Re:Penrose: Create diagrams by typing mathematical notation ...
by alanbernstein on Sunday May 10, 2020 @08:53PM
(#23137807)
Penrose is a neat name for this. At first I thought it was not a great choice because of other associated concepts. Then I noticed pen=write, rose=draw. Maybe not intentional, but cute.
Re:Penrose: Create diagrams by typing mathematical notation ...
by nikofeyn on Monday May 11, 2020 @03:57AM
(#23139833)
roger penrose is a famous mathematician, who did a lot of work in mathematical physics. he is also famous for a highly visual approach, such as penrose tiles.
Re:Penrose: Create diagrams by typing mathematical notation ...
by iainmerrick on Monday May 11, 2020 @04:59AM
(#23140165)
Just to close the loop here, Penrose is also famous (to his readers, anyway) for his beautiful hand-drawn diagrams and illustrations. So it’s a terrific name for this project.
Re:Penrose: Create diagrams by typing mathematical notation ...
by jcranmer on Monday May 11, 2020 @09:43AM
(#23142022)
And has five-fold symmetry, which is impossible for wallpaper groups.
The discovery of Penrose tilings is partially tangled up with the discovery of quasicrystals, where researchers found minerals with 5-fold and 10-fold symmetry that ought to have been mathematically impossible to exist. Turns out that the Crystallographic Restriction Theorem has an important caveat that people neglected up until the 1980s: it assumes that you have to have translational symmetry, which Penrose tilings do not.
Re:Penrose: Create diagrams by typing mathematical notation ...
by alanbernstein on Monday May 11, 2020 @12:06AM
(#23138820)
In my mind, it was obvious that I was thinking of this: https://mathworld.wolfram.com/Rose.html. Clearly I should have mentioned that. It's the type of shape drawn by a spirograph, a pen-rose.
Re:Penrose: Create diagrams by typing mathematical notation ...
by chii on Monday May 11, 2020 @12:17AM
(#23138859)
penrose is the name of a famous mathematician/logician, so it's a double entendre on both it being for drawing (the pen and the rose), as well as it being for maths.
It's not based just on text code, but on interactive manipulation as well, so it's not like e.g. TikZ which can be kept in the document source and version controlled, etc.
Re:Penrose: Create diagrams by typing mathematical notation ...
by maxkrieger on Sunday May 10, 2020 @10:01PM
(#23138187)
This is true but only at present.
An analogy would be the weights of pretrained machine learning models. They're still serializable, though it's true they're less semantically trackable than a well-written TikZ snippet (which, by the way, I haven't seen many of in practice!)
"Designers write programs that produce figures described in terms of relationships. These relationships are compiled down to constraints, which are then solved automatically."
Re:Penrose: Create diagrams by typing mathematical notation ...
by Koshkin on Sunday May 10, 2020 @08:29PM
(#23137646)
It is an interesting approach to drawing - it’s almost like what you do is, you make the tool understand the meaning of what you want to draw, and the tool is expected to do an optimal job at getting it done.
Re:Penrose: Create diagrams by typing mathematical notation ...
by crispyambulance on Monday May 11, 2020 @01:25PM
(#23144600)
That's nice.
It looks like a cool project and they say it's still in early stages. It's a pity they don't put pictures on the splash page. Pictures communicate better than anything else what the capabilities of the tool are.
Re:Penrose: Create diagrams by typing mathematical notation ...
by egwor on Monday May 11, 2020 @08:40AM
(#23141435)
I think a README.md with the render in each directory would be useful or maybe one README.md in the examples dir that refers to each example with a bit of context of why it might be something to do/use would suffice?
Looks cool, though.
https://dglittle.github.io/cdn/penrose002.html
It's an infinite plane of tiles, procedurally generated, that never repeats no matter how far you go.
The discovery of Penrose tilings is partially tangled up with the discovery of quasicrystals, where researchers found minerals with 5-fold and 10-fold symmetry that ought to have been mathematically impossible to exist. Turns out that the Crystallographic Restriction Theorem has an important caveat that people neglected up until the 1980s: it assumes that you have to have translational symmetry, which Penrose tilings do not.
The notation feels c like.
Maybe it could be interesting for knots and complicated algebraic geometry?
It's not based just on text code, but on interactive manipulation as well, so it's not like e.g. TikZ which can be kept in the document source and version controlled, etc.
An analogy would be the weights of pretrained machine learning models. They're still serializable, though it's true they're less semantically trackable than a well-written TikZ snippet (which, by the way, I haven't seen many of in practice!)
"Designers write programs that produce figures described in terms of relationships. These relationships are compiled down to constraints, which are then solved automatically."
Discussed here: https://news.ycombinator.com/item?id=21774018
Will we never learn??
https://github.com/penrose/penrose/tree/master/examples
The thing makes graphs for Christ's sake.
Your github link is just a dump of source code. One would have to go through a lot of steps to actually see what they even render.
It has rendered examples for all types of graphs.
It looks like a cool project and they say it's still in early stages. It's a pity they don't put pictures on the splash page. Pictures communicate better than anything else what the capabilities of the tool are.