Wang Laboratories

Wang logo circa 1980.

Wang logo circa 1970.

Usage restricted. Trademarks on this page belong to their owner. See .

Wang Laboratories was a computer company founded in 1951 by Dr. An Wang. The company was successively headquartered in Cambridge, Massachusetts (1954-1963), Tewksbury, Massachusetts (1963-1976) and Lowell, Massachusetts (1976-1992). At its peak in the 1980s, it was earning revenues of $3 billion/year and employed over 30,000 people.

The company was always directed by Dr. Wang, who played a personal role in setting business strategy and product strategy and thus must be credited both with the company s successes and failures.

Dr. Wang took steps to ensure that the Wang family would retain control of the company even after going public. He created a second class of stock, class B, with higher dividends but only one-tenth the voting power of class C. The public mostly bought class B shares; the Wang family retained most of the class C shares. (The letters B and C were used to ensure that brokerages would fill any Wang stock orders with class B shares unless class C was specifically requested). Wang stock had been listed in the New York Stock Exchange, but this maneuver was not quite acceptable under NYSE s rules, and Wang was forced to delist with NYSE and relist on the more liberal American Stock Exchange.

Under his direction, the company went through several distinct transitions between different product lines.

=Typesetters=

The company s first major project was an electronic phototypesetting, the Linasec, introduced in 1964. It was developed under contract to Compugraphic, which retained the rights to manufacture the machine without royalty. Compugraphic exercised these rights, effectively forcing Wang out of the market.

=Calculators=

The Wang LOCI-2 (there had been a LOCI-1 but it was not a real product) was introduced in 1965 and was probably the first desktop Calculator capable of computing logarithms, quite an achievement for a machine without any integrated circuits. The electronics included 1275 discrete transistors. It actually performed multiplication by adding logarithms, and roundoff in the display conversion was noticeable; 2 times 2 yielded 3.999999999.

From 1965 to about 1971, Wang was a calculator company, and a very well-regarded one. Wang calculators cost in the mid-four-figures, used Nixie tube readouts, performed transcendental functions, had varying degrees of programmability, and exploited magnetic core memory in ingenious ways. Competition included Hewlett-Packard, which introduced the HP9100A in 1968, and old-line calculator companies such as Monroe and Marchant.

Wang calculators were at first sold to scientists and engineers, but the company later won a solid niche in financial-services industries, which had previously relied on complicated printed tables for mortages and annuities.

One perhaps apocryphal story tells of a banker who spot-checked a Wang calculator against a mortgage table and found a discrepancy. The calculator was right, the printed tables were wrong, and the company s reputation was made.

In the early seventies, Dr. Wang believed that calculators would become unprofitable low-margin commodities, and decided to exit the calculator business.

=Word Processors=

The Wang word processor was designed by Harold Koplow and David Moros, who began by first writing the user s manual. This has long been known, and regarded as a brilliant design strategy. But it was apparently not a deliberate one. A 2002 Boston Globe article refers to Koplow as a wisecracking rebel who was waiting for dismissal when, in 1975, he developed the product that made computers popularly accessible.

In Koplow s words, Dr. Wang kicked me out of marketing. I, along with Dave Moros was relegated to Long Range Planning— LRPed . This, up until then, was tantamount to being fired: here is a temporary job until you find another one in some other company.

Although he and David Moros were indeed told to design a word processing machine, they were given no resources. They perceived the assignment as busywork. They went ahead anyway, wrote the manual, and convinced Dr. Wang to turn it into a real project. The word processing machine—the Wang WPS—was introduced in June 1976 and was an instant success, as was its successor, the 1977 Wang OIS (Office Information System).

These products were technological breakthrough in their day. They were multi-user systems. Each workstation looked like a typical computer terminal of its day, but contained its own Zilog Z80 microprocessor and 65K of Random access memory (roughly comparable in power to a typical 1982 IBM PC). Disk storage was centralized in a master unit that was shared by the workstations, and connection was via high-speed dual Coaxial cable. Multiple OIS masters could be networking to each other, allowing file sharing among hundreds of users. And the systems could be easily operated and administered by office personnel without special training (in the days before schools taught computer literacy ). It was an astonishing achievement.

All software for the systems was developed by Wang Laboratories, and the operating system, file formats, and electronic interface specification were closely held trade secret. Wang did not want third parties developing for or interconnecting with its systems. (This was relaxed somewhat in the late eighties).

=Wang VS minicomputer=

The Wang VS minicomputer computer was introduced in 1979, about the time as Digital Equipment Corporation s VAX. Its instruction set was compatible with the IBM 360 series but it did not run any IBM 360 software. The VS operating system and all system software were built from the ground up to support interactive users as well as batch operations. The VS was aimed directly at the business market in general, and IBM in particular. While many programming languages were available, the VS was typically programmed in COBOL, and Wang put the bulk of its compiler development resources behind its COBOL Compiler.

Dr. Wang felt a personal sense of rivalry with IBM, partly as a result of heavy-handed treatment by IBM in 1955-6 over the rights to his magnetic-core patents. (This encounter formed the subject of a long chapter in Wang s own book, Lessons. ) According to Charles C. Kenney, Jack Connors remembers being in Wang s office one day when the Doctor pulled out a chart on which he had plotted Wang s growth and projected that Wang Laboratories would overtake IBM sometime in the middle of the 1990s. He had kept it a long time, says Connors. And he believed it.

Wang was one of the first computer companies to advertise on television, and the first to run an ad during the Super Bowl. Their first ad literally cast Wang Laboratories as David and IBM as Goliath. A later ad depicted Wang Laboratories as a helicopter gunship taking aim at IBM.

Wang wanted to compete against IBM as a computer company, selling directly to Management information system departments. Before the VS, however, Wang Laboratories was not taken seriously as a computer company. The calculators, word processing systems and OIS were sold into individual departments, bypassing the corporate data-processing decision-makers. The chapter in Wang s book dealing with them shows that he saw them only as a beachhead in the Fortune 1000. The Wang VS was Wang s entrée into IT departments. In his book, Dr. Wang notes that, to sell the VS, we aggressively recruited salesmen with strong backgrounds in data processing... who had experience dealing with MIS executives, and who knew their way around Fortune 1000 companies. As the VS took hold, the word processor and OIS lines were phased out. The word processing software continued, in the form of a loadable-microcode environment that allowed VS workstations to take on the behavior of traditional Wang WP terminals to operate with the VS and use it as a document server.

=Decline and fall=

The decline and fall of Wang Laboratories can be seen from a number of viewpoints. A common view within the PC community is that the company failed because it specialized in computers designed specifically for word processing and did not foresee (and was unable to compete against) general personal computers with word processing Software in the 1980s. This view is skewed, though, because word processing was not the mainstay of Wang s business by the time PCs began to gain in popularity. Although Wang manufactured PCs, its main business by the 1980s was its VS line of minicomputer systems.

There were other factors, however. High among them was Dr. Wang s insistence that his son, Fred Wang, succeed him. Fred Wang was an intelligent, able, business-school graduate, but by almost any definition, wrote Charles C. Kenney, unsuited for the job in which his father had placed him. His assignment, first as head of Research and development, then as president of the company, led to jealousy and to resignations by key R&D and business personnel.

One turning point occurred when Fred Wang was head of R&D. In October 1983, Wang Laboratories announced over twenty major hardware and software products. The announcement was well received, but very few of the products were close to completion and many of them had not even been started. All were delivered late and some were never delivered at all. In retrospect this was referred to as the Vaporware announcement, and it hurt the credibility of Fred Wang and Wang Laboratories.

In 1986 Fred Wang, then 36 years old, was installed as president of Wang Laboratories. Things did not go well. The company was soon clearly in decline and, on August 4th, 1989, Dr. Wang had to fire his son.

Wang Laboratories eventually filed for bankruptcy protection on August 18, 1992. The three Wang towers in Lowell, which originally cost $60 million to build and housed 4,500 workers in over a million square feet (100,000 m²) of office space, were sold for $525,000. Wang itself would have bought the Towers property at the foreclosure sale but no one at Wang had anticipated that the final price would be so low, and so an opportunity to reacquire the Towers was lost.

The company emerged from bankruptcy a year or two later with $200 million in hand and embarked on a course of acquisition and self-reinvention, eschewing its former role as an innovative designer and manufacturer of computer and related systems. Later in the 1990s with the acquisition of the Olsy division of Olivetti the company changed its name to Wang Global. By then Wang had settled on network services as its new chosen business.

In 1999 Wang Global, by then back up to $3.5 billion in annual revenues, was acquired by Getronics N.V. of The Netherlands, a $1.5 billion network services company active only in certain parts of Europe.

The Wang VS mainframe computer product line, not actively marketed since the 1992 bankruptcy and now but a tiny portion of the Getronics business, survives to this day (Mar 2004) at a level somewhere in the vicinity of 500 to 1,500 systems worldwide. The most advanced VS model, capable of supporting over 1,000 users -- the VS18000 Model 950 -- was released in 1999, and smaller models based on the same CPU chip were released in 2000 -- the VS6760 and the VS6780.

=Rebirth of the Wang VS=

In 2005 [http://www.getronics.com/us/en-us/getronics/press/press_releases/press_releases/transvirtual_systems.htm Getronics announced] New VS, a product that is said to seamlessly run the VS OS and all VS software. It is based on a hardware abstraction layer for Intel x86 and IBM POWER. The product is a joint commercial effort of Getronics and [http://www.transvirtualsystems.com/ TransVirtual Systems], developer of the virtualization technology used.

Wang VS conversions (or migrations ) have historically been fraught with difficulties, risk, and cost. Wang VS OS, workstation, compiler language and database functionality have usually embodied extensions that have been difficult to duplicate in target systems for replatforming. VS software can be run on the New VS without program or data conversion.

The New VS consists of virtualization software running on specially configured hardware platforms. Though built from commodity hardware, the requirements for very specific configurations require that the hardware platform be provided by Getronics. The New VS is capable of interoperating with SCSI-based Wang VS tape and disk drives, which provide a means of restoring VS files from standard backup tapes or copying VS disk drives.

The New VS reportedly runs VS software with higher CPU and disk I/O performance than the fastest legacy VS, the VS18950, released in 1999.

=External links=

*[http://www.tjunker.com/ The Unofficial Wang VS Information Center] The longest running site specializing in data and background on Wang s VS product line *[http://www.transvirtualsystems.com/ TransVirtual Systems] The New VS, a commercial Linux-based hardware abstraction that runs the real VS OS *[http://home.planet.nl/~ernest/vs.html Emulating a Wang VS!] A free PC-based emulator of the Wang VS *[http://www.thebattles.net/wang/emu.html Wang 2200 Emulator] A true emulator of the Wang 2200 *[http://www.oldcalculatormuseum.com/wangloci.html The Wang LOCI-2] This site also has descriptions, photographs, and technical notes on many other Wang calculators *[http://www.oldcalculatormuseum.com/d-wangmem.html Wang 300-Series Calculator Memories] User-eye view of Wang calculators *[http://www.oldcalculatormuseum.com/d-wangmem2.html My Early Days at Wang Laboratories] Reminiscences by Dennis McNurland *[http://vs18k.dyndns.org/wang.html Wang OIS Information Center] *[http://www.harolds928people.org/ Harold s 928 people] Harold Koplow s website about the development of the 928 products (the Wang WPS and OIS). Includes reminiscences by many who were involved.