Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Knotty: A domain-specific language for knitting patterns

t0mpr1c3.github.io

Post reply on HN
Print view View original

21–30 of 57 posts

Re: Knotty: A domain-specific language for knitting patterns

#21

Post by pantulis »

bregma wrote: post #18
zokier wrote: post #11

Earlier quoted context omitted.

Some would be wrong. While the use of punch cards was significant, I struggle to see jacquard looms doing computation in any meaningful sense.

From a computer-science theoretic view, any any recursively enumerable transform is computing. It's pretty easy to see the translation between a Jaquard loom and an abstract Turing machine. Just because it smells of machine oil and lanolin instead of ozone and magic smoke doesn't mean an automated loom and a modern computer are not essentially the same device, mathematically speaking.

> Just because it smells of machine oil and lanolin instead of ozone and magic smoke

The magic smoke I've smelled before specially with a burnt cap but do computers really smell of ozone???

Re: Knotty: A domain-specific language for knitting patterns

#22

Post by dleeftink »

zokier wrote: post #11
haunter wrote: post #10

Some say the Jacquard machine was the first real computer using punched cards https://en.wikipedia.org/wiki/Jacquard_machine

Some would be wrong. While the use of punch cards was significant, I struggle to see jacquard looms doing computation in any meaningful sense.

Our modern-day nomenclature owes a lot to the 'arrays' and 'threads' that made up these weaving machines (see Howard Rheinhold's Tools for Thought)[0].

Evidently, the punch cards that stored Jacquard's weaving patterns were a direct inspiration for Babbage's analytical engine.

[0]: https://www.rheingold.com/texts/tft/02.html#:~:text=The%20we...

Re: Knotty: A domain-specific language for knitting patterns

#23

Post by eru »

bregma wrote: post #17
zokier wrote: post #12

Earlier quoted context omitted.

Worth noting that weaving and knitting are two different production processes.

Yes and no. The both involve implementing a sequence of instructions to perform mechanical motions on fiber to form textiles. It might be moving frames to open the shed for weaving, or it might be opening hooks to chain for knitting. The fiber used and the properties of the produced textile may have different properties and the machinery may differ in detail but the processes are essentially identical.

That's a bit like saying that welding and screwing pieces of metal together are 'essentially identical'.

Re: Knotty: A domain-specific language for knitting patterns

#25

Post by zokier »

bregma wrote: post #18
zokier wrote: post #11

Earlier quoted context omitted.

Some would be wrong. While the use of punch cards was significant, I struggle to see jacquard looms doing computation in any meaningful sense.

From a computer-science theoretic view, any any recursively enumerable transform is computing. It's pretty easy to see the translation between a Jaquard loom and an abstract Turing machine. Just because it smells of machine oil and lanolin instead of ozone and magic smoke doesn't mean an automated loom and a modern computer are not essentially the same device, mathematically speaking.

> It's pretty easy to see the translation between a Jaquard loom and an abstract Turing machine

I'm not really seeing that translation, could you expand how you envision that working?

Re: Knotty: A domain-specific language for knitting patterns

#28

Post by fipar »

zokier wrote: post #25
bregma wrote: post #18

Earlier quoted context omitted.

From a computer-science theoretic view, any any recursively enumerable transform is computing. It's pretty easy to see the translation between a Jaquard loom and an abstract Turing machine. Just because it smells of machine oil and lanolin instead of ozone and magic smoke doesn't mean an automated loom and a modern computer are not essentially the same device, mathematically speaking.

> It's pretty easy to see the translation between a Jaquard loom and an abstract Turing machine I'm not really seeing that translation, could you expand how you envision that working?

What I remember from school (late 90s) is that Babbage was inspired by Jacquard’s punched cards to use then in his Engine, and either this, or Jacquard directly, inspired the same in Hollerith. I don’t recall it as there being a direct line in terms of “modified loom is general purpose computer) but it was certainly an important influence.

In hindsight it seems easy for some of us to make the connection, but, at the time, it must have been quite the breakthrough.

Re: Knotty: A domain-specific language for knitting patterns

#29

Post by bluGill »

eru wrote: post #23
bregma wrote: post #17

Earlier quoted context omitted.

Yes and no. The both involve implementing a sequence of instructions to perform mechanical motions on fiber to form textiles. It might be moving frames to open the shed for weaving, or it might be opening hooks to chain for knitting. The fiber used and the properties of the produced textile may have different properties and the machinery may differ in detail but the processes are essentially identical.

That's a bit like saying that welding and screwing pieces of metal together are 'essentially identical'.

If you are drawing up blueprints for a machine the two are essentially identical. You choose whichever makes sense for the joint you need, but they are essentially identical on the blueprint.

If you are making the card reader for automated machines, knitting, weaving, and playing a piano are virtually identical - you just move some lever in response to a hole. Someone working on a different part of the machine cares about the difference, but to the card reader they are identical.

Re: Knotty: A domain-specific language for knitting patterns

#30

Post by bluGill »

delaminator wrote: post #13
haunter wrote: post #10

Some say the Jacquard machine was the first real computer using punched cards https://en.wikipedia.org/wiki/Jacquard_machine

Like many an invention, he was standing on the shoulders of others Basile Bouchon (1725) created one of the earliest automated looms using perforated paper tape to control the weaving pattern. Jean-Baptiste Falcon (1728) improved on Bouchon's design by using chains of punched cards instead of continuous paper tape, making the system more durable and easier to handle. Jacques Vaucanson (1745) further refined the conce…

Those systems likely took inspiration from music boxes from long before.
Post reply on HN