History of Franz Inc.

The roots of Franz Inc. began in the early seventies with a mathematical program at MIT named Macsyma. Macsyma, a symbolic manipulator used to solve complex math problems, was originally written in a version of the Lisp Programming Language called MACLisp. Macsyma was such a large program, however, that it could only run effectively on a specially-configured DEC 10 computer which had been loaded up with a huge, expensive memory (huge in those days meant about 2.5 megabytes!).

Professor Richard Fateman, one of Macsyma's original programmers, came to UC Berkeley in 1974 and continued to access the MIT system over the ARPAnet (the predecessor of today's Internet). In 1978, he learned of a new DEC computer, called the VAX-11/780, which could run Macsyma and many other programs more efficiently and at a much lower cost. The VAX could do so because it included a large and inexpensive Virtual Address Extension, a critical feature that gave a programmer the ability to run programs larger than the actual memory space of the computer.

Fateman, along with several other faculty members, received funding from the National Science Foundation to acquire the first of many VAX computers at Berkeley (which they christened Ernie CoVAX). This group went on to make significant modifications to the VAX to make it more useful. They developed a new operating system, VAX UNIX (based on the UNIX operating system) which offered a far more robust virtual memory. Through their research efforts, they created a portable operating system and environment that freed researchers from the proprietary operating systems of vendors. This enabled programmers to buy and use whatever hardware was fastest or least expensive.

Fateman also recruited students (including John Foderaro, Kevin Layer and Keith Sklower, future Franz Inc. founders) to build a version of Lisp that would form the substrate for the version of VAX Macsyma. The Lisp was named "Franz Lisp" after Franz Liszt (1811-1886), not only to fit in with the Berkeley-pun-filled environment, but also to emphasize that this was a quick implementation. As it turned out, Franz Lisp dominated subsequent VAX Lisps that appeared later.

By the mid 1980's, Franz Lisp had been distributed at no charge to thousands of universities and research laboratories, along with the Berkeley Software Distribution (BSD) of Unix. Moreover, new computer systems based on the Motorola 68000 CPU appeared which could also run a version of Berkeley UNIX, leading Kevin Layer to subsequently complete a 68000 port of Franz Lisp.

Fritz Kunze, a graduate student at UC Berkeley and a manager of scientific software support at the UC Berkeley Computer Center, saw a business opportunity: Fateman's UCB group was research-focused and unable to adequately support other Franz Lisp users. Yet, the creation and distribution of new workstation machines were on the rise, perhaps an external company could port and support Franz Lisp on these new, powerful computers. Despite a proliferation of free, but unsupported Lisp systems, Kunze believed that important commercial applications would need significant customer support, a high quality product and active development. Franz Inc. maintains this support philosophy to this day.

Encouraged by Professor Fateman, and joined by John Foderaro and Kevin Layer (and Keith Sklower, who later went back full-time to academia), Franz Inc. began doing business in 1984 in the spare bedroom of Fritz Kunze's house. The initial funding of $500 was provided by the founders, and was used to incorporate the company (thanks, Nolo Press), and buy blank stock certificates and an embosser. David Margolies and Charley Cox soon joined the company as well.

There was no money to buy the computers they needed, which in those days could cost over $10,000 -- so they negotiated a deal with Sun Microsystems which would give them a free computer if they ported Franz Lisp to their latest workstation. Kunze and Foderaro were so nervous that Sun would back out of the deal that they drove down that same day to pick up the computer. Franz continued this practice with every contract negotiated, and soon they had numerous workstations to work with.

In 1985, just as Franz was getting up and running and focusing their efforts on finding Franz Lisp porting contracts, the Department of Defense began to exert pressure on the Lisp Community to create a new unified language called Common Lisp. (Prior to that time, the DoD had been supporting research in artificial intelligence and expert systems in several different Lisp dialects: Franz Lisp by UC Berkeley, MacLisp and NIL (New Implementation of Lisp) by MIT, SPICE Lisp by Carnegie Mellon, and InterLisp by Xerox, USC-ISI and BBN)

While this new language was being finalized, a new venture capital backed company named Lucid was formed to implement a version of Common Lisp. Lucid had ties to Stanford University, the Common Lisp Committee, and the DoD. With approximately $25 million in government and venture capital funding, they quickly dominated the Lisp Market.

Franz Inc. was in crisis. They were being shut out of contracts; they had little support on the Standardization Committee; and Lucid had them in their radar. In addition, it was clear that Franz Lisp would need major revisions to become an implementation in Common Lisp. The founders met and decided that John Foderaro would work at home (to minimize interruptions) for a year and develop a Common Lisp version from scratch.

To make matters worse, money was tight and Franz needed a big contract to stay in business. Tektronix was making an AI machine and wanted a version of Lisp included in it. They became Franz Inc.'s first major contract, and kept the company going. Franz offered Tektronix a two-step option at a significantly lower price than Lucid: Franz Lisp initially, followed by a version of Common Lisp.

At the end of 1986, Franz Inc.'s version of Common Lisp was ready for distribution. Initially named ExCL (Extended Common Lisp) in order to distinguish it from Franz Lisp, the product was later renamed "Allegro CL" in 1988 to retain the company's musical theme.

Shortly after Allegro CL was released, Franz landed their second major contract -- creating an Allegro CL port to the Cray X-MP supercomputer on the Unicos operating system. The work was done by Duane Rettig, who had recently joined Franz from Amdahl Corp. While at Amdahl, Rettig had ported Franz Lisp to Amdahl's UTS operating system on IBM-370 compatible machines.

After the Cray contract was completed, Franz Inc. was able to garner a number of smaller customers, but lost a sequence of key sales to large companies (Sun, HP and IBM) who chose to re-label Lucid's Common Lisp as their house brand. The company faced significant competition from another Lisp vendor, DEC Common Lisp, as well.

Franz Inc. fought back by publicizing several simple benchmarks showing that Allegro CL ran faster than Lucid Lisp and could work on machines which didn't have a lot of memory. (Franz had developed their Lisp on a slow Motorola 68020 workstation vs. Lucid's development on powerful Sun workstations -- Allegro CL was smaller and more efficient). This extra speed, coupled with significantly faster compilation times over Lucid Lisp, enabled Franz to compete more effectively against the larger company.

Another factor that significantly contributed to Franz Inc.'s growing success was the company's decision to design hardware-independent software. Doing so, enabled Franz to provide direct sales and technical support to its customers, and allowed customers to choose their platform based on best price/performance versus being tied to a particular machine. This was in contrast with Lucid who, like many other software vendors of the era, customized and sold their product through hardware vendors.

The collapse of hardware-labeled software, an ill-fated venture into the world of C++, and the disillusionment of their venture-capital funders, forced Lucid to declare bankruptcy in 1994. Another Lisp Company, named Harlequin, ultimately picked up Lucid's technology and renamed it "Liquid Lisp."

Franz continued to grow, and along with the rest of the computer industry took advantage of the improved memory and speed of PC's, and the rise of Microsoft in the commercial development arena. Realizing that a PC version of Lisp was needed in addition to its Unix product, Franz purchased a Windows version of Lisp called Procyon Lisp with the intention of using it for a couple of years until Unix Allegro CL could be ported to Windows. In reality, Franz did not unite the two versions until 1998, when Allegro CL 5.0 was released.

1999 marked another year of transition for Franz Inc. The company's key competitor, Harlequin Inc., was sold to a Belgian company and the Lisp products were spun off into a new company called Xanalys. Although Xanalys is maintaining the Harlequin products, Franz remains the leading commercial Lisp vendor today.

As the new century unfolds, Lisp technology has begun experiencing a rebirth. Functionality such as garbage collection, which was an element of Lisp once derided by critics, is now one of the differentiating features of Java. The increased speed and memory capacity of today's computers, as well as improved software technology, have eliminated many of the concerns that developers had with Lisp in the past as well. Many developers are beginning to view it as the ideal technology for today's software development and delivery needs. The demands placed on software applications have become significantly more complex, and require increased flexibility, customization, and rapid development. Lisp is being used in all areas of development including the Internet, E-Commerce and B2B applications.

Franz Inc. has thrived by making it easy for Lisp users to play leading roles in the software industry, and continuously striving to introduce new functionality and features. Further, Franz Inc. understands that it is critical to grow the Lisp market -- special academic pricing, educational tools and scholarship programs, have all been implemented to ensure that the pool of Lisp users will continue to increase as new people discover this vibrant, dynamic language.