Comment by kazinator | Hacker News Reader

Comment by kazinator | Hacker News Reader

Originally, "code" referred to machine language. "Programming" meant designing the computation, and "coding" referred to translating it to machine language. "Automatic coding" was the term for translating a high level program to machine code, and it covered compiling and interpretation.

You can read all about it in Grace Hopper's 1955 paper Automatic Coding for Digital Computers.

http://www.bitsavers.org/pdf/univac/HopperAutoCodingPaper_19...

Though a woman, she talks the masculinized talk:

"The analysis is the responsibility of the mathematician or engineer, the methods or systems man."

[...]

"The job of the programmer is that of adapting the problem definition to the abilities and idiosyncracies of the particular computer. He will be vitally concerned with input and output and with the flow of operations through the computer. He must have a thorough knowledge of the computer components and their relative speeds and virtues."

[...]

"It is then the job of the coder to reduce the flow charts to the de- tailed list of computer instructions. At this point, an exact and comprehensive knowledge of the computer, its code, coding tricks, details of sentinels and of pulse code are required."

> Though a woman, she talks the masculinized talk

I don't think that's accurate. It used to be that "he" and "man" were understood to be gender neutral in abstract and impersonal cases. Such as "to boldly go where no man has gone before" and "mankind" and "mailman". During Hopper's life, English language changed to expunge the ambiguity.

I have to concede that I still talk in this now outmoded manner. I have to make a conscious effort to switch to modern gender neutral language (eg. they) when speaking or writing to anyone outside my family and friends.

> English language changed to expunge the ambiguity.

I think it's more accurate to say it adopted different ambiguities. The use of 'they' can create ambiguities of number, and just switching between he/she leaves the same ambiguity (is the gendering intentional or not?) albeit in a gender balanced way.

As a side-note, 'he' was used to refer to people, who could be regarded as interchangeable (man or woman) and 'she' was reserved for 'uniquely individual' things, which is why countries and ships (as two examples) are referred to as such. At least, this was may understanding when growing up and I've never lost this habit.

Not only understood, but widely accepted as the grammatically correct general reference gender. English isn’t the only language to do this.

Interestingly, the German translation "one" (as in the gender neutral pronoun: "One must not both eat and talk.") is "man".

This is a beautiful distinction (one that I'm aware of), but which appears to have been lost today. Consequently, these terms are now used interchangeably, leading many people to conflate these two processes. However, this distinction is indicative of how getting a computer to do something has two components: 1. Figuring out what needs doing, and 2. Expressing this in a form that the computer can carry out.

Thank you for the Grace Hopper quotes!

The distinction is lost because in many cases it doesn't hold relevance anymore. You don't need to write flow-charts, state or class diagrams before you start writing code. Todays tooling allows for more iterative approaches and also leaves room for making syntax errors without having to wait hours and days to get "compiler" errors.

But I agree it's still interesting to know of the distinction and know that both can require slightly different skillsets. So when you have someone that is very good at planing and has the experience to layout complex systems in advance, you can assign them an architect title, while someone who knows a programming language inside and outside, you might be better off keeping them in the "driving" seat.

The point here is that when higher level languages were introduced, writing the higher level program was understood to be part of "programming"; and its compilation or interpretation became "automatic coding".

Another thing worth paying attention to in Grace Hopper's paper is this: higher level program code is in fact called "code", but specifically "pseudo code".

That which we call "pseudo code" today doesn't execute. But we sometimes remark that certain high level languages are almost like "executable pseudocode".

At that time, the task of the compiler/interpreter writer was to take "pseudo code", formalize it more precisely so that then the machine could "automatically code" from "pseudo code" to "code".'

At the birth of Computer Science there is this notion that it's (as we might say today) turtles all the way down. Kurt Gödel, Alonzo Church, Alan Turing in the first half of the twentieth century come up with this weird stuff at the edge of the philosophy of mathematics to answer maddening questions, what is "computing" and what makes something "computable"? It seems to have no answer that isn't self-referential. The systems they imagine are endlessly defined in terms of themselves. But they are all theoreticians, and in theory this is perfectly satisfactory.

When actual digital computers get built, the engineers responsible for using them are not at first much interested in the theory of Computer Science, their machines resolutely do not come with an infinitely long paper tape, the problems they're solving are very specific and finite, these are practical people.

But Grace _applies_ the theory. Since the process of turning the high level (e.g. assembly language) into low level (e.g. holes in a paper tape) is mechanical, why not have the machine do it? And if this can be done, why not recurse, and produce even the assembly language from some higher level program. Productivity is improved enormously. And this is at the heart of everything today. Even our microprocessors are really full of microcode that implements more complicated instructions in terms of smaller ones, so that the "machine code" isn't the final form executed by the hardware in practice either.

> Though a woman, she talks the masculinized talk

You may or may not say "systems man" about a woman (depending on the language etc.), but I don't see that the sex of the _speaker_ has anything to do with it.

Reply on news.ycombinator.com