The Queen of Code, a FiveThirtyEight film on Grace Hopper


#1

[Permalink]


#2

Grace Hopper intended the world’s first compiler

Is this some defunct meaning of the word “intended”, like “super-intend” or something?


#3

No, she really intended to make a compiler, whereas gcc was intended to be a program that would print out pictures of gnus as ASCII art. You know how scope creep happens when marketing gets involved.


#4

A full length documentary is in the works, scheduled for a 2016 release:
Born With Curiosity: The Grace Hopper Documentary


#5

So if I take Intended here to mean invented, and not to take away from this great person’s work, I suspect Babbage, Turning, Tommy Flowers would have something to say about this.

if you mean programmed well again I suspect Ada Lovelace would be a bit put out.

This is a very poorly defined, researched and written sentence. As my professor would have said please clarify!


#6

No, they wouldn’t have anything to say about this. Babbage developed the first general computing engine. Lovelace was the first programmer. Turing one of the first great computer scientists. None of them developed, or indeed had access to, the compiler. (Although Turing lived just long enough to see it developed.)

The first compiler was written by Grace Hopper, in 1952 - largely the subsection of compilers we would call a “linker” today.

(Building on that work, the first complete modern compiler happened about 5 years later.)


#7

Sorry my bad. I was so thrown by the “intended” I misread compiler for computer! Still very poor grammar by our contributor here.


#8

Ah, that makes sense now. Completely understandable.


#9

As an old-time coder who began working in COBOL and Assembler back in the day… Admiral Hopper has always been one of my heroes!


#10

Turing got as far as envisaging a library of subroutines that would be linked to make programs. But by “library” he meant, literally, a room full of rolls of paper tape, and human librarians.

Kids today (get orf moi lawn) have no idea just how physical early computing was. I have fond memories of taking taxis across London with reels of tape and printout, reconstructing broken punch tape with a hand punch and glue, and having to call in two large men from Facilities to lift the triple disc drive onto the carpenter’s bench we needed to take the weight. Grace Hopper’s famous bug was just that. A program that managed programs was just…awesome when it happened.


#11

Tommy Flowers was not a computer scientist or a programmer. He was an engineer who saw how racks of valves (tubes) could be made sufficiently reliable to replace relays, and so set the stall out for the high speed digital computer. But the Venn diagram of his work and Grace Hopper’s doesn’t intersect.
His critical discovery was that since it was usually the heaters of tubes that failed, and since switching them on and off was the cause of the failure, the answer was to ramp up their supply voltage slowly and leave them on, even if at reduced power, all the time.
(I used to have a small box of EF50s, the wartime tube that Flowers worked with. I just discovered the other day that though I have no EF50s, I do have a box of EF91s and ECC92s. An ECC92 can just about be used to make a single NOR gate and has about four times the power consumption of a typical mobile phone. Isn’t progress wonderful.)


#12

The cathode heaters were also failing due to impurities in the filament. Special tungsten wires free of traces of silicon had to be used for the high-reliability tubes. (Don’t ask me why, I am still trying to find the material details.)


#13

You are right, and to make my forgetfulness worse I actually still have the monograph on tungsten to which I used to refer, and it has the details. Tungsten is magic stuff, almost as weird as plutonium (which I have never had anything to do with). Its crystal form depends very precisely on the presence of the right dopants. Under the right circumstances, more or less the entire length of a filament can gradually transform at high temperature to a single crystal, with the crystal planes actually continuous across the gap between the two sides of the filament, if that makes sense. (The planes are not distorted around the bends).
You can have an argument about whether it is paradox or irony that the downfall of tubes was silicon both times round.


#14

Author? Title? Year? Maybe it is on my favorite Russian servers…

Plutonium is fascinating. My most favorite metal. I wrote quite some parts of wikipedia dealing with it. The f-orbital related behavior is pretty complex and its phase behavior is a stuff straight from a metallurgist’s nightmare.

I have to read more about this!

Makes a lot of sense. I wasn’t aware about the not-distorted part. (Is the structure of the single-crystal influenced by the surfaces a lot, do they tend to form smooth planes and deform the inside, or does the inside take over and the filament surface becomes jagged with exposed plane edges?)


#15

Здравствуйте. Как дела?
I’ll dig it out later (it’s in the attic and I have to go out today) but I would be surprised if you found it online.
Tungsten and its compounds, G D RIECK, Pergamon 1967. Now I check, I think the full gory details of tube heaters is in one of the references, but as I don’t have access to that academic library anymore, and as this is in any case totally off topic, it’ll have to rest there.


#16

Спасибо, хорошо. :smile:

Book located (apparently it is somewhat well known reference? Older references are often found online, scanned and digitized, sometimes even OCRed and proofread!) and moved to my personal on-hand library. Looks interesting.

Todo: a wikipedia article about tungsten filaments. Will be a handy common material for incandescent tubes, vacuum tubes, hot cathodes, and maybe some other articles…


#17

Не за что. I hope the article goes well.


#18

This topic was automatically closed after 5 days. New replies are no longer allowed.