RE: Programming tools (was DDS Support)

-----Original Message-----
From: Rob Dixon [mailto:rob.dixon@erros.co.uk]
Sent: Tuesday, July 11, 2000 12:25 PM
To: MIDRANGE-L@midrange.com
Subject: Re: Programming tools (was DDS Support)

This thread has raised many interesting points but I have picked Karen's as it reminds us that it is our clients whom we have to please - we are in the service industry.  It can be amusing discussing the technical merits of different tools and methods but in the end we are in business to deliver infinitely flexible, robust, efficient, integrated, highly functional applications that can be delivered by more than one means - e.g. 5250 and HTML and within reasonable time and cost scales.  As Karen says " my clients are moving fast".  Current tools cannot keep up.

Many of the tools, on whatever platform, don't substantially alter the way we work - they just tinker with inefficient processes with little benefit for the users  If we could improve our productivity for development and maintenance by, say, 10 times or more, rather than soon being out of work as we ate up the application development backlog, we would be flooded with work as users would produce their lists that have been siting in their desk drawers for some long time.  With such an increase in productivity, cost justification for development would be so much easier.  But if to achieve these benefits, we have to change our methods of working, will anyone be brave enough to have a go?  Maybe a few will try the new tool, but will they really persevere to understand how to use it and  what it can do, or, at the first hurdle (there are bound to be some) will they stumble and return to their old bad ways even if these cannot satisfy user needs?  I fear that they will.  They have spent years learning and honing their existing skills and they will generally not be prepared to spend enough time trying the new ways to fully understand them.

So who will prove me wrong?

My company has a  AS/400 product, ERROS, that that can deliver productivity benefits well in excess of 10 times.  By storing data and application definitions as a business model in the same Neural Database in which the user data is stored , and by using the definitions to drive an ERROS "kernel"  program, new, totally integrated, applications with stunning response times can be created without any program generation (so it is NOT a case tool) and largely without any program coding.  I believe that ERROS eliminates about 90% of coding.  As ERROS generates HTML on the fly, applications can be used instantly over the Inter/Intranet (and ERROS can serve images).  No physical file design is required, no logical files are used (yet any item can be indexed 1,000,000,000 **2 ways) and the database structure and applications can be changed whilst the system is live.  The development and operational environments together are under 50 AS/400 objects and creation of new applications rarely changes this number (i.e. no new files are created!).

So, you are asking for new tools and here is one. Anyone want to try it at 2 in the morning?  For free?

Rob Dixon 

(snip)