|
|
Posted by Jerry Stuckle on 06/13/06 15:27
Peter Fox wrote:
> Following on from Jerry Stuckle's message. . .
>
>> Peter Fox wrote:
>>
>>> Following on from Jerry Stuckle's message. . .
>>>
>>>> Of course, the "best one" is one specifically written to meet your
>>>> friend's needs.
>>>>
>>> Umm. Not necessarily.
>>>
>>
>> Oh, definitely. A custom program is *always* the "best" - because it
>> can fit
>> the customer's needs *exactly* without any excess baggage. But it may
>> not
>> always be the most *practical*.
>>
> So it's late. It's full of bugs. It's not documented and supported on
> a poke 'n hope basis. The user's needs were neither understood nor
> articulated by the user and only 'sort-of' implemented by the
> programmer. Is that what you mean by "best"?
>
>
Not if it's planned well. It's on time, has few bugs and is documented. It's
called Project Management.
Over the years I've done dozens of such projects - ranging from one person to
teams of 20 or more programmers. Done right it's a success. But if it's not
managed properly, you will get the results you describe.
As I said - it's the *best* because it meets the customer's exact needs. But it
may not be the *most practical*. Spending $20K to $100K for an application when
there is a $200 commercial package on the market which fits the customer's needs
(and maybe has some bloat) is probably more practical.
But as a consultant, before I recommend something to a customer (whether
pre-packaged or custom), I first have to understand the customer's needs and how
he wants to use the software. Only then can I recommend the *most practical*
solution.
In this case it's for a machine shop. Well, that's a lot different need than
for a small retail clothing store. But it's still not enough info to recommend
the *most practical* solution to the problem.
--
==================
Remove the "x" from my email address
Jerry Stuckle
JDS Computer Training Corp.
jstucklex@attglobal.net
==================
Navigation:
[Reply to this message]
|