|Volume 1: No. 20|
Unisys had a bad quarter, and is planning to cut 10,000 jobs this year. HP is planning to trim just 2,000 jobs. [Computerworld, 7/29.]
Charles House is taking over as VP of Product Management and Development at Informix Software Inc. (Menlo Park, CA), after 29 years with HP. [SJ Mercury, 7/30.]
Donald Kennedy has resigned as president of Stanford, but will remain through the '91-92 school year. ONR claims that Stanford may have overbilled the government as much as $20M during the last ten years. Stanford's [accepted] practice was to bill for representative costs instead of detailed research expenses, but the choice of representative items was sometimes inappropriate. Faculty are divided over whether the President should be held accountable for accounting practices, and that controversy has led to the resignation.
Nearly all computer manufacturers are working on pen-based computing. Portia Isaacson, publisher of Pen-PC Report, says "There's never been a product in the industry that has so much techno-lust associated with it." Grid is selling a version now, but the real market is not expected to develop for several years. (The market is there, but current models are too heavy, too limited, too expensive, and too fragile.) Cursive handwriting recognition is expected to remain a bottleneck. Companies like IBM, Nestor (Providence, RI), and Communications Intelligence Corp. (Menlo Park, CA) are working on it. [Rory J. O'Connor, SJ Mercury, 7/28.]
The Software Publishers Association has released a study of the 1990 entertainment software industry in North America. Total sales were $355M, mostly at Christmas time. The Kuwait crisis boosted sales of simulation software, including flight simulators and war games, to 35% of the total; arcade/action games took 26% and role-playing games 22%. 73% of the software was for MS DOS, 10% Amiga, 8% Commodore, and much of the remaining 9% for Macintosh and Apple II. (Apple has failed to support entertainment developers, apparently fearing damage to the Mac's image as a business machine.) [Mike Markowitz, MicroTimes, 8/5.]
Electronic Arts sells entertainment software for all ages. In the U.S., its 1991 annual report touts software for 10-year- olds wanting to test reflexes, 21-year-olds managing simulated sports teams, and 35-year-olds test-piloting million-dollar aircraft. The company's best seller in Japan is Populous, a game where the player controls the fate of the world. The game's successor, PowerMonger, is a best seller in Europe. [David Needle, Computer Currents, 7/16.]
Bellcore has been changing under the guidance of George Heilmeier, its president for the last five months. The organization had been bogged down in committees, management by consensus, and attempts to please all seven of the local Bell operating companies. Heilmeier is asking scientists for priorities, advocacy, and accountability, all aimed at reduced bureaucracy, faster decision making, and higher quality. He also wants less physical science and more of the information sciences, especially OOP, multimedia, fuzzy logic, and databases. [Emily T. Smith, Business Week, 8/5.]
John S. May, the new president of AT&T Bell Labs, thinks along the same lines. He wants a profit-minded industrial lab, not a Nobel physics breeding ground. 90% of the Lab budget is already in development activities, but the $300M research section was reorganized last year to emphasize information science and other "practical" fields. It has become much harder to get long- term projects approved. [Peter Coy, Business Week, 8/5.]
Four Pentagon studies have shown Ada superior to C++ for military software projects. Ada is more mature, more standardized, has more vendors, and has richer development tools. (C++ is said to be about three years behind. It compiles faster and has better runtime efficiency, though, and has better support for object-oriented design.) TRW says that Ada coding is faster and with fewer errors, resulting in a 35% productivity and cost advantage, as well as a 70% maintenance advantage. The advantages may drop to 10% and 30% by 1994. [Gary H. Anthes, Computerworld, 7/29.] (A fifth study by the Naval Postgraduate School showed a 10-fold productivity advantage for fourth-generation languages.)
One of the hottest languages right now is Visual Basic, Microsoft's new $199 scripting language for Windows. "If I were buying one programming language today for a student, for quick prototyping in Windows, even for trying to cash in on the commercial fascination with Windows programs, I'd go for Visual Basic." [Phillip Robinson, SF Mercury, 8/4.] The language provides the same power and ease of use as HyperCard, but works at the "desktop" level so that you can invoke and control other programs. (In the Mac world, there's a new language called Frontier that serves much the same purpose.) Visual Basic uses the same syntax as Microsoft's other Basics, but provides menu- driven programming of icons, forms, dialog boxes, pull-down menus, controls, buttons, etc.
Smalltalk is making a comeback now that hardware is powerful enough to support it. ParcPlace Systems (Mountain View, CA) has been selling the original Smalltalk, which offers the strongest object-oriented support tools of any language. Digitalk Inc. (Los Angeles, CA) is selling a lower-cost Smalltalk for OS/2 and Windows, and both IBM and Microsoft have endorsed and licensed Smalltalk links to their environments. Several corporate MIS shops are now using Smalltalk for critical applications, especially client/server GUIs. [Jeff Moad, Datamation, 7/15.]
Sequent Computer Systems Inc. (Beaverton, OR) is offering the first object-oriented multiprocessing system. The Symmetry 2000 system combines Objectworks/Smalltalk from ParcPlace Systems with an OODB system from Versant Object Technology (Menlo Park, CA), at $3,500 each. (503) 578-9855. [EE Times, 7/15.]
HP's Visual Computing Department (Palo Alto, CA) needs a Ph.D. image-analysis researcher to work on scientific visualization in C/UNIX. Alexander Drukarev (firstname.lastname@example.org). [ba.jobs.offered, 8/1.] HP's Computer Peripherals Laboratory (Palo Alto, CA) also needs a PhD image-compression research scientist. Jeanne Wiseman (email@example.com). [ba.jobs.offered, 8/5.]
Lawrence Livermore National Laboratory (Livermore, CA) has been advertising for several computer scientists to work in image processing, simulation, control, and other real-time applications.
Dynacs Engineering (Palm Harbor, FL) needs a MS/PhD specialist in neural-network control and simulation. Citizenship required. Ravi Venugopal, (813) 784-4035. [ misc.jobs.offered, 8/1.]
Syracuse Center for Computational Science (Syracuse, NY) needs a PhD research scientist for their Fortran 90D MIMD compiler project. [Wennie Shu (firstname.lastname@example.org). Min-You Wu (email@example.com). misc.jobs.offered, 8/5.]
MRJ, Inc. (Oakton, VA) seeks citizens for development of parallel supercomputing applications, including large databases, text retrieval, image understanding, simulation, and optimization. Contact firstname.lastname@example.org. [Alan J. Broder (email@example.com). misc.jobs.offered, 8/7.]
The Lockheed Space Operations Company (Titusville, FL) has an AI opening in its Advanced Projects group to work with NASA Ames on scheduling tools for Space Shuttle repair and maintenance. Michael J. Deale (firstname.lastname@example.org). [misc.jobs.offered, 8/6.]
The Human Genome Mapping Center at UCSF seeks an MS/CS senior information specialist to build systems for molecular biology. Dr. Richard M. Myers, Fax (415) 476-8391. [Leslie Taylor (email@example.com). ba.jobs.offered, 8/1.]
Looking for a career? Kathryn and Ross Petras have just pub- lished Jobs '91 (Prentice Hall Press), a guide to salaries, entry requirements, and industry organizations in thousands of fields. The Petras also wrote Inside Track, discussed at length in V1 N11. I haven't seen Jobs '91, but I'll bet it beats the Commerce Department's Career Outlook Handbook.
Half of the U.S. workforce and a third of the GNP are in the small business sector. Indeed, 95% of all Bay Area businesses have fewer than 50 employees according to the San Francisco Economic Development Corporation. Yet lenders are wary of helping struggling small businesses, especially during a recession. (Your home equity may also drop, reducing the amount you can raise from that source.) About 90% of recent business failures are small companies killed by cutbacks at the larger companies they serve. 80% of small businesses fail in their first year, even during better economic times. Wells Fargo has begun courting small businesses, but most banks see more risk than profit in loans of less than $50,000. The SBA has raised the interest rate that banks can charge, but even SBA- guaranteed loans are hard to get until you have two or three years of profitable growth. The average size of a 1990 SBA loan in SF was $267,488. [Beatrice Motamedi, SF Chronicle, 5/13.]
Entrepreneur Magazine has just published 111 Businesses You Can Start for Under $10,000. I kind of fancy being a private investigator in San Francisco, like a professor I heard of who got tired of "watching life from a distance." A translation service would be good, but I'm not fluent in any of the languages I've studied. (BTW, an MBA with a second language can have her pick of jobs these days.) Other CS-related jobs are computer consulting, information brokering, free-lance writing, self- publishing, meeting planning, seminar promotion, and newsletter publishing. You can get how-to books at your public library. For a free brochure on self-publishing guides, write to Para Publishing, P.O. Box 4232-882, Santa Barbara, CA 93140. Or for about $70 you can buy any of Entrepreneur Magazine's 165 business guides -- (800) 421-2300, or (800) 352-7449 in California. These include desktop publishing, information brokering, consulting, newsletter publishing, software locator service, and mail-order business.
Educators and information professionals who must survive in industry sometimes take on commercial training roles. Chris Syed (firstname.lastname@example.org) has been head of customer and internal training for a major vendor of integrated library systems. Chris has tracked people who began as instructors in the last ten years. Two are now candidates for Ph.D.s in library and information science; three are training and project managers; two are with a CD-ROM vendor as sales rep. and regional sales manager; one is a journalist; and one became a public library branch manager. Two also worked for a major online-information company, and each of the others had at least one intermediate position. Other customer-support positions at this company led to careers as systems librarians at major libraries. Whether this is advancement or instability depends on your viewpoint, but Chris recommends the library-systems training field as a good way to gain contacts in the information industry. (Your mileage may vary. Watch out for contracts that restrict your right to work for former clients. Also, Chris notes that some of the major vendors are having financial problems -- and that client training involves a lot of nights in lonely hotel rooms.) [PACS-L, 7/24.]
Dan Dabney (email@example.com) points out that librarians work with longer time frames than do vendors, and hence learn more about system implementation and evolution. He thinks it's better to start as a librarian and then go to industry, if necessary. He warns, though, that it is no longer possible for a professional librarian to master the library trade; there is always more to learn. You must spend half your career keeping up and the other half working under pressure. [PACS-L, 8/1.]
Many women have entered engineering in the last decade. Entry salaries are slightly higher than for men, but median salaries are $6,000 lower -- $46.8K vs. $52.8K in 1990. Women also receive fewer than half as many promotions into management. The gap has been widening since 1984, and is due in large part to the relative youth of the female workforce -- 5 years median experience vs. 13.5 for all engineers. Authorities differ on other reasons for the slower promotion of women. [Margaret Ryan, EE Times, 7/17.]
Immigrant engineers were 3.6% of all 1985 U.S. engineers and 14% of new hires in computer/electronic firms, according to the NSF. Only 10% were managers, an under-representation that is again due partly to youth and inexperience. A Rutgers study shows that foreign-born engineers are often held back by cultural training. Many find it hard to promote their own work, and particularly hard to take a firm stand and argue with bosses. They may also be embarrassed by praise. Engineers with poor English or with foreign accents encounter additional obstacles. [Margaret Ryan, EE Times, 6/24.]
There are associations and services to help with such problems. Some are generic, such as Toastmasters or the IEEE Career Fairs. Others serve specific ethnic groups, such as the Silicon Valley Indian Professionals Association (Santa Clara) and the Soviet-American Scientists Division of Bnai Zion. A recent book to help people promote themselves is "Marketing Yourself: The Ultimate Job Seeker's Skill," by Dorothy Leeds. (Harper Collins, $19.95.)
I've heard it said that office politics is fierce at the universities. Marilyn Moats Kennedy discusses this (lumped with charities, hospitals, trade associations, government, and other nonprofits) in "Office Politics: Seizing Power, Wielding Clout," 1980. She says that "The process of office politics takes up much more time in nonprofits. It's used more destructively, if more creatively, and the rules are different."
Nonprofits don't pay well. Because they have a mission, and because there's never enough money to go around, they put the money into service rather than salaries. People who join understand the pay scale and are willing to trade salary for lofty ideals, easy work, and relative job security. (I know, the work doesn't seem easy -- especially for what's paid -- but it's not the ulcerous pressure of a law firm, advertising agency, major newspaper, or retail sales organization. The ulcers come from office politics, incompetent people, and an irrational system. Those who accumulate enough seniority can expect to plateau, or coast, without directly competing with energetic young idealists.)
People are rarely fired, even for very good cause. The more years you've "given up" to the cause, the less likely you will be fired -- unless your foibles make the evening news. Government workers are protected by civil service; professors by tenure; other jobs by custom. Downsizing and restructuring do happen, but nonprofit workers are more likely to be fired en masse than individually. A few large companies -- IBM in particular -- have tried to guarantee lifetime employment, but even IBM would demote or move an incompetent individual; nonprofits often would not.
Nonprofits are accustomed to getting bargains, and managers seek this in their personnel. They expect dedicated, motivated, competent workers who have chosen mission -- or academic environment -- over pay and rapid advancement. Instead, they typically get unmotivated people who have chosen security over good pay. This creates a difficult management problem, exacerbated by the inability to fire, transfer, or punish an unproductive worker. If appeals to loyalty and altruism fail, psychological torment is all that's left. (A professor may be given a heavy class load, with early and late classes every day and duties on Saturday. Graders are unavailable. Requests are denied. Committee work is assigned. Meeting announcements arrive late. Suggestions are ridiculed.)
Other workers in the office will often join in the torture, cutting the victim out of their information networks and social groupings. (The professor's clerical work may be done last, with no attention to quality. Her every mistake or snub may be discussed.) One subconscious reason is that anyone forced out by office politics opens up a job slot, increasing the seniority and security of those who remain. Staff snubs are also motivated by general resentment and frustration at pervasive incompetence. And former friends may fear that the victim's harassment will be extended to them.
New hires (and volunteers) tend to believe that a nonprofit is a "family," united in pursuit of a common good. They expect managers to know them personally and to genuinely care about them. They also expect to put in unpaid overtime when critical deadlines approach. Managers will use these expectations to draw out extra effort. Power tends to be very centralized in a nonprofit, but the real work is done by the people at the fringes. (In a university, that would be the graduate students and assistant professors.)
Eventually the worker "wises up," but not before learning to describe every problem and achievement in politically correct terms. Learning to play the game means learning when platitudes can be substituted for accomplishment. Org-speak reassures the boss that you're not out to change the system. To change procedures you must show that no real change in values is involved. Trying to change the system without loving it is as unforgivable as criticizing another person's child.
Although nonprofits exist to serve their clientele, many do it ungraciously. Clients may or may not be aware of the workers' contempt for them, although they are likely to sense inefficiency and a lack of courtesy. The nonprofit typically has a monopoly -- and even its own managers can't discipline uncooperative employees -- so clients who are dissatisfied have little recourse except political warfare. Upper management may spend much of its time dealing with such outside threats.
Two types of managers thrive: those who cannot delegate and those who cannot make a decision. The former have extreme power, the latter duck all responsibility. Marilyn Moats Kennedy says that both styles result in cynicism, low staff productivity, and high political activity.
There are exceptions, of course. I really can't fit the Girl Scouts into this mold, although there are frictions between the national staff and the local volunteers. (U.S. Girl Scouts were named the Best Managed Volunteer Organization a few years ago.) I'll also have to excuse SRI from much of this diatribe -- they pay pretty well and operate on reasonably competitive business principles. SRI encourages procedural discussions, and I ran across very little office politics. Jobs are remarkably stable, but people in the lower ranks can advance by bringing in new business. It's close to being the best of the academic and business worlds -- when projects are available.
Why work? If you're not motivated, you won't do a good job. And when you look elsewhere, you'll find employers wondering whether they could motivate you any better than your past boss did. Failure to advance is seen as lack of motivation, and that makes you a bad risk.
Different people are motivated in different ways. A poor manager will try to motivate all employees through criticism and threat of firing. This works much of the time, reinforcing the manager's behavior. People who can't produce in such an environment will be fired, demonstrating that the threats have teeth. A fear-motivated team will evolve, becoming proficient at ducking criticism. Work will be done in a routine manner, and no one will dare to show initiative. This is characteristic of bureaucracies.
A better manager will set up a "racetrack," motivating employees to compete for visible goals. Most people thrive on attention and praise, so the manager makes it clear what accomplishments will win these rewards. Dale Carnegie wrote about the time Charles Schwab visited a mill where the men were slacking. The manager had coaxed the men, sworn at them, and threatened them; nothing worked. Schwab asked how many "heats" had been produced that shift, took a piece of chalk, and wrote a big "6" on the floor. He said nothing more. By morning, the night shift had rubbed it out and replaced it with a "7." The next day, the shift raised it to "10." Soon the mill was the most productive in the factory. (Schwab attributed this to a desire to excel.)
The best managers will find individual motivation for each employee. One who is ambitious will be given opportunity for advancement. One who craves power will be given supervisory tasks. One who works for money will be spurred with commissions or contingent bonuses. One who works to help others will be given service tasks. This may appear unfair (and often sexist), but it gets the greatest output from the least staff, with a minimum of turnover.
You can't do your best unless you know what motivates you. If you want security and a gold watch, look for a bureaucracy. If you have a dream, you may do well in a nonprofit. If you enjoy helping others, look to education or the service sector. If you enjoy competing, find a competitive environment such as sales. If you want rapid advancement, start in a small company and transfer if blocked. And if you want full autonomy, start your own business.
Any company can earn more by paying its workers less, and it can pay them less if it can "de-skill" the jobs so that fewer skills are needed to perform them. (Paradox: If all companies do this, workers will have so little to spend that all companies will make less money, not more.) The alternative -- INCREASED complexity, productivity, and pay -- is rarely tried, but it can work if pay is tied to performance rather than hours on the job.
Jim McNair [SJ Mercury, 8/4] wrote about the team of 12 that developed the IBM PC. It was 100-hour weeks with desktop naps, yet people fought to join the team. "Where I worked, there were 20,000 people who looked just like me," said marketing manager Doug Johns. "The odds of doing anything substantial in that environment were pretty slim." Now he's VP of Corporate Marketing for Compaq. Dan Wilkie, the PC manufacturing manager, has said he desperately wanted to relive the experience. Three companies later, he's president of Solbourne Computer.
Other lives were also changed by the high-pressure freedom. Dennis Andrews is now president of XSoft, a Xerox business unit. Joseph Bauman is dean of the Kansas University School of Business. Jim D'Arezzo is VP of marketing for Banyan Systems. William Lowe is president and CEO of Gulfstream Aerospace Corp. Bill Lyons is chairman and CEO of Ashton-Tate. Mike Maples is Senior VP of Applications at Microsoft. Joseph Sarubbi is an independent consultant. Mike Shabazian is president of Intelligent Electronics. H.L. Sparks is VP of worldwide sales and service for Athenix Corp. William Sydnes is VP of development programs for Commodore. Max Toy is VP of sales and worldwide marketing for Ashton-Tate. And Joyce Wrenn is VP of information systems technology for AMR Corp.'s Sabre Computer Services unit.
Nothing succeeds like success. But note that people -- even Americans -- are eager to work hard if there's a good reason. And that once you've enjoyed the freedom of goal-driven development, there's no going back.
The Cygnus contact information in V1 N17 was incorrect. It should have been (415) 322-3811; (415) 322-3270 Fax; firstname.lastname@example.org. [MicroTimes, 8/5.]