Defect Cost Increase

Kent Beck

2004-12-27 16:58:19 UTC

June,

I apologize that the graphs aren't clear. They are so clear to me while I'm
drawing them that I sometimes forget that everyone doesn't share my
perspective :-)

The y axis is intended to represent the number of defects. The solid bars
are defect removal activities.

You're right that for a complete picture it would be necessary to integrate
the cost of fixing defects so as to show the economic superiority of
frequent testing. I can't think of how to do that graphically. Do you have
any ideas?

Kent Beck
Three Rivers Institute

-----Original Message-----
From: June Kim [mailto:juneaftn-***@public.gmane.org]
Sent: Thursday, December 16, 2004 4:38 PM
To: xpbookdiscussiongroup-***@public.gmane.org
Subject: [xpe2e] Defect Cost Increase

Hello.

On page 99. in XPE2E, there's an explanation on DCI and two graphs
describing it.

What is the y-axis of the graphs? I guess it is the number of defects.
The x-axis is time. So the thin line means the total number of defects
in the system over the time and the bar the number of found and fixed
defects during testing. Am I right on this?

The problem is that the diagrams are a bit confusing w.r.t the
context. On the page, the book is explaining what DCI is and then the
diagrams show up. I thought, at fisrt, the y-axis had been about the
cost of defect (finding and fixing) since we are talking about d-C-i,
but the curve has to be exponential rather than linear. "What am I
missing?"

If there were another line overlapped in those two diagrams, depicting
the exponential increase of the cost, wouldn't it be better to understand?

Yahoo! Groups Links

------------------------ Yahoo! Groups Sponsor --------------------~-->
Make a clean sweep of pop-up ads. Yahoo! Companion Toolbar.
Now with Pop-Up Blocker. Get it for free!
http://us.click.yahoo.com/L5YrjA/eSIIAA/yQLSAA/nhFolB/TM
--------------------------------------------------------------------~->