Forgive me if there’s an obvious answer I can’t figure out. I will probably facepalm myself.
We are surrounded with straight lines and flat surfaces: your computer screen, IKEA furniture, bathroom tiles and so on. But how they were made? The factory machines must have been made by other factory machines, and so on. Where is the origin of all that?
I can’t think of any straight lines appearing in nature, save for the horizon line, Sun rays and a frozen lake, perhaps.
A perfect circle is much more achievable; it’s relatively easy to make a compass or a spinning wheel for clay pottery (although manufacturing of a perfect sphere is beyond me).
I’m aware Ancient Greeks were using a straightedge. But when did it all started? How human society - in the Stone Age, perhaps - managed to make something perfectly flat by using stone tools and bones?
By “perfectly” flat I mean reasonably flat, not on a microscopic level.
And this begs another question: what is the most perfect flat surface/straight line we are able to make nowadays (using laser technology and whatnot)? I mean material object, not the laser beam itself?
A straightedge can be made “from scratch”, as long as you’re okay with making 3 of them. By checking different pairs of 3 straightedges against each other (or rubbing them together in a process known as lapping), they will all eventually acquire straightness. If you use only 2 edges rubbed against each other, they will mate perfectly, but no necessarily along a straight line. See this page for a description: http://home.comcast.net/~jaswensen/machines/straight_edge/straight_edge.html
Creating flat plates is similar…it involves using a trio of plates rubbed in two dimensions until they are all flat.
So basically how could someone have created a straight line or flat surface early on?
Dig a long trench, fill it with water, let it come to rest. There’s a flat, level surface that you can use to mark off the foundation of your (building, pyramid, etc).
Or pull a string tight and mark along it. Nice straight line. Cut along the line carefully.
Take two reasonably flat surfaces (e.g. two clay bricks) and rub them against each other, they should wear down to a couple of nice flat surfaces.
Once you’ve got one flat reference it’s much easier to make more of them.
The surface of still water is for most intents and purposes flat. Since the middle of the last century window glass has been made flat by floating the molten glass on bath of molten tin.
It’s pretty easy to make flat glass by spreading molten glass over a liquid:
Also, a string with a weight on it will make a perfectly straight vertical line.
As for the flattest surface ever made, a quick Google search brings up this articlefrom 1998 that says that it would be achieved in the next 5 years (don’t know if it happened; however, they recently made the most perfect sphere, so that should be possible):
Carpenters today still set out a straight line using a string and powdered chalk.
A snap line uses a string coated with powdered, usually colored chalk. The string is stretched taut, then you pull up on the line ‘snapping’ a mark in whatever you intend to cut or attach something to.
Example #1: In the mid-70’s, when Pascal was the predominant Computer Science Religion, it was an article of faith that Pascal compilers were written in Pascal. According to my then-roommate who was a student in that field, it was a blasphemy, and indeed it was even beyond imagination, to suggest or question that anyone would write a Pascal compiler in any language other than Pascal.
So how was the first Pascal compiler written? Well, in Pascal, apparently.
(Refugees from that era may remember that one of the earliest implementations was for the Control Data 6000/7000 series machines, and that it was so horrendously awful that a whole generation of Computer Science students’ minds were poisoned into believing that Pascal was surely a Gift From The Gods, in order to rationalize in their minds why they would ever use such an abomination.)
Example #2: I recently picked up a stack of old (but possibly still useful) computer programming books at a thrift store for 10 cents each, including SQL Visual Quickstart Guide by Chris Fehily (Peachpit Press, 2005). In the introduction, on page xviii he lists topics not covered in the book:
(What’s that leave to cover anyway? This book is 450 pages long.)
The footnote clarifies why he doesn’t cover recursion:
It’s straight edges, factory machines, and recursion all the way down.
As a side note to Michael632129’s post about float glass, an engineer I used to work with left the company briefly to work for a glass manufacturer.
He said that due to the energy costs of heating up the float metal, that if something broke down on the packaging/finished product end of the production line, they had an arm that would just come out and start breaking the glass off the end of the line.
It was easier to recycle the broken glass because all of the raw materials and most of the energy, had been used to make it the first time, than to shut down the whole process and start it up again.
Thanks for that link, Dave. Very interesting read.
There’s a book title that immediately caught my attention…“Foundations of Mechanical Accuracy”. (if I’ll be able to find it somewhere with a wallet-friendly price).
I’m little skeptical about rubbing two stones together; my muscle movement is not uniform - a surface would be smooth, but probably slightly bulging. However, other techniques mentioned here are really ingenious…and when I see some things still working perfectly for carpenters and tailors after all these years of human progress, somehow it makes me feel glad.
The trick to producing a flat plate is making them in threes per aerodave’s post. Rubbing two surfaces together until they fit perfectly only guarantees spherical surfaces, with the radius depending on how you moved them. Only when you iterate between three do you drive the radius to infinity and so generate flat surfaces.
With no snarkiness intended to the OP, can I just say that this thread had produced some of the most interesting content from the most unpromising of starting questions.
