I'll throw my 2 cents in agreement with Golias. :) Which is actually quite
a bit more than the .02 cents Golias was throwing in in the first place. :)
Darkfox Reven'tsol
> Let me say that I agree with you 100 percent on the testing bit. They
don't
> test in a competent manner, but at the same time they claim to.
> However, I want to say something about the perception that they are
ignoring the
> large issues in favor of the small ones.
>
> I'm currently working on a fairly large data warehousing project, with a
fairly
> complex ETL (extraction/transform/load) structure.
>
> I have a list of changes that need to be made.
> The system is in production.
> The data used will F up the financials if I screw up.
>
> What do I do? What would YOU do? I make the small safe changes first.
Tweak
> a field value, change a calc. No structure changes, no multi tier
changes.
>
> Things as a whole get slightly better, my list is shorter. I still have
large
> issues.
>
> How do you solve large issues? What's the solution to issue X? Who does
that
> solution hurt unintentionally? How long does it take to get a workable
> solution?
> Can you even get one with no downsides? I've almost NEVER solved a
complex
> issue with no drawbacks in a complex system.
>
> Now, I have a team working with me.
>
> What do we do? Some folks make MANY small changes fairly fast.
> Some folks work long hours for months on EACH large issue and fix them
slowly.
>
> Both of these things happen at the same time.
>
> I know most of us are programmers/data specialists/IS folks, etc. And as
mad as
> we are, you know that in the real world this is how things work.
>
> Add to this that we have NO idea what the code ramifications to change
are. For
> all I know, the same module that controls spell X also produces cascade
data for
> mob behavior Y. With a zillion spells and a zillion mobs, making ANY
structural
> or rate change can be HUGELY time consuming, even to track the effects,
much
> less implement the change.
>
> I think Verant has dropped the ball in several cases. Are they a bunch of
> Keystone Kops with no clue? I doubt it.
>
> Complain away about specific issues, that's the only way to get them
fixed.
>
> But nobody on this list really knows enough about the code or dynamics to
call
> Verant incompetent.
>
> Just my .02 cents, having been there...
>
> Golias the Staunch!