This is a test of the use of LibSyn PDF upload/download capability.
I decided to use one of my old articles from a magazine which is no longer published, but should be.
Computer Language Magazine was a good magazine (and I not saying that just because they published some of my articles.)
I chose this particular article because its sort of relevant to what I'm going through on my day job.
People are still struggling to adequately describe on-line, interactive systems for design, implementation and testing.
Rovira diagrams are one means of describing a system's behavior, using dialog syntax diagrams, from the screen "in", without getting bogged down in the actual implementation details. (It just focuses on the system's behavior.)
There are two main reasons the technique is called Rovira Diagrams:
- Larry O'Brien, the editor of that magazine, was a distant friend,
- the abbreviation of dialog syntax diagrams (DSD) was identical to Data Structure Diagrams and Larry felt it would be confusing, so they're named after "moi" (There are lots of techniques named after their formulators: Bachman Diagrams etc. :-).
Back in 1983, I was given the task by my boss at the time, (a hospital IS administrator in Montreal, Quebec Canada,) of finding out how a Hospital Care System system had been implemented.
It was a mess as only IBM could have delivered it. It consisted of pieces of PL/I, Assembler, CICS macroes, some Cobol, as well as something that was IBM's answer to the Holy Grail, their Application Development System, a.k.a. ADS. (IBM later gave up on that particular, or should I say peculiar, version of their ADS. Later still, they tried using Smalltalk and then Java as their ADS. The search continues...*)
To complete the dog's breakfast, it was an online system which the hospital relied upon for their customer care application. It absolutely needed to be documented.
I was the, and I quote, "genius they kept locked in a back room", in another part of the hospital, (occasionally humorously referred to as the "Psych Ward," [it may very well have once been]) and I would occasionally get these "interesting" problems tossed my way.
Given my previous experience at computer languages, (I excelled at formal languages and automata and I had created a recursive descent Pascal compiler for a Keronics Point/Four machine,) I came up with a syntax to describe the interactive dialogs between human and machine.
Because of the mess of the actual implementation and of the functionality of 327x terminals, where light pens could be used as well as cursor position and single character input could be used to trigger some functionality, a feature of the syntax was that it DIDN'T rely on triggers, instead relegating them to something unimportant for the overall functioning and description of the system/application.
A few years later, for the January 1990 issue of Computer Language Magazine, I was asked by Larry O'Brien of "Computer Language Magazine" to come up with a visual representation of the dialog syntax language that I had created.
That gave rise to something he called "Rovira Diagrams."
A few years later, in January 1994, Medical Devices & Diagnostic Industry magazine ran an article by Ken Niehoff on "Using Rovira Diagrams To Specify the User Interface" (I believe they still have reprints they can send you by fax.)
Basically, it repeated my article and stated that I had come up with a good idea, with an example of how they were using it themselves.
*) And it will likely continue, since language (even computer languages) are perennially evolving things.