…Moving them to another machine, also, caused the programs to behave
as expected…

>> Does this mean they worked on the second machine, or not?

…I have taken the traditional position that it is the responsibility
of the developer to provide programs which work in our environment.

>> Windows is a complex and finicky environment and it is rare to
develop on the deployment server. Even Unix development is often done
using cheaper, user-friendlier, Windows tools (i.e., Java development)
and the code moved/migrated to the deployment server.  Also, an
application may be deployed on more than one server.

Unless you have specifically negotiated ongoing support, with a
contractor, the line has to be drawn somewhere.  Is it fair to keep
the contractor/vendor on the hook to make it work every time you
deploy to another server or upgrade the system? You should negotiate a
"user acceptance test" with the contractor. Vendors usually demand it
so that they can get final closure.  A mutually agreeable test plan is
developed that can include testing on the deployment server (if it is
a new system- generally, testing should not be done on production
systems).  Once you sign off (accept the application) on the test
plan, when you are satisfied everything works, responsibility ends
unless you have contracted for deployment. You can negotiate a
warranty period, in case something was missed in the acceptance
testing, but this will add to the cost.  There is no free lunch.

…Is it fair to expect that if one calls herself a VB/C++/Delphi
developer that they are experts in Windows' internals and networking?…

>> No. This type of person may exist but is rare.  The environments
are just too complex to be on top of everything all the time. Even
Microsoft has different certifications for different types of skills.
Application developers used to just need to know language syntax to
write the code.  Today, applications are usually "assembled" from
pre-defined classes and components with only unique logic actually
being coded.  Developers need to be familiar with hundreds, even
thousands of API calls, objects, and components.  Not a lot of room
for other things like network configuration, security, etc.

…Does future development really have to be done by a committee of
security, O/S, network and programming experts?…

>> Not necessarily, but, for complex applications/environments, it's
recommended.  In today's multi-tier application development, it's not
unusual to separate business logic from data access logic to leverage
expertise in different tools and technology. A VB expert may not be an
expert SQL coder and, if unique database features are to be leveraged,
for things like performance, an expert in the specific database may be
desirable. Welcome to the New Age!



Chip Dorman <[log in to unmask]> wrote in message news:<[log in to unmask]>...
> Hi y'all...
>
> I have a situation with a Windows developer for which I am seeking
> advice.  The general issue is that  in programming contractor
> relationships, who is actually responsible for getting the developed
> program to work in our environment?  Before you jump to any immediate
> conclusion allow me to explain.
>
> We contracted a set of programs for a Windows 2000 server.  This was not
> an hourly contract rather a contract for a specified solution. Once the
> programs were installed, we have had nothing but troubles with them.
> Well, when the programs were demonstrated they worked fine on the
> demonstration W2k machine but not so on our W2k machine.  Moving them
> to another machine, also, caused the programs to behave as expected.
>
> Being an old fart in these matters, I have taken the traditional
> position that it is the responsibility of the developer to provide
> programs which work in our environment.  This meaning to me that it
> is the developers responsibility to debug the problem.  Of course, the
> longer I think about the traditional position the more I wonder about it.
>
> Since this is a server based program there are potentially several
> extra sources for problems such as network and server configurations.
> Plus there is _always_ some sort of Windows issue be it COM+ or MTS or...
> If you've done any Windows development, you are aware of how ...uh...
> "challenging" the environment can be.
>
> So my friends, in this new-fangled age of development where DOES the
> responsibilities of each party begin and end.  Is it fair to expect that
> if one calls herself a VB/C++/Delphi developer that they are experts
> in Windows' internals and networking?  Does future development really
> have to be done by a committee of security, O/S, network and
> programming experts?
>
> I am really looking for opinion and actual experiences from both sides
> of the programming contractor relationship.
>
>
> TIA,
> Chip Dorman
> Kennametal, Inc.
>
> * To join/leave the list, search archives, change list settings, *
> * etc., please visit http://raven.utc.edu/archives/hp3000-l.html *

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