A personal History of dBASE
by Jean-Pierre Martel, Editor of The dBASE Developers Bulletin
My first encounter with dBASE was with dBASE-II on an Apple-II clone (a 64K RAM computer). I was in charge of an Association and we had found a young programmer willing to make a custom database application for us. He chose to write it in dBASE. He was very busy at the time, and he would come to my apartment four hours late, type in some code for 45 minutes to an hour, then leave suddenly to work on another contract. Because our contract with him had some penalty clauses for any delay, we were absolutely sure we would get our application on time but as the weeks passed after the deadline, I became more and more nervous.

I had learned how to load dBASE on my computer. When I started dBASE it would appear as a black screen with a nice flashing dot at the bottom left corner. I also knew how to call our unfinished (and buggy) application. It would crash after one or two seconds, but would display an error message about a bug in a specific line. I would then consult the printout of the code, go to the erroneous line and try to find out what was wrong. Just by looking at the good code that had been accepted by dBASE up to that line and comparing it to the bad line, I soon realized that when we open quotes, we have to close them; and every “if” is followed by an “endif”, sometimes with an “else” in between. This is how I started learning the dBASE language.

This language was quite simple with its English-like commands (use, find, append, list, etc.). It was invented by Wayne Ratliff, a young programmer at the Jet Propulsion Laboratories in Pasadena, California. Ratliff had the idea of a database program for desktop computer based on the one used on mainframes at work. He implemented it on his kit-built computer at home. Not only could his software create a table, save and load data, but it could also create pure ASCII programs (like DOS batch files) which could display and print data. Ratliff called his software Vulcan and began to sell it by mail order.

Not long after, a guy named George Tate saw the potential for that software. He contacted Ratliff and convinced him to create a company to sell the software. This company could have been called Ratliff-Tate but George Tate wanted a more respectable name - Ashton-Tate - even though there was nobody named Ashton among the shareholders nor among the employees. Secondly the name Vulcan might be appropriate for Spock's home planet, but for a database software it would be meaningless. So he changed it to dBASE-II, even though it was not the second version at all: just Vulcan with a more established name. In Canada, it was sold for 900$ (about $650 US).

When the first IBM-compatible desktop computers appeared, a version for DOS was created. I never saw dBASE-III, just dBASE-III Plus. The “Plus” was for a new DOS interface (called Assist) with its drop-down menu (à la WordStar - the first popular word processor on IBM-PC). We could access all the power of dBASE-III through its interactive menu. Being now more user-friendly, dBASE-III Plus became a huge success. Millions of people used it daily. dBASE was not only a database program, it was also a programming language. With its large number of commands, you could not only create pure database applications, but also accounting applications (something no spreadsheet software could do), or even scientific applications.

At that time, nearly every computer had WordPerfect, Lotus 1-2-3 and dBASE-III Plus. In 1987, just before the release of dBASE-IV, 35 years old Edward Esber Jr. (A-T's Chairman, President & CEO) earned 874 K$ a year (more than 51 years old Andrew Grove, Intel's President & CEO). These were the most popular softwares:

 In 1987  Rank   Product  Relative Sales 
 Top IBM-PC Software Titles
1
 Lotus 123
100
2
 WordPerfect
45
3
 PFS: First Choice 
31
4
 dBASE-III
31
 Top Macintosh Software Titles 
1
 Word
100
2
 Excel
76
3
 MacDraw
68
4
 dBASE-III
66

In 1988, the five-year sales growth was 76.2% for A-T, 63.8% for Microsoft, 54.6% for Lotus.

Because its programming language was so easy to learn, millions of people were dBASE programmers without knowing it. To use Randy Solton's words: “In essence, dBASE brought programming power to the masses.” This popularity spawned a cottage industry of dBASE compilers or clones. dBXL, QuickSilver, Arago, Clipper and Force allowed developers to compile dBASE-III Plus programs into DOS exe files. FoxBase was a clone that was faster than the original.

Secretly Ashton-Tate (A-T) had approached Fox Software to buy their software, but the negotiations broke down and soon A-T sued Fox for alleged copyright infringement. I was always for competition through innovation but mostly hostile to competition through court actions. As an end-user, the same day I learned the News, I've decided to get FoxBase. A-T also wanted the tribunals to declare the dBASE language proprietary. This move was a big mistake as developers became concerned that A-T would ask for a royalty for each copy of an application sold. Many programmers began to look at alternatives.

Meanwhile, rival companies were nurseries for talent (from which one person would stand out: Randy Solton at WordTech, as we will soon see). Against the competition, A-T counter-attacked with dBASE-IV 1.0 in October 1988. This version was supposed to include everything the competition had to offer and more. But it was released prematurely; the promised compiler was not included and the software had many memory problems. Instead of fixing the problems in the DOS version, the developers were instructed to port dBASE to the 32 bit operating system of the future: IBM OS/2. For a year and an half dBASE users waited for dBASE IV 1.1, and meanwhile, competitors were making better products than dBASE itself.

In the spring of 1990 Microsoft released Windows 3.0, and it  became an overnight success. Every computer magazine was talking about Windows and every user was asking for software that would work under Windows. A-T was unable to meet this demand. In 1991, a small and dynamic company named Borland began showing a database software program that displayed a dot prompt in Windows that could run existing dBASE III Plus programs. They were able to convince A-T's managers that Borland 's engineers were better qualified to insure the future of dBASE. In the fall of that same year, Borland bought the software giant A-T.

On the DOS front, Borland released in March 1993 a new version of dBASE that fulfilled the promises made years before by A-T. Under the guidance of Tom Burt, dBASE-IV 2.0 became a reliable product and had the long awaited compiler.

On the Windows front, George Fruend headed the dBASE for Windows project. But it was not very user friendly. Meanwhile at WordTech, Randy Solton and his team were doing Arago for Windows. This software was well in advance of its time. It used a virtual machine, an idea that Java developers copied many years later. The company just had to make a virtual machine for the Mac, for Unix, etc., and the same application could be used on all these platforms without any changes. It was revolutionary.

When some Borland officials heard about and first saw a prototype of Arago for Windows (demonstrated at  Comdex), Borland headquarters began looking like Panic City: Borland had to buy Arago. But above all, Borland had to get those geniuses at WordTech. The easiest way to have them was to buy their employer. By doing so, Borland had two different versions of dBASE for Windows. Arago for Windows became the foundation of dBASE for Windows 5.0, released in August 1994. The only thing salvaged from the previous dBASE project was the debugger. Meanwhile Microsoft had released Access, a user-friendly database software: more than half a million copies of Access were sold the first year. When Borland released dBASE for Windows 5.0, the Company discovered that the market had moved elsewhere.

Commercially, Borland paid dearly for the long development time required for a version of dBASE. But for those using it now, it means a software with solid foundations, a robust set of stock classes and an outstanding technology not even approached by rival database products.

For those unfamiliar with the ins-and-outs of the development of a piece of software, it  may be surprising to learn that the team that brought us dBASE for Windows consisted of very small number of people. In a rare interview given to the dBulletin on dBASE2000's WishList newsgroup, Randy T. Solton declared May 21st: “Just counting developers, the Arago team consisted of about nine developers (five core), all of which came to Borland. At Borland, we added about four developers. There were a few others at times that worked on the code base briefly. Of the original team of nine, and ten years later, all but one remained through VdB 7.01. Of the four we picked up at Borland, one remained through VdB 7.01. Naturally, there were numerous other very important and talented people in management, QA and writing that were very dedicated as well.”

Last February, when Inprise (Borland's new name) started dismantling and laying off the golden team, the News had a devastating effect on the dBASE community. What none of us knew (but a few suspected) was that Inprise was secretly negotiating the sale of dBASE to KSoft. KSoft is headed by Alan A. Katz, a long-time dBASE advocate and a well-respected man in the dBASE Community. Like the millionaire who liked his electric razor so much that he bought the company, Mr. Katz made so much money making dBASE applications that he could afford (or his banker could afford) to buy the software from which he earns a living. More seriously, this is the first time in History that a major software is given into the hands of some of its users.

dBASE changed hands this last March. Alan created a subsidiary of KSoft to take care of the legendary software, provided a number of newsgroups to insure customer support, built a new confidence in the direction he intended to take the product, and began hiring competent people to get the company moving ahead. But this means a new chapter is being written in the history of dBASE. From what we hear and see, this chapter promises to be very interesting. Ken Mayer's tutorial is a major milestone. Previously there were no books on VdB 7/dBASE2000? Now three are under way (one by Alan Katz, one by Ted Blue and one by Michael Rotteck). Before, there was no magazine about dBASE. Now there is the Xbase Files Magazine and the dBASE Developers Bulletin. And just six weeks ago I had no idea that the dBulletin would be published monthly. Things are moving fast for dBASE, very fast, like the World around us. And in this fast moving World, anything can happen...


For details about the different versions of dBASE, click here