M7 Discussion

Legal status of software engineering Capers Jones Poware Productivity Research

s the term “software engineering” a misnomer? That question has long been debated within the computer science, programming, and software engineering

community. Naysayers point to the software activity’s large trial-and-error component and its notable lack of solid intellectual and ethical underpinnings. On the affir- mative side, ACM and the IEEE Computer Society recently joined forces to move software engineering toward pro- fessional status.

Barry Boehm, TRW professor of software engineering at USC and an ACM member of the IEEE CS/ACM task force, summarized some new-and serious-implications of the debate in the September 1994 Software Engineering Technical Council Newsletter.

Currently, software engineering is not one of the 36 engineering professions recognized and licensed in the United States. This situ- ation is more serious than you might think, because 48 states have laws on their books that prohibit anyone

T ennessee now actively prohibits the use of “software

who is not licensed from using the term “engineer”

engineering” in business

in describing his occupa- tion and work.

As Boehm points out, these laws are beginning to be enforced. The state of

literature and advertising.

Texas has forced universities to stop offering master’s degrees in software engineering. Tennessee now actively prohibits the use of “software engineering” in business lit- erature and advertising. New Jersey considered, but did not pass, a regulation that would have required licensing of all software professionals employed within the state. (The fact that this regulation would essentially have shut down all software businesses and forced them to leave the state was eventually recognized.)

This legal phenomenon affects all of us involved in soft- ware. It makes us vulnerable to the probability of increas- ingly onerous rules and regulations passed by well-meaning-but clumsy and often ill-informed-leg- islative bodies. Consider, for example, the hazardous situ- ation now faced by computer and software consultants. As a result of legislation and regulations to determine who is or is not an employee, many software consultants have come close to losing their ability to practice independently.

Computer

What makes an engineering profession? Let’s consider some of the attributes of the recognized

engineering professions, such as electrical engineering, mechanical engineering, and civil engineering. What do they have that software engineering lacks? Also, what characteristics do the nonengineering professions, such as medicine and law, have that make them true professions instead of mere occupations?

In the broadest sense, the factors associated with rec- ognized engineering professions and other formal profes- sions include

l a well-defined body of knowledge, and often many sub- sets of more specialized knowledge;

l academic curricula that transfer the body of knowledge to students well enough so that a significant percentage can pass qualifying examinations;

l qualifying examinations that certify at least minimal competence for general practice of the profession;

l a formal set of subspecialties, each with a substantial body of knowledge and some form of certification, often created by accreditation boards within each specialty;

l continuing education for those within the profession to maintain currency in the overall profession and their chosen subspecialty;

l a code of ethical responsibilities for those engaged in the profession and its specialties;

l strong professional associations capable of creating qual- ifying examinations and certifying specialties in con- junction with state and other governmental agencies;

l a recognized canon of standard practices for common conditions, against which claims of professional mal- practice can be evaluated;

l methods for monitoring and dealing with instances of professional malpractice, and a formal mechanism for decertifying those found guilty of professional mal- practice; and

l liability insurance coverage for professionals that pro- tects them and their employers from a portion of the financial consequences of losing law suits for profes- sional malpractice.

To a greater or lesser degree, the software community fails to meet any of these criteria. However, IEEE Computer Society and ACM task forces are exploring these topics with the idea of making software engineering the 37th engi- neering profession, perhaps by the end of the century.

Following the path of the medical profession A very instructive book, which I recommend to task

force members a n d a n y o n e else interested in creating a highly regarded profession, is Paul Starr’s The Social Transformation ofAmerican Medicine (Basic Books, N e w York, 1982).

It’s b e e n only 1 5 0 years or so since medical practice faced challenges similar to those software faces today: It was a n amorphous a n d fragmented community with many questionable a n d unproven practices, a n d academic training s p a n n e d every possibility from state of the art to totally inept. In fact, it was just over 1 0 0 years a g o that Johns Hopkins University took the unprecedented step of requiring medical students to have college degrees as a precondition for admission.

In every drive toward professionalism, there are com- peting forces a n d p o w e r groups with vested interests. At the very minimum, there are universities, professional associations, legislative bodies, the profession’s knowl- e d g e workers, the knowledge workers’ employers, a n d of course the clients the knowledge workers serve. To this basic set, w e can also a d d insurance companies a n d attor- neys w h o specialize in providing services to the profes- sion’s members a n d enterprises.

Each participant has its o w n vested interests, so com- petition a n d politics are the order of the day. However, for any occupation to succeed to the same d e g r e e as medical practice, a strong professional association must emerge to s h a p e events.

In particular, the profes- sional association must have a strong voice in establishing basic acade- mic curricula (individual universities are too chaotic a n d d o n ’t grasp the big pic- ture); in establishing licensing or accreditation criteria (legislatures will botch it up), a n d in moni- toring a n d eliminating malpractice (otherwise, liability insurance will b e unobtainable). A related but more subtle topic is the role of professional associ- ations in minimizing com-

T he professional association must have a strong voice in establishing basic academic curricula, in establishing licensing or accreditation criteria, and in monitoring and eliminating malpractice.

petition from nonmembers w h o lack accreditation. Right now, software engineering is in about the same

condition as medical practice was in the 1890s. Software is already a n important topic, but the practice of creating software is undisciplined a n d often “unprofessional” in any serious use of the term. Academic training is spotty, a n d the thought of qualifying examinations and/or licens- ing remains highly unpalatable.

If the software engineering community cannot rise to the level of becoming a recognized profession a n d engi- neering discipline, w e face a n uncertain future with ever- mounting prospects of unfriendly legislation a n d harmful government actions.

A Guided Tour of Multimedia Systems

and Applications edited by Borko Furht and Milan Milenkovic

Looks into the goal of fully integrating both audio a n d video so that they can b e created, edited, a n d combined with other data types to form compound objects. This book explores the techniques, standards, a n d applications behind today’s advances in multimedia systems. The five papers in the tirst chapter provide a n overview of the basic principles involved in multimedia technology a n d applications. The next chapter consists of seven papers o n multimedia compression techniques a n d standards. The next three chapters include 2 4 papers covering multimedia communication a n d synchronization, storage a n d retrieval, a n d tools a n d applications.

400 pages. Mad 1995. Sofme~: ISBN O-81 86 Tow 1. Catalog # BP07014 - hlenzbels $40.00 /List $5’0.00

SOCIETY

Now Available on IEEE Computer Society On-Line

l This month in Computer: Article Summaries, Binary Critic, Hot Topics, Letters to the Editor, Software Challenges, a n d Table of Contents (a n e w option off the main menu) l Abstracts a n d tables of contents of Computer Society publications (weeks before publication) l Conference calendar l Calls for papers l Career opportunities l Volunteer directory l General membership a n d subscription information l Author guidelines a n d copyright forms l Computer Society Press Catalog l Staff contact list l Senior/staff manager list l 1 9 9 4 IEEE fellows

The server is available with a g o p h e r client at info.com- puter.org or a W client at http://www. computer.org. For more detailed information, send questions o n e-mail to on-line.access@ computer.org.