Inc., kicks up the daisys...
Well, what do you think? Dreadful B movie material? How many marks out of ten do I get for sheer awfulness? Can you do better?
All flippancy and flights of fancy aside, the VC in my mythical Golden Turkey B Movie would not be the first to have nightmares about the middleware market. After all, one of the rueful definitions of middleware is 'the software that nobody wants to pay for'. Given that middleware is everywhere and all around us, this is an interesting state of affairs. Personally, I think that middleware's biggest problem is its name.
Let me explain by analogy. Movie plot number two. Picture the scene. A homesteader sets up a farm in wide open prairie. Harvests are good and the family prospers. Pretty soon more homesteaders come along and before long, a little town - Foo Town - comes into existence. Foo Town is essentially one street, known as Main Street. Everybody thinks of it as being the 'middle' of town. All traffic and commerce ultimately takes place on Main Street. All transport into Foo Town starts and ends on Main Street.
Time passes and prosperity brings with it more growth. Gradually, Main Street proves too limiting for the social and commercial structures that Foo Town needs. New streets just as big as Main Street are formed. After a while, Foo Town no longer has an easily identifiable 'middle' really. It has distinct areas that are each self-sufficient. More prosperity brings a bigger transportation infrastructure. Transportation Termini spring up at many different locations in Foo Town - not just on Main Street. What used to be the Main Street Shuttle Bus is now known as the Foo Town Metro System.
In a growing town, middles are necessarily transient structures. They serve a purpose in the early stages of growth. As growth continues 'middles' must metamorphose into 'metros'. Otherwise, growth is unsustainable. Now which do you want for your growing town? Middleware or Metroware?
See what I mean?