Microsoft Word Hopper Grace oral history. 1980. 102702026. final doc

 

 

CHM Ref: 

X5142.2009

                    © 1980 Computer History Museum                           Page 

44

 of 54

 

 

 

 

again. And many of the concepts are now coming back again, just like the structured 

programming. 

We had an infinite memory so we didn’t have to do any of those things anymore. So along with 

the vacuum-tube computers we threw away everything we had learned instead of carrying it 

with us. We wrote everything in modules from day one, independent modules, now that’s 

coming back again. The reason that didn’t get passed on is that those of us who learned it were 

all so busy writing programs, and we didn’t do the teaching. We left the teaching to some other 

people who didn’t realize it. 

Pantages:  

You changed that. 

Hopper:  

Tried to. There’s a good deal of divergence now between the academic computer 

scientist and the data processing people, and you’ll even find it throughout the industry. And I 

think we may get somewhere. The biggest fault of the computer scientist is that they don’t teach 

documentation. Once you’ve written a program and got it running, that’s considered it; you don’t 

have to report on it. And when I took a science in college, every time I performed an experiment, 

I had to write a report on it. Chemistry, physics. And they ought to do the same on computers. 

That’s an integral part of the experiment, but we’ll get around to it. 

Pantages:  

I find you very accepting of human nature, but yet very impatient with it. 

Hopper:  

It’s a combination of the two. You can’t help getting impatient, even though you 

know it’s not going to do any good. Like I said, there have to be a few people who get out in 

front and holler. And they have to learn not to be frustrated. Because getting frustrated only 

stops you. So you have to know that if you have a nice new idea it’s going to take five years to 

get it across. Then you don’t get frustrated. Every time you move an inch you feel happy about 

it. You just have to realize there is a world out there and this is the way it is. 

I’ll never forget Burdette and Bloch wanted Commander Aiken to do something at some point 

and the Commander wouldn’t do it. And I had to try and tell Burdette that the Commander is 

exactly like the computer. He was wired a certain way and he’s not going to change that. And 

this is how you have to go at it and not get frustrated.  

Pantages:  

What happened when you were pushing against operating systems and you were 

up there at the University of Pennsylvania? 

Hopper:  

I didn’t really push against them. I was still at the stage of trying to figure out how 

I could get them apart.  

---

 

 

CHM Ref: 

X5142.2009

                    © 1980 Computer History Museum                           Page 

45

 of 54

 

 

 

 

Pantages:  

At this point you were at with the research group at the University of 

Pennsylvania, right? 

Hopper:  

It wasn’t until a few years after I got down here in the Navy that I began to really 

see what I could do with it. But the effort came from trying to get rid of all that general- purpose 

stuff in a program, in execution I mean. I began to look at the different parts of the operating 

system and see if they could be replaced by generators. That could generate the specific code 

for a specific problem. And then what would I have to know, and so on and so forth.  

The generator of course comes from way back in the beginning, when I mentioned Betty 

Holberton’s software generator. But what the generators really were was specialized compilers. 

They generated only a part of a program, or specialized program. They weren’t general 

purpose. And actually, the first COBOL had an input-output generator in it as part of it. And then 

I realized that what we have now is a general-purpose interface to the operating system that did 

that function. All right, pull that piece out of the operating system and make it a generator again. 

And the same thing is true on the interrupts. You could for any particular program…it’s a subset 

of the interrupt process that you want so you need a generator to generate that.  

The concept of generator disappeared with those old machines, where we had to save space. I 

think it’s time to bring it back again to save time. Once you get on to that you got on to the idea I 

was starting up this week that you can break up your compiler into separate generators and try 

to get rid of the operating system. It’s going to take another five or 10 years, but we’ll get there. 

Pantages:  

What came out of your work at the University of Pennsylvania? 

Hopper:  

Nothing that was really published except better COBOL compilers. Just a lot of 

thinking. A lot of people who moved ahead and got some new ideas.  

1980:  Back to the Navy 

Teaching the Next Generation 

Pantages:  

And then you said a couple of years after you got to the Navy… 

Hopper: 

Norm Ream asked me to come back. 

Pantages:  

You started to do more work in that area… 

Hopper:  

No, the work was on the test routines, but I was thinking in the other directions. I 

haven’t finished it yet.  

---