Is TPF Ready For A CASE Tool? - Part 1
by Jeff Robinson

It's no secret... TPF is losing the battle of popular transaction operating systems. From all appearances, IBM has all but pronounced it dead and has even joined forces with other interested parties to develop an ORACLE-based alternative to TPF. On the home front, TPF/ALCS shops are shifting loyalties as fast as you can say "client-server". New application development efforts are being curtailed while an emphasis on maintenance and platform transitions are highlighted. At least the TPF contractors are happy; as one headhunter told a friend of mine: "TPF is jumping with (contract) opportunities right now".

Our Own Worst Enemy?
So is it all over for TPF? Should TPFers wise-up and just play dead? Although that's always an option, I believe that there might be a glimmer of hope for TPF and I think that we in the TPF arena can start fighting back. But let's don't get confused as to whom we're fighting. It's not IBM - they've rarely, if ever, pushed TPF to the heights most TPFers have felt it can reach. Disagree? Then just take a moment and think of the most recent innovations in TPF - TPFDF, CMS/TPF, etc. - and you can agree that most of them were not begun by Big Blue. Furthermore, the enemy is not our employers and their new corporate madness for the GUI (Graphical User Interface) look and feel. They have a fiduciary obligation to keep their companies and their products competitive and marketable.

In fact, we (the TPF user community) may be our own worst enemy. The past ten years have seen rapid technological and methodological advances in the data processing industry; but how many of those advances have crept into TPF? A whole lot of us still don't know what Object-Oriented Design/Programming (OOD and OOP) is nor ever care to find out. Worst still, how many of us would be allowed to exercise such new approaches on a project-level development effort with the full support of co-workers and management? Unfortunately, a lot of TPFers are still struggling to accept the now almost commonplace techniques of Structured Analysis and Design.

TPF Needs A Paradigm Shift
I don't think it's too late for TPF to accommodate a few of those technological advances. And by taking in some of those advances, someone, somewhere may decide that TPF can be one of the operating systems of the future. Look at it this way: either it (or should I say "we"?) accommodate change or TPF is dead. But in order to accommodate, a lot of us will have to start "thinking" differently about our work, our jobs, and our careers. This kind of change scares a lot of people - not just those in TPF. But the past has shown that change usually and eventually produces a win-win (everyone benefits) situation. All it takes is that initial first step.

What I would like to propose as a step toward TPF's paradigm shift is the development of a CASE tool for the TPF environment. What is a CASE tool? Simply put, it's a computer-based tool which allows programmers to control the development cycle of their projects. CASE is an acronym for Computer-Aided Software Engineering. A CASE tool can be used to ease the analysis and design of a project and in most optimum situations it will even generate your code for you. You may have heard of CASE before and you may have even used a CASE tool at some time in your career. But before you pass judgment on this CASE idea, at least give this proposal a fresh, unbiased look.

TPF/CASE: A First Generation CASE Tool
Being a TPF programmer myself, I know that the TPF community will NOT just accept any old CASE tool for their environment. Why? Because almost every TPFer I know has that innate need to know just how a piece of code works, why it works the way it does, and what made the idiot who coded it use a CLI rather than a TM (humor intended).

That's why I'm proposing what I call a first generation CASE tool: a tool that will definitely aid the TPFer in the analysis and design stage but only give him as much help as he will allow during the coding stage. This means that a programmer can generate simple program shells or in-depth, fully-coded modules. Naturally, because the CASE tool will include graphical based tools and a user-friendly, easy-to-use environment, it will be hosted on the PC in a Microsoft Windows environment. This widely used environment will allow the programmer to generate and maintain accurate diagrams and visual representations of his projects.

Sounds exciting? I hope so because I want this to be a tool that you, the TPFer can use. And in order to ensure that you will want to use it, I am soliciting suggestions and ideas from you as to what you would like to see in this CASE tool. Here are the tools that I think may be handy in this CASE venture:

Over the coming months, I will continue to address development and finalization of this tool to the readers of ACPTPF Today. As development progresses, the tool will be made available to the public for use in their own project development efforts. All suggestions and comments are welcomed - especially if you think this is not a suitable approach toward the paradigm shift.

Please feel free to contact me via the newsletter, by way of the ACPTPF Today Development Forum on the WEB, or via the Internet at: You can email ACPTPF Today on the Internet at: You can reach the ACPTPF Today Forums at: http:/

Utilities Update

We want to thank those readers who requested the free TPF Code Tracker utility which was offered in the November '95 issue. We also want to let you know that we're working on the bugs you reported, along with making general improvements in the application for the next version release. When the next version is complete, we will reissue the utility to those of you who would like an upgrade.

IBM, TPF, and TPFDF are trademarks of IBM Corporation
CMS/TPF is a trademark of TPF Software, Inc.
Microsoft is a registered trademark and Windows are trademarks of Microsoft Corporation
ORACLE is a trademark of the Oracle Corporation.