RE: Prototypes, process, and ID (RE:[interactionarchitects] Re: process ... the nextgeneration)

30 Oct 2003 - 4:40am
10 years ago
2 replies
736 reads
livlab
2003

: "Data indicate that 60-80% of the cost of software
: development is rework--that is, fixing defects that
: are found during testing.* While software must still
: be tested, testing and rework costs would be reduced
: if better design and implementation practices were used"

Fixing defects that are found during testing... if there really was testing
then it's already a significant and important design practice. As John said,
the software industry is used to testing, it's part of the process. But not
all Web practitioners consider testing vital and, thought appalling, some
projects cut-out testing before other things in defense of 'cost saving'.
The most counter-productive decision possible in my opinion...

There is also a significant difference between rework because of test
results and rework because of mistakes after implementation. To test and
modify/improve is iteration; to release a product and create patches to fix
bugs is bad design.

Livia Labate
liv at livlab.com | http://livlab.com | http://aifia.org

Comments

30 Oct 2003 - 11:20am
vutpakdi
2003

--- Livia Labate <liv at livlab.com> wrote:
> Fixing defects that are found during testing... if there really was
> testing
> then it's already a significant and important design practice. As John
> said,
> the software industry is used to testing, it's part of the process. But
> not
> all Web practitioners consider testing vital and, thought appalling, some
> projects cut-out testing before other things in defense of 'cost saving'.
> The most counter-productive decision possible in my opinion...

But, quite a few producers of more traditional software applications
consider testing to be an afterthought or a drag on the process of
shipping. Even in some organizations where there is a strong tradition of
testing, there are pendulum swings from where testing is considered
important to where testing is considered an impediment to shipping "on
time."

In my company, the current fad is to allow some "early adopter" products to
ship without being "certified" (ie, not extensively tested).

On the bright side, we're still doing better than a to be unamed
compensation management software company who seemed to consider testing to
be a "well, it ran once one the developer's data set so we can ship it"
proposition.

Ron

=====
============================================================================
Ron Vutpakdi
vutpakdi at acm.org

__________________________________
Do you Yahoo!?
Exclusive Video Premiere - Britney Spears
http://launch.yahoo.com/promos/britneyspears/

30 Oct 2003 - 8:30pm
John O'Donovan
2004

----- Original Message -----
From: "Livia Labate" <liv at livlab.com>

> some projects cut-out testing before other things in defense of 'cost
saving'.
> The most counter-productive decision possible in my opinion...

I'm afraid I have worked on far too many projects where the testing period
is actually considered a buffer overflow period for development...more last
minute extra development and bug fixing is done than testing of a stable
system.

Iterative development is designed to address the need to have long final
testing periods, but it has to be enforced rigorously to work well.

Cheers,

jod

Syndicate content Get the feed