Tuesday, 28 July 2009

A professional malpractice after 1995

Reading through Assessment and Control of Software Risks by Capers Jones from 1994 is giving me a few chuckles

"The US software industry is now reaching the end of the start-up phase where consumers will buy almost anything because it is new, different and exciting. In the next mature phase the software industry will be dealing with sophisticated and educated clients who are going to demand high quality, low costs and full support after delivery. "

"Improving customer support, improving HELP function, and improving user documentation are also beneficial, but not easy to accomplish. However, software products such as some Windows applications that require 20 megabytes of hard-disk storage and 8 megabytes of memory to execute are forging new kinds of dissatisfaction"

There was some learning for me as well - I learnt that not only do Brits and Americans differ on how to spell humour, what football is and how to cook a steak but also on how to count Function Points.
A little Googling of this and I found a site that told me all about IPFUG, Mk II Function Points and a new kid on the block, Boeing and their 3D function points.

The site also told me that the first three letters in function points are FUN.
"People who enjoy function point counting and can justify it on that basis should do so"
People who enjoy function point counting need to get out more

But then onto the serious stuff and this quote from the book

"Usage of innacurate metrics is the most serious risk of all, since mistakes derived from this problem can slow productivity and quality progress to a standstill. The usage of inaccurate and paradoxical metrics such as "lines of code" has been a major obstacle to software engineering since the industry began.

This problem is so common and so severe that until it can be overcome, it is unlikely that the phrase "software engineering" will be anything other than an oxymoron. A strong case can be made that the usage of "lines of code" metric should be declared to be a professional malpractice after 1995"

Strong words.
And what has been one of the current buzzes in the blogsphere at the moment ?
Coding Horror and his latest post - Software Engineering: Dead? after reading the latest article from Tom DeMarco Software Engineering:An Idea Whose Time Has Come and Gone? with the quote
"Do I still believe that metrics are a must for any successful software development effort? My answers are no, no, and no."

Matthew Heusser has been done a 3 part series about Metrics, schmetrics, Michael Bolton also recently blogged about Three Kinds of Measurement and Two Ways to Use Them ( and also noticed the DeMarco article ) and even Linda Wilkinson was thinking about a Metric of the Month

So 15 years on from the Capers Jones book the metric argument still rages.

But at least we've solved the problem of high quality applications for sophisticated clients and no longer have Windows applications that require 20 megabytes of hard-disk storage and 8 megabytes of memory to execute...

1 comment:

Rob Lambert said...

Nice post. I reckon in 20 years time that will still be relevant.