Douglas T. Ross Quotes

The artificial intelligence people, the artificial intelligencia,... kept choosing little games to play, and little things that they could master, right? But my whole philosophy has always been give me a really tough problem that's just beyond the state-of-the-art, and give me a whole bunch of users beating on me to get it done. In other words, the real core of what being an engineer is.

The first paper I ever wrote was "Gestalt Programming" and that was in 1955. The whole idea there was to replace the laborious writing out of detailed programs and all those steps by having analyzed a problem area well enough so that you had what I later came to call a "systematized solution." Then you could compose different problems of this class by just plugging together pieces of program, and they would in turn be controlled by a pushbutton language. The user would make a number of discreet selections. It's just like nowadays it's done with menus, and when you had indicated all the pieces that you wanted to put together--by these mnemonic names and words for things associated with buttons, switches, with one meaning "period," essentially, for that sentence, you see--all these things would be brought together and that would be the man/machine, manual-intervention mode of problem-solving. I took over the term from studying Gestalt psychology, meaning that everything was brought together at once, as a unit, instead of this laborious step-by-step build-up.

There is a rigorous science, just waiting to be recognized and developed, which encompasses the whole of 'the software problem,' as defined, including the hardware, software, languages, devices, logic, data, knowledge, users, users, and effectiveness, etc. for end-users, providers, enablers, commissioners, and sponsors, alike.

(SA) combines blueprint-like graphic language with the nouns and verbs of any other language to provide a hierarchic, top-down, gradual exposition of detail in the form of an SA model. The things and happenings of a subject are expressed in a data decomposition and an activity decomposition, both of which employ the same graphic building block, the SA box, to represent a part of a whole. SA arrows, representing input, output, control, and mechanism, express the relation of each part to the whole.

From the computer application point of view the primary problem [of Computer-Aided Design] is not how to solve problems, but how to state them.

The natural law of good communications takes the following, quite different, form in SA:

Neither Watt's steam engine nor Whitney's standardized parts really started the Industrial Revolution, although each has been awarded that claim, in the past. The real start was the awakening of scientific and technological thoughts during the Renaissance, with the idea that the lawful behavior of nature can be understood, analyzed, and manipulated to accomplish useful ends. That idea itself, alone, was not enough, however, for not until the creation and evolution of blueprints was it possible to express exactly how power and parts were to be combined for each specific task at hand.

The real core of what being an engineer is. You have a scientific basis, but when you don't have the science, you put in some bugger factors, some safety factors, and so forth and you get smart enough, and you get the job done anyhow, right? Economically, and as close to on time as you can make it. And if the customer is asking for something that's outlandish, give him what you can, and educate him back to what it is.

This report summarizes the activities of the M.I.T. Computer-Aided Design (CAD) Project from 1 December 1959 through 3 May 1967 in the development of a generalized 'system of software systems' for generating specialized problem-solving systems using high-level language techniques and advanced computer graphics. Known as the AED Approach (for Automated Engineering Design) the Project results are applicable not only to mechanical design, as an extension of earlier development of the APT System for numerical control, but to arbitrary scientific, engineering, management, and production systems as well.

Automatic design has the computer do too much and the human do too little, whereas automatic programming has the human do too much and the computer do too little. Both techniques are important, but are not representative for what we wish to mean by computer-aided design.

I realized after I'd been at MIT for a while that I had never even known the semantics of the word "engineering". You see, all my relatives and contacts were medical doctors or biology and chemistry professors. In fact, I'm almost the "black sheep" in the family for not being an MD or Ph.D. because everybody was doing that sort of thing. There was no contact at all with engineering. I didn't even know what the word meant...

Computer-aided design is not automatic design, although it must include many automatic design features. By automatic design we mean design procedures which are capable of being completely specified in a form which a computer can execute without human intervention.

The assertion that a problem unstated is a problem unsolved seem to have escaped many builders... All too often, design and implementation begins before the real needs and system functions are fully known. The results are skyrocketing costs, missed scheduled, waste and duplication, disgruntled users and endless series of patches and repairs euphemistically called "systems maintenance"

The APT and CAD Projects had one Air Force sponsorship, but it was actually a combination of about five or six projects, all going in parallel on both hardware and software matters, and I couldn't keep track of things, so I would take Polaroid pictures of the blackboard and then say what we had talked about into my dictating machine and my secretary would type it up.

A general theme for what I'm trying to convey and what actually drove me and my very industrious and creative project members over all these years, is... that there is much more to it than pictures. It has to be a picture language. There has to be meaning there, and the meaning is useful. You're trying to solve problems. So it really comes down to man machine problem solving . Better means of communication and expression is what always has driven our work.