XXXXXXXXXXXXXXXXXXX XXXXXXXXXXXXXXXXXX  XXX             XXX
  XXX                  XXX             XXX XXX             XXX
  XXX                  XXXXXXXXXXXXXXXXXX  XXX             XXX
   XXXXXXXXXXXXXXXXXXX XXX                  XXXXXXXXXXXXXXXXX 

CPU is a moderated forum dedicated to sharing information among workers in the computer industry.

CONTENTS
1. ABOUT BOX

2. /*COMMENTS*/

3. BOOK REVIEW: THE DECLINE AND FALL OF THE AMERICAN PROGRAMMER

4. FEATURE: WORKS COUNCILS IN AUSTRIA

5. FEATURE: INTERNATIONAL NETWORK OF WOMEN IN TECHNOLOGY

6. BOOK REVIEW: THE 6TH ANNUAL COMPUTER INDUSTRY ALMANAC

7. TOOLBOX: AMATEUR COMPUTERIST

8. LABOR BYTES: MISCELLANEA

9. EOF: "THE MERCENARY'S 10 COMMANDMENTS"

Online subscriptions to CPU are available at no cost by e-mailing listserv @ cpsr.org , with a single line in the message:

SUBSCRIBE CPSR-CPU

For example:

SUBSCRIBE CPSR-CPU Tom Mooney

CPU can also be found via anonymous FTP from cpsr.org in /cpsr/work. Hard copy subscriptions are available for $30 per year.

The moderators may be contacted by voice at (510) 601-6740, by email to stack@starnine.com, or by U.S. mail at:

P.O. Box 3181
Oakland, CA 94609

Post freely, especially at work. CPU material may be reprinted for non-profit purposes as long as the source is cited. We welcome submissions and commentary.

CPU is a project of the "Working in the Computer Industry" working group of Computer Professionals for Social Responsibility / berkeley Chapter.

Editors for this issue: Michael Stack and Jim Davis. We may be contacted by voice at (510) 601-6740, by email to stack@starnine.com, or by USPS at PO Box 3181, Oakland, CA 94609.

The views expressed herein do not necessarily represent those of the editors or CPSR. On the other hand, some do.

Welcome to our Labor Day 1993 issue of CPU. Some news updates during our long absence:

-- Since the last issue, we have mention of CPU in the _San Jose Mercury News_, and a nice write-up in the _San Francisco Examiner_, with pictures and everything.

-- At last count we had 978 subscribers. Although it is difficult to tell exactly who our subscribers are, we did receive a note from an Austrian engineer saying he knows of 10 CPU subscribers there.

-- Next issue we will print a talk given by Lenny Siegel of the Pacific Studies Center (and publisher of _Global Electronics_) on current trends in the electronics industry.

We'd like to hear from you, what's happening at your company or in your work situation (employed or otherwise), what the issues facing you and your fellow workers,

jd

_The Decline and Fall of the American Programmer_
Edward Yourdon
Prentice Hall, 1992
352 pages, $ 24.95 (hardbound)

SUMMARY

In his 1992 book "The Decline and Fall of the American Programmer", Edward Yourdon exhibits multiple personalities: There's Yourdon as Geraldo, Yourdon as Computer Science Pop Professor, Yourdon does National Geographic and Yourdon the bibliophile.

The first chapter is all afternoon TV talk-show style. This is not to say that his proposition -- that "The American programmer is about to share the fate of the dodo bird" -- is without substance, it's more his supportive evidence that stinks.

Thereafter follows the meat of the book; a ranging discussion of current software development technologies. He offers no definite conclusions, nor has he a particular technology preference, since he agrees with Fred Brooks that there is no "Silver Bullet" to slay monster software projects of missed schedules, blown budgets and flawed end-product. That is:

"There is no single development, in either [software] technology or in management technique, that by itself promises even one order-of-magnitude improvement in productivity, in reliability, in simplicity." From "No Silver Bullet," IEEE Computer, 1987, by Fred Brooks, quoted on p. 23.

The long appendix describing the Indian Software Industry is the best thing in the book (interestingly providing ammunition to undo arguments made in chapter one) and Yourdon's bibliography has expansive commentary on what he considers the quintessential tomes of programming practise.

Caper Jones, an author and consultant cited often by Yourdon, in an April ComputerWorld interview said "The Decline and Fall of the American Programmer" described the "worst case scenario".

THE BOOK

Yourdon begins:

"By the end of the decade, I foresee massive unemployment among the ranks of American programmers, systems analysts, and software engineers. Not because fifth generation computers will eliminate the need for programming, or because users will begin writing their own programs. No, the reason will be far simpler: International competition will put American programmers out of work, just as Japanese competition put American automobile workers out of work in the 1970s" (p. 1).

Beneath this opening paragraph, there is a drawing of a big old dodo bearing the caption "The American programmer, circa 1999". Yourdon argues such will be the case because:

1. "American" programmers earn five to ten times more than their foreign counterparts.

2. "American" programmers are relatively unproductive.

3. "Americans" do not write high-quality software.

Point one is undeniable and with each passing quarter, off run the Sun's and EDS's to enjoy the advantage (programmers _are_ dodo's if they are not conscious of this fact, once quipped an associate). Foreign programmers and systems analysts speak English, are highly-educated and can electronically transport their software products anywhere in the world, says Yourdon [see CPU 001 -- Ed].

But his justification for claims of low-quality and low- productivity are specious. He writes: "More important than the claim that India-based software is 30 percent cheaper than American software is the likelihood that it has 10 times fewer bugs and can be maintained 10 times more easily" (p. 16) though previously he admits "nobody (to my knowledge) has yet attempted any large-scale surveys of software quality on a national basis" (p. 7 ). He leans heavily on self-observation of a particular cast: "...I see so much evidence of sloppy, low-quality work all around me in my day-to-day life -- in all fields, not just in software development" and then he dodges providing other supporting argument by suggesting the reader look up "horror stories" in journals of the extreme (Peter Neumann's software bug documentary, Jon Jacky's critique of probably the largest software project to date -- SDI -- or having the reader contact CPSR). Regarding productivity, "... I have not been impressed with the energy level of the average programmer in the vast majority of DP [Data Processing] shops I've visited in the United States. Most of them have a difficult time remaining in an upright position all day. I'm convinced that many organizations play muzak to hide the sound of snoring" (p. 6) -- and this just after citing his consulting comrade, Capers Jones, claiming the average American programmer puts in 50-hour work weeks.

This early section is the low-point of the book. It's Camille Paglia controversy. An academic woman attacks feminism voicing the views of conservatives. She gets heavy media coverage. Yourdon courts management. Those pot-bellied, birkenstock programmers with the too-much hair (but fading from where it used to sprout) not only look sloppy -- they really are ... and expensive too! Irv Wendel, a consultant in Oakland, California writes: "You blame U.S. programmers for rotten systems, crap code, and not having a decent DP education. Bullshit. Blame the people who control the situation -- management." (p. 11).

The rest of the chapter is taken up with a mishmash of views from a Yourdon Compuserve forum calling for commentary on his dodo thesis. Points of note are that competition seems to be coming from projects overseen and funded by foreign governments, whether it is the Indian government approaching a forum-contributing consultant to ask his opinion on how best to set up an establishment capable of training N x 100,000 young, educated people in modern software engineering techniques or Yourdon's citing nationally-funded programs: England's Alvey Directorate, Europe's Esprit program, and Japan's SIGMA project "among many others" (p. 10).

Interesting comments are made by Leon Levy of Bell Labs:

"Our present methods of developing software were developed two decades ago when machines were expensive and programmer's weren't... With the changing relative cost of machines vs. programmers there will be increasing pressure to mechanize." (p. 11).

Levy thinks the above-mentioned Fred Brooks quote of there being no Silver Bullet doesn't tell the pertinent story. He says there will always be difficult software issues but "[d]evelopment of the transaction type applications that are the bread-and-butter of the application programmers of today will be sufficiently automated that only very few applications programmers will be needed to supply the demand." (p.11).

Yourdon concludes Chapter One with what he sees as the means of keeping ahead of the software Golden Horde flooding from Asia, the Indian sub-continent and Ireland. Buddha Yourdon will have the software corporation achieve Nirvana -- or as he terms it, a state of "world-class software organization". (In his chapter on Software Processes he spends time on the Software Engineering Institute's five steps toward the Ultimate). Before launching on enterprise rescue (for "... the efforts of a productive few are often overwhelmed in companies where 25 percent of large system development projects are canceled before completion." (p. 17)), he has a parting comment for the individual programmer and system analyst, They must vote with their feet to follow those software enterprises that are subscribing to his advice ("You can change jobs freely" p. 18).

With no single Silver Bullet, chapter 2 begins by asserting that what's necessary is multiple software techniques applied simultaneously. Yourdon outlines the software technology veins that he is to trace in coming sections, also explaining why he devotes whole chapters to certain topics while leaving out others that many might think of import. For instance, he quickly dismisses the topic of computer languages. He believes that an interest in the language used in the writing of programs will soon be regarded an anarchronism: "Programming languages will still be a matter of some academic interest, but the world-class software organization will greet the announcement of yet another new and sexy language... with a loud yawn." (p. 27). The language used will be hidden from view by CASE tools. Object-oriented Programming also gets short shrift. He is not interested in object-oriented "programming" per-se; discussion of the type "Is Smalltalk better than C++?" or "What's the best way to implement a scroll bar in Microsoft Windows?". He believes this low-level of discussion will have little effect on productivity and quality. He advises focussing attention on object-oriented analysis, object-oriented design and object-oriented databases. "If you don't understand the user's requirements, it doesn't matter how you code it." (p. 35).

Now Yourdon is in his stride. He's talking about what he knows and it's informative reading (especially me being one of the DP technique-ignorants referred to earlier in the book). He devotes a chapter each to the importance of hiring and properly managing the right people, software processes, software methodologies, CASE tools, software metrics, quality, reusability, software re- engineering and future trends. This makes up the bulk of the book.

In the chapter on "Peopleware" (titling it so allows him to treat people as though they were a technology in line with the headings of his other chapters) he notes: "Attention to peopleware issues can literally cause 10-fold productivity improvements, while investments in CASE, methodologies, or other technologies rarely cause more than a 30-40 percent improvement." (p. 28) He also quotes DeMarco and Lister saying "Most managers are willing to concede the idea that they've got more people worries than technical worries. But they seldom manage that way." (p. 28). Employees seemingly are where most gains can be made and at the same time are the most unpredictable of the "technologies" listed. He describes a test given by Sackman, Erickson, and Grant back in 1968 to 12 experienced programmers. They found wide variance with the best person in the group finishing coding 28 times faster than the worst person, and the best program was approximately 10 times more efficient (in terms of memory and CPU cycles). There was no correlation with years of programming experience or scores on standard programming aptitude tests. By contrast, for programming teams, Capers Jones reports that development and maintenance costs of projects using experienced people were half that of projects using inexperienced people. Yourdon spends time on how proper management is key to unlocking "peopleware" productivity, with this point reinforced by more solicited Compuserve views. He next moves to a discussion of software processes dwelling on the Software Engineering Institute's five levels of maturity for a software development organization. These are:

-- The initial level (Ad hoc, Little formalization of processes with software tools informally applied).

-- The repeatable level (A stable process with a repeatable level of statistical control).

-- The defined level (Achieved foundation for major and continuing progress).

-- The managed level (Substantial quality improvements over previous levels, comprehensive software process measurement).

-- The optimizing level (Major quality and quantity improvements).

The SEI holds that it takes months and usually years to go between levels, and levels cannot be skipped (It would not be thought unusual that it would take 10 years to go between levels 1 and 5). A combination of organizational shake-up and software technologies applied differentiates levels. Hardly any companies are operating at levels 4 or 5 on an enterprise-wide basis according to SEI survey, though some internal projects may be (companies measured by SEI were such as Hughes Aircraft, GTE, IBM, Ford Aerospace, McDonnell Douglas). Yourdon would have the "world-class software organization" operating at a level 5. This whole topic is meant for organizations of a 100+ or 1,000+ engineers. This discussion came across as a trifle unreal in an industry where multiple generations of software products, methods and companies might come and go in a decade. Yourdon quotes Tom DeMarco as saying, "according to the SEI model, Apple Computer should not exist".

The chapter on Software Methodologies spends time on the nature of scientific revolutions as per Thomas Kuhn (he's read that book). Figure 5.12 has a simple graphing of major methodological software transitions from "structured techniques" to "information engineering" to "object-oriented." But Yourdon writes that "the typical software engineer needs 1 to 2 years to become familiar and comfortable with a methodology. An obvious consequence is that organizations are unlikely to change their methodology at the drop of a hat; having chosen one, they will typically stick with it for 5 to 10 years before choosing a new one. It also suggests that the typical software engineer is unlikely to become proficient in multiple methodologies; learning multiple methodologies is more complex than learning multiple programming languages." (p. 107).

This is also a problem when choosing CASE tools. Methodologies implemented are behind the current state of the art -- even decades behind, observes Yourdon. This will be the case until the "CASE tools themselves become the driving force for methodology creation and evolution." (p. 33). Other issues are that it's expensive kitting-out each engineer with their own CASE development environment -- but prices are plummeting. CASE has a steep learning curve with a corresponding fall-off in productivity after the first months of usage ( "At the end of 1990, IBM's AD/Cycle, the ultimate CASE environment for carrying out a software engineering methodology, already had 24 manuals! If the trend continues, it will be necessary to start training our future software engineers when they are in kindergarten if they are to ever finish their training!" (p. 106)) But Yourdon banks heavily on CASE as the way of the future radically changing the way software is developed. He quotes Carma McClure:

Pre-CASE:                      Post-CASE:
-------------------------------------------------------------- 
Emphasis on coding, testing    Emphasis on analysis, design. 
Paper models                   Rapid prototyping. 
Manual coding                  Automated code generation. 
Manual documentation           Automatic documentation. 
Testing of code                Testing of specs and design 
Maintaining code               Maintaining specs and design.

This chapter also has a practical list of questions to ask of your CASE vendor. "But in this field, as in so many other walks of life, everyone lies; it's just a question of when they lie, where they lie and how much they lie." (p. 136).

Software Metrics discusses methods of measuring productivity for "You can't improve if you don't know the current situation". The chapter on Software Reusability outlines what seems the most straight-forward of the Yourdon-cited "technologies" often realizing substantial pay-backs. He suggests setting up software part's-shops with their own set of engineers working on nothing but producing code for reuse. Having been around a while, Yourdon notes this is a fashion that alternatively thrives and wanes. He recommends offering incentives -- financial etc. -- to encourage engineers in thinking in terms of reuse. He suggests old applications as a resource worth mining for tested, reusable code and devotes a chapter to maintenance of programs written years ago in languages forgotten by people long gone. Mainly its a sympathetic lament promising no magic fix. This is relevant Yourdon writes, because many organizations spend more than half of their MIS resources keeping old software alive.

The last chapter is on future trends, and he concludes that no one can predict what will happen in the future.

CONCLUSION

I read this book to see what's in it for the everyday Joe Blow programmer ("Software Engineer", "System's Analyst" etc.). I learned from the chapters on software technologies (e.g., CASE tools are going to change everything but they are not here just yet). But this is a book busy saving the "American" software enterprise (This smacks of the "Buy American" campaign in the face of the latest Honda Accord made of 80 percent American parts assembled with American labor. Profits accrue to multinationals based in "American" cities rather than those overseas). Yourdon would have companies listen to Robert Reich and invest in training and tools for Joe Blow back home to make an increased productivity out-weigh the local higher costs [CPU would like to hear about companies spending on their employees -- Ed]. Simultaneously, he would have them set-up liaisons with such as the described Indian software shops in Appendix A to take advantage of cheaper labor (Yourdon provides a long listing of Indian contacts).

But in this Yourdon's book is out of touch. Companies are opting NOT to invest in their "peopleware". Layoffs, out-sourcing and the tendency toward contracting programming labor are what we are getting instead. CASE tools are a large capital investment and you can hire multiple cheap Indian, Irish, FOC's (Fresh Out of College) or whatever programmers for the same price -- and not be stuck with an obsolete shelf-full of dated manuals and software six months later (The relative high-cost of CASE is why little is done with this technology in India, Yourdon writes). Have labor make the capital investment themselves either by self-financing education and then getting hired or have some foreign government subsidize a programmer acquiring current skills. That Yourdon discusses little these tendencies is a major omission.

ms

1) BACKGROUND

All general regulations concerning work (working duration per week [37,5 - 40 hours][*2], working environment concerning health & safety, length of holidays [5 weeks minimum], social insurance [fixed by law], working restrictions for pregnant women, ...) are defined by law. In addition to that there are two further possible agreements:

• a) Collective agreements between unions and representatives of employers. Essential contents are the definition of lowest levels of income, improvements of the working environment with respect to law. These agreements exist separately for every trade. Almost every trade has its collective agreement.

• b) Agreements between the works council of a single company and representatives of this company. We will call that "company agreements". Areas which may be covered by "company agreements" are defined by laws.

Both collective agreements and "company agreements" are not restricted to union members; i.e. both are valid for all employees of one trade with respect to one company and have a status similar to law. The collective agreement may only contain improvements for the employees with regard to laws and such; the "company agreement" may only contain improvements to the collective agreement. Any written drawback is invalid -- that's the law. So we have a hierarchy of regulations: laws, collective agreements and "company agreements". In addition to that individuals may have a special contract with extra benefits.

2) WORKS COUNCIL

Every company with five or more employees is obliged by law to have a works council. The works council has a certain number of members depending on the number of employees; council members do their committee work as part of the paid working time. If the company is big enough, one or more council members may work exclusively for the interests of the employees. All members of the work council are permanent (they cannot be fired). And, council members need no be unions members if they don't like. [*3]

• a) Elections. The works council is voted upon in a secret election every four years. Of course, several parties may apply for the election.

• b) Rights and Duties of the Works Council.
  • Representation of employees in all affairs.
  • Participation in "company agreements". There are some

areas where company representatives may be forced on the basis of law to agree and sign such agreements (e.g. such as we had in our company on the definition of flexible working hours), and some areas where works council and company representatives agree on a "good will" (e.g. such as we had on the refund of costs for glasses and contact lenses).

• The council is responsible to check whether laws and collective treatments are respected.
• The company is not allowed to change the "status quo" in some aspects, unless the work council agrees. E.g. the company intends to install a automatic time supervision of telephone control. In such cases the work council shall agree only if the collected data is not be used against employees in any case.
• The members have the right to do council work as part of their job and have three extra paid weeks in four years for further education.
• The company has to provide room(s), law books, telephone, etc. for the committee.
• The company has to inform the council about the commercial situation of the company (balance sheet) and about other important issues (new strategic plannings, new and leaving employees, etc. etc.) These rights sound very comfortable at the first glance but in many companies it is hard to get them; with a works council you have the best means to enforce your rights. Only if your company has a works council is it possible to have "company agreements" and to oppose effectively negative developments in the company (e.g. notice of termination of employment)

(Engagierte Computer ExpertInnen, Austria)

FOOTNOTES
[*1] Regulations in other West European countries are similar as far as we know.
[*2] In [brackets] some numbers for Austria.
[*3] In fact, not all companies in Austria, even with some hundred employees have a work council. Often no one dares to do the first step because s/he is afraid cause of the company representatives. The unions only act on behalf of the law to enforce a work council, if someone has blamed the company for having no council.

The recent Department of Labor "Glass Ceiling Report" concluded women are not successfully pushing beyond mid-management boundaries - only 3% difference in the last 3 years. It is because we are less capable, less educated, less effective? Of course not. The Clarence Thomas hearings left many people with the mistaken impression that there are only two kinds of women - men lovers and men haters. Is there anyone out there representing your voice? Take a look at WIT.

The International Network of Women in Technology (WIT) represents a centered alternative. We are a grass roots organization representing women working in technology companies who prefer to define ourselves, individually through our consciousness and actions. WIT is dedicated to improving the status of women in technology and helping women advance to the highest levels of responsibility possible.

We don't want to be promoted (or held back) because we are women, but want to be accepted on our own merit. Rather than adopting a "victim" mentality. We chose to empower ourselves - rather than asking for depending on, or expecting someone to make our lives better, we choose to take responsibility and do it ourselves.

{We take the obstacles we, ourselves place in our own paths as seriously as the obstacles imposed by others. We find both intolerable.}

We are committed to growth and self-improvement - women who care about each other and want to make the world a better place. We consider ourselves principled, effective and capable of assuming great leadership roles. We believe we can contribute to solving world problems and making things better in very small and very big ways. Many of us aren't "joiners," but see the value of significantly expanding our network.

{Who we know and how fast we can get critical information significantly contributes to our empowerment!} All of us have our own networks. WIT will help us multiply those networks efficiently to leverage off and access valuable information. Involving women working in a variety of technologies and backgrounds will directly contribute to our individual and collective power base.

All WIT members will be part of an electronic mail network (under the WIT alias) which will empower each of us by allowing us to broadcast and receive information to and from members all over the world in a time critical manner. This network will be invaluable and profile successful women. All WIT members will have access to the WIT job hotline dedicated to help women get ahead. Our Membership Directory will be an outstanding source of contacts.

If you wish to be included in our network, Directory and have access to our job hotline, we welcome your support, friendship, and membership. Please mail the attached Membership application with your check to:

        International Network of Women in Technology
        4641 Burnet Avenue
        Sherman Oaks, CA 91403

For more info, contact:

Carolyn Leighton, Criterion Research, 818-990-1987,
MCI ID: 439-6466

Dr. Carolyn Turbyfill, Sun Microsystems Inc., President Northern California Council, Carolyn.Turbyfill@Eng.sun.com, 415-336-1007

Kathleen Bernard, North Carolina Supercomputer Center, President Washington, D.C. area Council, kcb@ncsc.org

_The 6th Annual Computer Industry Almanac_
Karen Petska Juliussen and Egil Juliussen
Computer Industry Almanac Inc. 1993
$45 (paper) $55 (hardcover)
Available directly from the publisher (Computer Industry Almanac, Inc. 225 Allen Way, Incline Village, NV 89451-9608, 800-377-6810.

Finding good data on the fast-changing computer industry is no easy task. Government industry classifications are hopelessly dated, and government data is often a few years old by the time the general public gets a hold of it. Other sources -- the various industry organizations and publications that collect data -- are scattered, often hard and/or expensive to obtain. For the high- tech job seeker, finding potential employers or getting useful salary data is an equally daunting task. So _The 6th Annual Computer Industry Almanac_ is quite a treasure.

The almanac is over 800 pages of recent data about the computer industry. Compiling research from over 90 sources, the almanac include chapters on computer companies, products and technologies, education, employment and salaries, industry forecasts, and more. There is a wealth of data in this book that will be useful for the computer industry worker.

For the job-seeker, the almanac includes an directory of 2,909 computer companies, including address, telephone and fax numbers, number of employees, annual revenue, product focus, type of ownership, and the company president's name. (This data is also available on disk for $200). Unfortunately, these companies are not indexed by product focus, or by location. This would be a very useful addition to future editions. However, other listings of the leaders in respective industries can give the job-seeker a headstart in at least identifying the leaders in a respective field. The almanac includes company rankings from various computer and other business publications, including _Datamation_, _Electronic Business_, _Forbes_, _Fortune_, _Soft*Letter_, and more.

For help in negotiating salaries, or determining the wage scale in your area, the almanac includes the results of several different industry salary surveys, including ones from _Computerworld_, _Datamation_, Source Edp, Positive Support Review and Source Engineering. In case you were interested, the median salary for a software designer/developer with 4-6 years experience in the San Francisco/San Jose area is $54,100 (that's from the Source Engineering survey). In Portland, the median salary for the same is $43,700. The national median is $48,000. Unfortunately (and this isn't the fault of the almanac, but the original surveys themselves), the surveys don't have fine enough job descriptions. For example, how do salaries compare for work on different platforms (e.g., Mac vs. Windows vs. UNIX), or for specialty areas or languages?

For career planning, the almanac includes an employment trends section, with perspectives on what skills are in high demand now, or likely to be down the road.

The almanac also includes the top stock options, the top 150 industry executives in terms of salaries and compensation, the Fortune billionaires that are connected with the computer and telecommunications industry, etc. etc. Curious how much your company is making off of you? There's a listing of profits per employee (in the "Computers and Communications: Major Systems" category, AST is first, at $28,300, Apple is listed second at $22,500). At Microsoft, profits per employee is $83,900. The largest profits-per-employee ratio of any of the computer companies is cisco Systems (networking-related products), at $154,700.

There are a few holes in their collection; or rather, things-that- would-be-useful-additions. Things I'd like to see: a survey of how many contract and temporary employees companies, and pay differentials; an international comparison of salaries; how much offshore development work is done by companies; inclusion of information on the non-engineering/non-management positions (e.g., clerical, manufacturing and custodial positions); information on the status of women in the industry, an analysis of various nationalities and ethnic groups in the industry (the Pacific Studies Center, Mountain View, has done some interesting research in this area); and the inclusion of electronic publications in the publications listing.

Still, _The Computer Industry Almanac_ is the best single source for computer industry information. It comprises the most up-to- date collection of information on the industry that exists. It's certainly a book to encourage your public library or company library to get, if you can't afford your own copy. We'll especially find it useful here at CPU in future months.

jd

CPU recommends _AMATEUR COMPUTERIST_. Its not your usual net- fare, carrying considered articles of substance (There's history even!). Here follows a quote from their self-description: "The newsletter is dedicated to support for grassroots efforts and movements like the 'computers for the people movement' that gave birth to the personal computer in the 1970s and 1980s. Hard efforts of many people over hundreds of years led to the production of a working computer in the 1940s and then a personal computer that people could afford in the 1970s. This history has been serialized in several issues of the newsletter .... Most recently the newsletter has begun an online edition that is available free. We are beginning to document the progressive impact of democratic developments like Usenet News and the Internet..."

Their most recent issue (Winter/Spring '93) has a good framing of the Internet's open and free development, something the hovering commercializing sharks waiting on the sidelines are about to squash and meter. There are articles on the "Social Forces behind Usenet", an interview with Henry Spencer, one of the early participants, authors and pioneers of Usenet News discussing News origins and the writing of the C news program, an obituary of John G. Kemeny, co-inventor of the computer language BASIC and of the Dartmouth Time Sharing System (DTSS) and advocate of universal education in programming, and another on the tradition of May 1, 1848 -- how the first 10 hour bill law in Britain caused capital consternation.

_AC_ was described by Andrew Ross and Constance Penley in their book _Technoculture_ (Univ. of Minnesota Press, 1991, p. 125):

"When worker education classes in computer programming were discontinued by management at the Ford Rouge Plant in Dearborn, Michigan [February 1987], United Auto Workers members began to publish a newsletter called the Amateur Computerist to fill the gap. Among the columnists and correspondents in the magazine have been veterans of the Flint sit-down strikes who see a clear historical continuity between the problem of labor organization in the thirties and the problem of automation and de-skilling today. Workers' computer literacy is seen as essential not only to the demystification of the computer and the re-skilling of workers, but also to labor's capacity to intervene in decisions about new technologies that might result in shorter hours and thus in `work efficiency' rather than worker efficiency."

Try the _AC_ news-group 'alt.amateur-comp' for a sample or to obtain a copy, send E-mail to:

au329@cleveland.freenet.edu

or:

ae547@yfn.ysu.edu

ms

EMPLOYMENT TRENDS IN ELECTRONICS INDUSTRY: "Domestic employment in the U.S. electronics industry fell for the fourth consecutive year in 1992. December, 1992 electronics employment was 2,291M or 99,000 (4.1%) less than the 2.39M reported for December, 1991.

"The only industry segment that experienced growth in 1992 was Prepackaged Software, with a modest 2,270 new jobs. On the other hand, Defense/Commercial Guidance Systems lost 30,000 jobs last year. With one exception, U.S. electronics employment showed no month-to-month growth for 30 consecutive months. Since August, 1989, our industry has lost 309,000 jobs. And, when the industry's healthy software segment is removed from the total, domestic electronics employment dropped by more than 380,000 in the same period." -- American Electronics Association, as quoted in the _1993 Computer Industry Almanac_

[Also, pay in most areas dipped slightly over the previous area, according to a _Computerworld_ survey. The AEA reported recently that employment in the industry rose by 12,000 in June. "'Although this represents some modest improvement in our industry's employment level, we still have concerns about the long-term prospects,' said AEA Chairman Arnold N. Silverman." (_SF Examiner_, 8/26/93)]

LAYOFF MONITOR:
As expected, IBM announced large layoffs at the end of July, this time it was 35,000 workers, and announced closings of some unspecified factories. Reports the _New York Times_ (7/28/93): "[IBM chairman louis V.] Gerstner acknowledged the 'unbelievable burden' on the IBM work force in recent years and how employee morale had suffered as the payroll was reduced by more than 100,000 people... Yesterday's announcement of further cuts will do little to revive morale."
FUJITSU, the leading Japanese mainframe manufacturer, is cutting 6,000 workers (11%) over the next two years. (_Wall Street Journal_ 7/15/93)...
TANDEM
Tandem will cut 15% of its employees over the next year, that's 1,600 to 1,800 people; they also instituted a permanent 5% cut in pay for all employees (WSJ 7/29/93)...
DIGITAL EQUIPMENT
Digital Equipment is closing a 600- worker disk-drive factory in Germany (NYT 8/20/93)...
Disk drive maker QUANTUM announced in August that it was laying off 7% of its workforce or 147 people (_SF Examiner_ 8/3/93)...
EVEREX SYSTEMS, in Chapter 11, is cutting 125 manufacturing workers (_San Jose Mercury News_ 8/27/93).
MEAD DATA CENTRAL (provider of Nexis and Lexis databases) is "streamlining" -- 400 workers to be dumped (NYT, 8/20/93)...
In the SPINNAKER/SOFTKEY/WORDSTAR merger, 30 jobs will be lost (_SF Examiner_ 8/18/93). In telecommunications, AT&T announced it would close 40 operator centers (WSJ, 8/31/93), phasing out 4,000 operator jobs; SPRINT is cutting 1,000 jobs in its long distance unit (_SF Examiner_ 8/27/93); AMERITECH announced it would cut 1,200 to 1,500 management positions (NYT 8/21/93).

FIRED FOR INTERNET POSTINGS?: Gregory Steshenko, a systems support engineer at MICROSOFT, claims that he was fired in June for messages he posted on the Internet. He asserts that outspoken postings about the Ukraine prompted complaints directed at his employer. Microsoft initially asked him to stop posting from work, which he did, but Microsoft continued to receive complaints, so the company fired him on June 29, citing "lack of discretion, judgement and professionalism." Microsoft's employee handbook evidently proscribes offensive statements on electronic bulletin boards. Steshenko considers this a free speech matter, and undue control over his personal, "extracurricular" activities. He is considering suing Microsoft. (This story adapted from the _SJ Mercury News_, 7/22/93).

STRIKE AGAINST DEC IN GERMANY: For the first time in history, the most powerful German Union "IG Metall" used strike measures (lasting two weeks) against a computer company. The German DEC company had not signed a collective agreement which has been in place for the DEC subsidiary Digital-Kienzle (among others).

Digital, currently heavily restructuring their German subsidiaries, signed the contract; as a result, the weekly working time is now 36 hours per week.

[Please note, that what I call a "collective agreement" is an agreement between representatives of a Union (e.g. Steel Union) and representatives of ALL companies whose employees are represented by that Union. So all steel workers in Germany, Austria have the SAME collective agreement. They may also have certain company agreements (only improvements) on a per company basis. -from Engagierte Computer ExpertInnen, Austria]

CRACKDOWN ON IMPORTED CONTRACT PROGRAMMING? Federal and state authorities are moving against companies that hire contract programmers from other countries and pay them less than the prevailing wage, according to the _San Jose Mercury News_ (9/6/93). Placement firms take advantage of provisions in immigration laws to bring programmers to the U.S. for extended work stints, of up to six years. Most companies deny paying less than the prevailing wage. "But one [agency], asking not to be identified, said that wages, especially for independent contract programmers, have gotten too high and that the international temporaries offer some relief," reports the _SJMN_.

The U.S. Department of Labor recently fined a Michigan placement firm $180,000 for failing to meet several requirements for hiring offshore workers, including failure to pay prevailing wages to 41 software engineers. The company is appealing the decision, and disputes the Department's prevailing wage figures. The Department of Labor asserts that temporary computer programmers in the San Jose area should be paid anywhere from $35,734 for a junior programmer/analyst with no experience, to $85,000 for a programmer with 10 years experience. The Michigan firm, COMPLETE BUSINESS SOLUTIONS, believes those figures are too high (though at least some are in line with the _Computer Industry Almanac_'s surveys).

In addition, the California Employment Development Department (EDD) has fined six international contract programming agencies a total of about $1 million in back taxes on their contractor's wages. The state says it is trying to bring the industry back into "compliance."

According to HEWLETT-PACKARD figures, H-P pays about $60,000 to an agency for a programmer on a temporary visa. The temp engineer's pay varies. California's EDD says the prevailing wage is from $32,700 to $44,520 a year, depending on job title (note the discrepancy between their figures and the Department of Labor's figures). One firm says it pays temp engineers about $3,000 a month when taxes are included, but according to several recent lawsuits by some Indian programmers, they have complained that they have only been getting $1,400 to $1,800 a month.

Kick Them While They're Down Department

Concerning recent announcements of a layoff of 2,500 employees by Apple, Jay Leno recently quipped they were achieving their longtime dream of becoming the same size as IBM. Hi - oooooh!

=-=-=-=-

Q: What is the difference between Jurassic Park and IBM?
A: One is a high-tech theme park for dinosaurs and the other is a
   movie by Steven Spielberg. Ba-bing!


Kick Them Even If They're Not Down Department

Q: What is the difference between Jurassic Park and Microsoft?
A: One is a high-tech theme park dominated by expensive, nasty,
   hungry, predatory monsters that will destroy anything they can 
   get their teeth into... and the other is a movie by Stephen 
   Spielberg.

[from "Bits and Bytes Online". Bits and Bytes Online is a weekly electronic newsletter. Email Subscriptions are available at no cost from slakmaster@aol.com or jmachado@pacs.pha.pa.us. Put "SUBSCRIBE" in the subject header.]

Workplace Index	CPSR home page	E-mail webmaster

Web pages edited by Dave Williams
Last Updated: Thur 4 April 1996 10:20 PST