In article <[log in to unmask]>, Mark Wonsil <wonsil@4M-
ENT.COM> writes
>Birket writes:
>> 0) You must know what the business plan is for the next 3-5
>> years and you must
>> know what applications support the business functions today.
>>   [Loop for each application]
>> 1) Does the application meet the needs of the business for
>
><snip />
>
>Trying not to add to the chicken-egg thread, I would quickly mention that I
>agree with Birket's steps even without the 11/14 announcement.  Every year a
>company should revise the business plan for the next 3-5 years and make sure
>the infrastructure will support the plan.  If the infrastructure is not
>there then you follow the steps further.  However, I cannot decide to move
>to MPE/iX running on IA-64 natively or emulated under Linux without checking
>if the tools are available to do that first.  So clearly, it is an iterative
>process and all of the steps listed by Birket are important, but you may
>have to revisit them depending on the tools available at the time.  I would
>also add that Duane Percox had some good thoughts when he suggested that one
>should look at your business processes aside from the current
>software/hardware.  In this way, one can migrate different parts of your
>applications at different times.  So in the first step you can replace
>application with "the user interface" or "the data store" and follow the
>same steps.
>
>FWIW,
>
>Mark Wonsil
>
I agree entirely with Mark's and |Birket's points as well as with Ray
Shahan; the analysis and planning is paramount but iteration and review
of the technology and tools as well as the business and its wider
environment is equally important. Additionally protecting investment,
achieving low or least cost solutions, and obtaining a good ROI must
also be part of the equation.

Focusing on the tools and technology is important because poor choice
can lead to vendor lock in. Also, with legacy systems evolution or
migration, continuing to promote the use of obsolescent or vendor-
centric tools, methodologies, languages, and platforms when alternative,
cost competitive open system, open software, vendor-neutral, and modern
re-engineering methodologies  are available neither protects investment
nor achieves long term least costs.

In my view, the demise of MPE presents opportunities for businesses to
take the initiative to re-engineer their legacy systems into fully
integrated designs enabling competitive advantage to be obtained from
the use of open system standards and platform neutral tools, services,
databases, and languages, as well as from component software and
evolving Web Services. This approach doesn't require wholesale
migration. Instead it allows specific components to be evolved or re-
engineered without disrupting the business processes and operations.

We don't talk about migrating lines of code; which achieves nothing in
terms of improving the application design or its integration potential.
Rather we focus on modules, subsystems, interfaces, data flows and data
stores, business logic, transaction management, etc. whilst at the same
time evaluating alternatives and possibilities, improvements and future
needs.


On our web site at http://90.0.0.3/adbcwebix/migration.htm we have a
detailed analysis and review which we believe presents a good balance
and which we believe will help to focus evolution, migration, or re-
engineering strategic planning.
--
Chris Thompson
The Internet Agency, UK
http://www.the-internet-agency.com
European Distributors for Advanced Networks Systems Inc.
Distributors of CCS TRAX and CCS C-iX 'C' compiler for MPE
MPE and RTE re-engineering and migration tools and services
IBM Development Partner
Voice:  +44 7836 364575
Fax:    +44 1202 418209
Email   [log in to unmask]

ANSI -
Advanced Network Systems Inc., USA
Specialist in JAVA and J2EE for MPE
http://www.advnetsys.com
Voice:  +1 908-638-3330
Fax:    +1 908-638-3331
Email   [log in to unmask]

-----------------------------------------

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *