Ancient Babylonian Algorithms: The Earliest Programs | Hacker News Reader

Ancient Babylonian Algorithms: The Earliest Programs

historyofinformation.com

125 pointsrobbbminson56 comments

Link to Knuth pdf https://web.archive.org/web/20160303185340/https://steiner.m...

The more I read about math history the less I care about electronic computers. There was a lot of interesting problems and abstractions even 400 years ago. I find it funny that these people had no choice but to understand large and complex with nothing but pen and paper.. and it seems like the less computing power the smarter the thinking.

> I find it funny that these people had no choice but to understand large and complex with nothing but pen and paper.

Some took it even further. The reason Plato gives for why Socrates wrote nothing down is that writing is an aid to memory and a form of crutch for the mind. And just as walking around on crutches will weaken a healthy man’s legs over time, so will leaning on writing weaken his memory.

right i remember this, it was framed as a history of anti-tech and purism .. in any case I think most brains, even the smartest will need a bit of storage to avoid looping around old ideas that were proven wrong hence paper

The alternative view is that the right tools are more like a bicycle and less like crutches. The art is knowing which is which I suppose.

Was your use of this example a reference to Steve Job's referring of the computer as the "Bicycle of the mind"? I think he might have been quoting Stewart Brand.

Learning by Shipping article claims Jobs credits "March 1973 Scientific American." [0, 1]

Jobs says it in the presence of Brand in this Wired book excerpt [2], to which Ken Kesey [!] replies "It’s the next thing after acid!" [3]

0. https://medium.learningbyshipping.com/bicycle-121262546097

1. https://www.jstor.org/stable/24923004?seq=1#metadata_info_ta...

2. https://www.wired.com/story/how-doug-engelbart-pulled-off-th...

3. https://en.wikipedia.org/wiki/Ken_Kesey

More zeppelins than bikes.

Definitely extending the usefulness of the analogy, but when I am riding my bike a lot and then take a break for repairs and walk more instead, I quickly get sore. Even after walking distances that I dont normally get sore from if I am bike free for a while and walking more regularly.

So can we test against an illiterate population and test their memory, recall and problem solving techniques ?

There have been such studies* of at least the first two, showing that pre-literate people are far better at memorizing (e.g., long epic poems) than literate people.

* Yes, citation needed. Searching...

How Do We Know That Epic Poems Were Recited from Memory? https://daily.jstor.org/how-do-we-know-that-epic-poems-were-...

Previously on HN: https://news.ycombinator.com/item?id=14640519

They were not "recited from memory" as we would understand that phrase today, though. Oral epic poets do not use the same wording every time they deliver the same poem.

There are a lot more people today that do think very hard to come up with smarter algorithms. Most likely more than at any point in human history.

Yes, because technical interviews select for people who can do it on the fly :-p

People forget that prior to electronic computers we had mechanical computers: abacus, slide rules and other devices.

Slide rules go back all the way to the 17th century, and the abacus goes back thousands of years.

Romans engineers, for example, were known to carry portable hand abaci.

Slide rules are actually my favorite thing.. two pieces of wood to leverage logarithmic relationships (quite abstract IMO) and compute faster. Just through gradual marks and a few concepts.

I’m with you on “just” for “a few concepts”, but not for the marking:

Slide Rule for the Modern Day:

https://www.instructables.com/Slide-Rule-for-the-Modern-Day/

I mean, the “knowledge” of where the marks go is not the same as the “know-how” required to put them there ;-)

Alright it's indeed subtler than expected. Still a seemingly looking thing that is actually packed with mathematics. The point was a bit that sophistication doesn't come from the underlying technology.

What I like about slide rules is that they're not a black box. You can see it and manipulate as a physical object and you can develop an intuition around it.

Greeks had a mechanical cube-root finder.

https://www.jstor.org/stable/3615516?seq=1

(btw. are scihub links allowed on HN?)

A 1979 article on ancient catapults, includes the description of the cube-root finding device:

http://jaanmarss.planet.ee/juhendid/Kaugrelvad_Antiik-Kreeka...

You can easily use any checkered board as a binary calculator. I'm quite convinced that all the games like chess started life as things to do on your calculator when you were bored.

I agree. The over reliance on technology is making us less aware of our innate abilities to reason things out.

Easy access to information might make us "smarter" in a sense that we know how to find this information. But let's face it, we don't actually know how to find stuff. Google does that.

We just type simple questions that any 10 year old can do.

Another point, I have diplomas, am patient, like precision... then I watched youtube videos about electronics, woodwork etc.. and I saw how things are done. I expected easy transfer, yet when I tried to do them I struggled.

The reading knowledge is not know-how.. these are two distinct categories. The web gives you metadata in a way.. but so far it seems only experience gives you know-how. You can see all the diagrams about how to cut wood straight .. but until you saw all the ways you can fail your edge, curve your blade .. it's not 'can do' knowledge, only potential.

And I found this in dev too.. I've done tutorials, moocs, books about non trivial stuff.. but making an app was a totally different kind of effort. It wasn't intellectually hard, just different and you don't know that until you do it. Like hundreds of stupid details to juggle and go around to finish something properly.

The gulf between theory and practice is larger in practice than in theory.

maybe we should rename some 'theory' classes as 'pretty ok approximation of future problems'

Love it! I just googled that quote and you seem to have originated it. I have a little list of quotes and their authors. Do you mind revealing your name so I can attribute it? Or emailing me to the address in my profile?

It's just a modification of this quote: https://pubmed.ncbi.nlm.nih.gov/24320775/

How did I miss that. Thank you very much

"Knowledge is only rumor until it's in the muscle." (I've heard that that is a New Guinea proverb but I don't know for sure.)

The more you learn, the more the surface area of your inquiry expands. Each answer Spurs several new questions. A 20 year old will have far more sophisticated questions they want to ask and problems they need to solve than they would have had at 10.

That link only got me the first page (on mobile Safari). This link has entire pdf: http://www.realtechsupport.org/UB/NP/Numeracy_BabylonianAlgo...

Writing things on paper is just a term rewriting system [0] which is just another universal computational system.

[0] https://en.wikipedia.org/wiki/Abstract_rewriting_system

Turing was explicitly trying to abstract from the sort of symbol manipulation mathematicians of his time did with pen and paper.

Presumably his angelic counterparts have no need of an infinite tape of finite symbols, and can use a single square with an infinite repertoire of symbols instead.

Turing was interested in mechanical manipulation. The church turning thesis showed that he neededn't have bothered since lambda calculus, a term rewriting system, is equivalent to a Turing machine and dozens of other methods of effective computation. You can pick which ever is the most aesthetically pleasing.

Wonder if Knuth’s work in 1972 influenced Neal Stephenson’s “Snow Crash” in 1992? Certainly seems like it did!

Solid essay on ancient myth and “Snow Crash” here: https://www.cs.unc.edu/~taylorr/snowcrash.html

They also invented Fourier transform. https://en.wikipedia.org/wiki/Fourier_analysis

“A primitive form of harmonic series dates back to ancient Babylonian mathematics, where they were used to compute ephemerides (tables of astronomical positions).”

'This in the procedure' make it really sound like a 'END PROGRAM'

>One of the ways to help make computer science respectable is to show that is deeply rooted in history, not just a short-lived phenomenon. Therefore it is natural to turn to the earliest surviving documents which deal with computation, and to study how people approached the subject nearly 4000 years ago.

Interesting.

This is why I prefer the term Computing Science. Naming the field after the tool is like calling astronomy Telescope Science.

I never thought about it, this is a great observation!

This is actually a pretty funny fact that I think most, including myself, have never thought of.

CS in Swedish is usually “data science” (datavetenskap).

Some universities use “Department of Computing Science” as well as the English name.

Dijkstra: "Computer Science is no more about computers than astronomy is about telescopes." (Since nobody's said it explicitly and in case you were referencing it.)

That is, telescopes are, in practice, the most common way to learn about stars(/celstial bodies), but the stars are the target of study, with telescopes an incidental tool.

Computer science is about e.g. resource scaling for solving particular problems, with electronic computers just being one way those insights can be applied.

That Dijkstra's phrase is genius because at first sight it just seems a pedantic point made by the professor, a_boutade_ if you will, but the more you think about it the truer it is, even obvious.

Abelson makes this point in the opening seconds of his 1986 Structure and Interpretation of Computer Programs course.

Lecture 1 https://www.youtube.com/watch?v=-J_xL4IGhJA

He explains that Computer Science is a terrible name – it's not a science, and "not about computers in the same sense that physics is not really about particle accelerators, and biology is not really about microscopes and petri dishes".

Computer isn’t just the device though, it’s the activity. Before electronic computers, computer was a job, people were hired to be computers so it’s more like calling astronomy “Astronomer Science”. Still awkward, but not quite so silly.

Isn't that field that superset computer science called computational thinking? (which is quite equivalent to semi-formal applied problem solving)

Can someone clarify the calculation in the given example? What are stuff like “8,20”?

FTA:

> In this case, 50 stands for 5/6 and 8,20 stands for 8 1/2

Except this is clearly a typo (it's not clear whose), and the real values are 5/6 and 8 1/3 as mentioned in another comment.

So you win a point for demonstrating you read the article, and lose fifty for demonstrating you didn't grasp its meaning here.

Babylonians used a base-60 number system. Here is the wikipedia link: https://en.wikipedia.org/wiki/Sexagesimal

8,20 is 8+20/60 = 8 1/3

we do too. Though these days it’s mostly for time only.

“...and is still used—in a modified form—for measuring time, angles, and geographic coordinates.”

> One of the ways to help make computer science respectable is to show that is deeply rooted in history.

This is not a sure thing, by any means:

https://en.wikipedia.org/wiki/Oldest_profession_(phrase)

People do not automatically respect new professions.

Professions that very new are not respected due to lack of recognition; but some old professions are not highly respected either.

"When G*d created Eve from Adam's rib, he was doing surgery, and so surgery is the oldest profession" said the surgeon.

"Before that, He created the Heavens and the Earth out of Chaos, so architecture is the oldest profession" said the architect.

"Well, yes, but where do you think the Chaos came from?" asked the software developer.

Reply on news.ycombinator.com