A computer scientist and computer corporate executive, Adele Goldberg is best known for her work with Alan Kay and others in developing the object-oriented computing language Smalltalk in the 1970s and 1980s. For this work she has won both the 1987 Software Systems Award from the Association for Computing Machinery and PC Magazine's 1990 Lifetime Achievement Award. Goldberg was born on July 7, 1945, in Cleveland, Ohio, and grew up in Chicago, Illinois. Her parents were Lillian and Morris Goldberg. She received her bachelor's degree in mathematics from the University of Michigan at Ann Arbor, her master's degree in information science from the University of Chicago, and her Ph.D. in information science in 1973 from the University of Chicago. Her doctoral dissertation, titled "Computer-Assisted Instruction: The Application of Theorem-Proving to Adaptive Response Analysis," was prepared at Stanford University while she was a visiting research associate there. She has two daughters, Rachel and Rebecca.
Goldberg began working as a researcher for Xerox in 1973 at its famous think-tank, the Palo Alto Research Center (PARC) in California. One of her most important accomplishments at PARC was managing the System Concepts Laboratory, whose team developed Smalltalk-80, an object-oriented programming language (object-oriented programming languages contrast with procedure-oriented programming languages such as BASIC, FORTRAN, and COBOL).
When Smalltalk was designed, a set of programming tools and a user interface were also invented. The user interface was the first to use pictures that allowed programmers to interact with the computer; a mouse could be utilized to interact with overlapping windows on graphical display screens. These windows contained the tools, menus, or lists of items which programmers employed to send messages to the system. Systems designed for procedure-oriented languages had previously restricted their users to keyboards and to typing in commands on lines of the computer screen. Because of this kind of input system, the interfaces of procedure-oriented programs are often called "command-line interfaces." Many consider object-oriented programming interfaces to be easier for programmers to use.
Object-oriented programming is organized around several important ideas. One of these ideas is modularity. In object-oriented programs, users create autonomous "objects" or modules, each of which contains its own private memory and set of operations. These objects or modules can pass messages to each other requesting information as actions. The objects in object-oriented programming are often represented by an icon and they can look very simple; while appearing very simple on a graphical interface, however, the icon, object, or module can conceal a very complex structure of data and functions. An important virtue of these autonomous objects is that they can be readily customized, copied, and placed where a user wants them. A programmer, for example, may have to assemble relatively few objects to create a program, which can save a lot of time.
Object-oriented programming is organized around a second important idea: that objects correspond to the identifiable parts in a problem situation. For example, if a problem is composed of ten parts, then the programmer can develop ten objects corresponding to each part of the problem, and these objects can send messages to each other to solve the problem. The objects have several important benefits. One advantage is that each can be reused in other situations in which similar problems occur; for example, a company could have many production lines for manufacturing its products, and a program with ten modules could be developed for troubleshooting problems with one production line. If the company needed a troubleshooting program on a second production line, then a programmer would not have to reinvent ten modules to solve this problem. He or she could simply import the useful modules from the first production line, customize them to solve the problems specific to the second production line, and quickly put the program to work. Because the objects in this type of programming are reusable and customizable, and because the objects correspond to the identifiable parts of a problem situation, they can be transferred from one situation to another, saving time and money.
A third important concept behind object-oriented programming is the reusable interface. An interface is the common boundary between two entities; these entities might include people, computer screens, or programs inside a computer system. If an interface is reusable, then it does not have to be redesigned each time it is used in a different situation. Reusable interfaces in object-oriented programs allow data to be passed more easily between objects or modules inside a computer system; they allow data to be passed more easily between a user and the computer interface (examples would be windows and pull-down menus), and they allow data to be passed more easily across a variety of applications in different business environments. Because of the idea of the reusable interface, Smalltalk-80's interface is much more economical and accessible than procedure-oriented interfaces. The windows software interfaces that became so popular in the late 1980s were direct descendants of the innovations developed at PARC by Goldberg, Alan Kay, and others.
After spending fourteen years at Xerox's Palo Alto Research Center developing object-oriented programming, Goldberg wanted to be sure that Smalltalk got a wider audience. She worked out a technology exchange agreement with Xerox, and in 1988 she cofounded ParcPlace Systems, a company that sells development tools for Smalltalk-based applications. Goldberg served as president and chief executive officer of ParcPlace from March, 1988, to April, 1992. Since then she has served the company as chief strategist and chairman of the board. ParcPlace went public on February 1, 1994, and saw its opening stock offer rise six dollars a share before the day was over, a very strong showing. Its revenues in the early 1990s were estimated at over twenty million dollars a year.
Goldberg was president of the Association for Computing Machinery (ACM) from 1984 to 1986, and she won the 1987 ACM Software Systems Award with Alan Kay and Dan Ingalls. She served as ACM's national secretary, and as editor-in-chief of the ACM journal Computing Surveys; she has also been a member of several ACM boards. She was a member of the scientific advisory board for the German National Research Centers, and was a fellow of the ACM. She wrote a regular column for Object magazine called "Wishful Thinking." One of her most significant honors was PC Magazine's 1990 Lifetime Achievement Award.
Since 1990, Goldberg has concentrated on issues that help programmers become more effective in using object-oriented technology to solve their problems. She has written extensively and lectured worldwide on project management as well as on the analytical and design methods needed to implement and advance object-oriented technology. The object-oriented programs sold by ParcPlace will considerably improve the productivity of programmers, because they give programmers greater opportunities to reuse existing objects and because they let programmers construct graphical user interfaces and database applications with visual construction tools.
This is the complete article, containing 1,129 words
(approx. 4 pages at 300 words per page).
