On Mon, 28 Dec 2009 23:33:00 -0500, "Diane Poremsky [MVP]"
wrote:

the "cost" in speed and resources is too high and the
benefit too low.

That is short-term management think, and the primary cause of poor
software quality. "It's too expensive to make an app as important as
a widely used e-mail client reliable"; this mentality has a much
greater long term cost, as users get fed up with the software and
switch to another vendor offering. I would agree that not every bug
is worth chasing down, but e-mail is critical for most users, and if
it is not rock solid (or at least degrade gracefully), users will
simply seek alternatives. Microsoft cannot afford much more lost
goodwill due to poor software quality, and if they continue to employ
management that relies more heavily on "business metrics" and pretty
charts and graphs to guide their direction, their reign will be short
lived. That is a cost that a software vendor cannot afford. Do it
right the first time and the health of the company will remain strong.
Google, Apple and others have been eroding MS marketshare in every
category for some time now, and it is up to MS to re-think the cost of
quality versus the cost of low quality, or the competitors will simply
keep coming until there's nothing left.

Crashes can send data back to Microsoft (if you allow) so
they can analyze and fix the problem but hangs don't generate doc watson
files.

If they felt the cost was too high to fix it the first time, why would
they invest time to fix it after the fact? It is a known software
engineering fact that the least expensive time to fix a bug is in
early development and testing, and the cost skyrockets exponentially
after the product has shipped or otherwise late in the lifecycle.

I've never received a followup from MS that said "hey we fixed your
bug, just grab the latest update". I agree they can fix the problem
but for reasons mentioned above they often don't.