Ashton-Tate Buyout Would Aid Users
The initial reaction to Borland International’s intended takeover of Ashton-Tate in this corner was, frankly, not favorable.
The flagship database programs from each company, Borland’s Paradox 3.5 and Ashton-Tate’s dBase IV, version 1.1 (the bug-eradicated version), are good programs and one reason, undoubtedly, is the competition between them.
Also, there are millions of dBase users out there who don’t want to have to switch to Paradox or anything else. What are they going to be, orphans?
No, they won’t be orphans, according to Rob Dickerson, vice president and general manager of the database division at Borland in Scotts Valley, Calif.
The day of the takeover announcement last week, Dickerson met with a roomful of securities analysts in New York to show them the fruits of what amounted to Borland’s plan to take control of the dBase world, whether it acquired Ashton-Tate or not.
From the database consumer’s point of view, it looks like a pretty good plan. It plays off the current weaknesses of both programs, the experience of Borland, and recognizes the chauvinism of the customers.
A little background helps in understanding what is planned.
First off, neither Paradox 3.5 nor dBase IV are Windows-based programs. If we’re to believe that the graphical mode of working enabled by Windows is the way of the future, neither program is now suitable for that future.
Second, there is virtually nothing in common between Paradox and dBase except that they both manage data.
They are, in fact, fundamentally opposite in the way they work. For one thing, dBase typically manipulates data through a series of indexes that can show data in different ways without actually changing the order in which it is stored. Paradox must first store data in a particular order before showing it that way. The same set of data may have to be stored in multiple tables.
In dBase, operations are controlled by a series of programming commands that are typed next to a period on the left side of the screen, called the “dot prompt.” Recent versions give users an alternative way to work by choosing among menu choices and manipulating graphical depictions of tables to create forms and queries. But the dot prompt commands are really being executed in the background.
Paradox is controlled by menu choices and by manipulating a series of permanent and temporary tables of data. A Paradox program is a script that essentially plays out the menu choices in prescribed order and under prescribed conditions. Before you can have much success writing program scripts, you have to thoroughly understand how to use the program through its menus.
Despite their sophisticated programming languages, neither Paradox nor dBase can now handle graphic images or digitized sound or other non-traditional information objects.
So what we really have are two leading database programs from the 1980s that are not up to the demands of the 1990s.
But they are backed by legions of loyal users who sweated hard to learn those programs and aren’t likely to want to learn some other new program as a replacement.
Borland knows that. Its founder and chairman, Philippe Kahn, built the company on sales of a $99 programming language called Turbo Pascal, which spawned legions of new programmers.
Today, in addition to Turbo Pascal, Borland sells Turbo C ++, a modern object-oriented programming language.
What is important about those two dissimilar languages is that they use the same editor and other components that measure performance, find errors and process sections of code into completed programs.
The same concept of sharing common features while producing what outwardly appears to be quite different products is what Borland’s Dickerson showed the New York analysts.
Tentatively called Object dBase and Paradox for Windows, the two Windows programs share the same method of operation and can work interchangeably with dBase, Paradox and other common database file formats.
What really distinguishes them, according to Dickerson, is the programming language they use. Paradox programmers would buy their version and dBase programmers would buy their version, but they would both be able to perform exactly the same tasks on exactly the same data files.
Of course, each language would have added extensions to handle the new data types that Windows makes possible, such as graphics images and even files from other kinds of programs linked as embedded objects in the database files.
For instance, a field in a dBase database could actually be a graph produced by a Quattro spreadsheet.
“We’re trying to do the same thing in the database world that we have done in the languages world,” Dickerson explained.
So, it seems that Borland was planning to lure Ashton-Tate’s dBase customers away anyway. But it will be easier if they buy the company because then they will have their names.
“We have a very aggressive direct-mail marketing program,” Dickerson said.
When will we see the new products? Dickerson won’t say, except to explain that both are in beta test now, which is the final testing phase before marketing.
Completion of the Ashton-Tate takeover has been estimated to take two or three months. Perhaps we’ll see something called “dBase for Windows” hit the streets the same day.
Computer File welcomes reader comments but regrets that the author cannot respond individually to letters. Write to Richard O’Reilly, Computer File, Los Angeles Times, Times Mirror Square, Los Angeles, Calif. 90053.
