Response to draft of Chapter 9

greenspun.com : LUSENET : HumptyDumptyY2K : One Thread

August 17, 1999

1658 hours CDT

Dear Ed:

Three thoughts came to mind as I read your draft of Chapter 9. First, I am reminded of a story attributed to Tom Peters about a software programmer in Silicon Valley who said something to the effect of "our software would be fine if it weren't for those blankety blank end users". End users are often (not always, but often) unqualified to use the software that they allegely must have to do their work.

There are three reasons for this situation. First, organizations view training an an evil to be tolerated so long as money remains in the project budget after the inevitable cost overruns on hardware and software have been absorbed. If there is anything to be cut from a project, end user training is the place to start.

Second, there is a dearth of truly qualified trainers and/or outstanding training materials (I am speaking here of off-the-shelf software). Training is an art, not a science, and training materials are often oriented toward learning by rote. Learning in this manner only works if the employee does not have to think outside of the box (ie, perform a task that the training manual does not cover). Documentation is usually written at a reading level (a discussion for another forum!) and an experience level that is beyond the typical end user's time or abilities. As a former Education Center manager for a Valcom franchise from 1984-1988, I encountered this problem on a daily basis, and since becoming an independent consultant (and sometimes trainer), I have found that the problem has becom worse. Finally, even if training is made available to end users and it is reasonable effective (ie., it does what is needed to enable the end user to do his/her job with a minimum of outside support), the training is scheduled for times that either cut into an employee's non-work time (evenings and weekends) or cut into their work day, leaving the employee with a backlog of work as a "reward" for going to training.

Second, the notion of licensing software programmers reminds me of my days as a hearing aid audiologist and what I have heard about the CPA examination. In both cases, those who were already licensed appeared to be acting as gatekeepers (ie., keeping the great unwashed masses out of their sandbox). Whenever you have industry "insiders" regulating the industry (ie., the insurance industry), you have an industry regulated by the self-interests of the insiders. that is why Congress turned to the computer industry to help them draft the y2k liabilty legislation. Free market advocates like Milton Friedman would have a fit with licensing of this type (another governemnt intrusion into the private sector); their belief is that enlightened self interest will eventually pervale in a market-based economy. Y2k will show the flaws in a market-based and -driven economy; to quote the late Pogo, "we have met the enemy, and it is us!"

Finally, the real problem with software (again, I am referring to off-the-shelf PC software) is that it is marketing-driven, not need driven. If companies truly examined their needs and based their purchases of all products and services (even those not related to technology) based on their needs, much of the software and a majority of the upgrades would not be purchased. Marketing forces compel software vendors to make new releases, and those same marketing forces determine th timing of those new releases. As we may see with Windows 2000, end users purchasing software are really paying for the privilege of advanced beta or release candidate testing for the software vendor. If software vendors (and even developers) were driven by enlightened self interest, they would realize that their approach to commercial development needs to change if they want to avoid the eventual backlash from end users that Y2k might usher in.

My apologies for being so long-winded. Thanks for listening.

Sincerely,

Oliver

-- Oliver K. Burrows III (oburrows@dwave.net), August 17, 1999

Answers

Oliver,

Interesting points ... I guess my basic response is: would you expect to see society tolerate the same state affairs in the field of medicine? Or in the design, construction, and testing of commercial aircraft? If not, why should we tolerate the level of unprofessional behavior (in terms of training and everything else) in the software field?

Ed

-- Ed Yourdon (HumptyDumptyY2K@yourdon.com), August 21, 1999.


(1) I must agree about the dearth of qualified data processing trainers. When I ran a training team and needed to staff up, I had more difficulty in finding adequate people than when staffing for any other edp position (such as programmers, analysts, computer security, EDP Audit, etc.).

(2) Ed makes the assumption that there will be an objective inquiry by Congress (or whoever) into who is to blame. I would probably give no more than 50-50 odds that such an inquiry will be objective. His points about who are actually to blame are right on.

(3) Re: Licensing/Certification. You might also want to add the International Systems Audit Control Association (Formerly the EDP Auditors' Association). They test, evaluate experience, and certify EDP Auditors as Certified Information Systems Auditors (CISAs). At least they did the last time I had contact with them. They also require continuing education and continuing experience for recertification. I think that they can be reached at their web page.

(4) As far as standards within corporations, I have felt like a voice crying in the wilderness about software configuration management. Like, please guys, don't change my production software without telling me...and maybe even giving me some user documentation!

-- Mad Monk (madmonk@hawaiian.net), August 17, 1999.


Moderation questions? read the FAQ