• CliffStoll a day ago

    Outstanding work and a delightful read.

    • chriskw a day ago

      Thanks Cliff, it means a ton coming from you! The videos from you and all the other folks on Numberphile always inspired me to see the beauty in math growing up :)

      • speeder 12 hours ago

        Please you two, make an awesome YouTube vídeo out of this. It is fascinating and beatiful and deserves a chance to viralize a little :)

      • sakesun 17 hours ago

        Wow

      • mckeed 3 hours ago

        Fun post! I drew the first 5 iterations by hand myself and I'm finding it easiest to think of as a self-similar coloring of a square tesselation.

        If you start with the shape of iteration 3, it tessellates as a 5x5 square tile. Make an infinite grid of those tile shapes with one iteration 3 version in the center. Treat that center tile as the center square in the iteration 3 pattern and color the tiles around it according to how the 2nd and 3rd iterations were built of squares. This gives you the 4th and 5th iteration and you can continue to iterate on the coloring outwards to color the grid of tiles in the wallflower pattern.

        • nico a day ago

          Amazing insightful and thoughtful write up, thank you!

          Loved the 3d visualizations

          It reminds me of this thing I built some time ago while playing with recursive decimation to generate effects similar to fractals from any image

          You can play with it here: https://jsfiddle.net/nicobrenner/a1t869qf/

          Just press Blursort 2x2 a couple of times to generate a few frames and then click Animate

          You can also copy/paste images into it

          There’s no backend, it all just runs on the browser

          Don’t recommend it on mobile

          • Iwan-Zotow a day ago

            Curious if it would work in 3D

            • nico a day ago

              Very interesting! I wonder what that would look like

              Right now, roughly, the algorithm recursively divides the image by doing decimation (ie. picking every other pixel), and keeps the decimated pixels as a second image

              Not sure how that algorithm would apply to a 3d data structure

              Do you know how 3d objects/images are usually represented?

              It would be cool to recursively decompose a 3d object into smaller versions of itself :)

          • Cogito 15 hours ago

            Thought I'd check the arithmetic for 2 two-digit numbers, and it works!

            I expect 41+14 to be 12 (two right plus two up equals two right and two up).

            Long addition in long form below uses:

            '=' to show equivalent lines (reordering of terms (1+2=2+1), spliting numbers (41=40+1), adding single digits (1+4=22))

            '->' for when the algorithm gives a digit

            '<' for when we move over a column

                41+14
                = (40+1)+(10+4)
                = 40 + 10 + (1+4)
                = 40 + 10 + 22
                -> 1s digit = 2
                < 4 + 1 + 2
                = 22 + 2
                = 20 + 2 + 2
                = 20 + 41
                -> 10s digit = 1
                < 2 + 4
                = 0
                -> done
                == 12
            
            [edit] Just noticed the article has two different numbering systems, one where 10, 20, 30, 40 are clockwise and one where they are anticlockwise. In both, 1, 2, 3, 4 are clockwise. My addition is on the second, where 10s are anticlockwise (this is what is used in the addition table).

            It still works in the alternative system (14+21 should equal 12)

                14+21
                =10+20+42
                ->2
                <1+2+4
                =13+4
                =10+3+4
                =10+31
                ->1
                <1+3
                =0
                ==12
            • taeric a day ago

              Holy cow, I was expecting a quick read. Wound up having to skim some, as I need to get some work today. Will be coming back to this to play with some. Really well done!

              • baq 14 hours ago

                This went much deeper and harder than expected. One has to admire the dedication.

                Question to the author: what would you recommend to hang on my kid’s wall today?

                • chriskw 13 hours ago

                  I'm by no means a parenting expert, but my answer would be anything related to something they feel passion or wonder for in the moment. I snuck in a paragraph near the end about burnout. At the root of the problem for me was that I lost the feeling of fascination and curiosity I had for math and programming, and doing this write-up helped me tap into that feeling of childlike wonder that used to come easily.

                • leni536 a day ago

                  Got a bit nerd-sniped by this and came up with an L-system that fills out (I think) "the wallflower":

                  https://onlinetools.com/math/l-system-generator?draw=AB&skip...

                  edit: On second thought, this probably generates the other fractal, but I'm not sure.

                  • leni536 10 hours ago

                    Found a space-filling curve for the wallflower:

                    https://onlinetools.com/math/l-system-generator?draw=ABCD&sk...

                    The previous one fills out the Koch island.

                    • chriskw 3 hours ago

                      That's really cool! I tried to get something to work last week on pen and paper but couldn't get anything to stick. Is there a strategy you used or did you just go by feel?

                      Edit: just noticed how you encoded a flip (AB <--> CD) between iterations like how the matrix flips the orientation of space. Super neat!

                      • leni536 33 minutes ago

                        > noticed how you encoded a flip (AB <--> CD)

                        Exactly! There is also a less obvious relationship between A and B too: B is a A "backwards" (A rotated 180°, starting the curve from the opposite end).

                        The strategy was to put 5 lines on the plus sign on the sides of the 5 cells, with the idea that each line eventually fills out a neighboring cell in subsequent iterations. I found one such path that had a chance of working. Not sure if this makes sense.

                  • CBLT a day ago

                    Well written! Would you mind sharing how you came up with the "middle out" numbering system? I can never seem to come up with something this inspired when I'm doing math problems by myself.

                    • chriskw a day ago

                      The post presents it a bit out of order, but it was mostly from realizing at some point that the way the fractal grows by a factor of 5, base 5 number systems, and the "spiral" mentioned in the post can all fit together. I also thought a lot about how to programmatically draw the fractal and a natural way would be to start from the middle and zoom out.

                      There's an apocryphal story about Richard Feynman about how he used to keep a dozen or so random problems in the back of his mind and made a little bit of progress on them every time he saw a connection, until finally he'd solve one and everyone would think he magically figured it out instantly. This was a bit similar except I'm not nearly at that level and I've only been able to do that for one problem instead of a dozen.

                    • wistlo 4 hours ago

                      This is so much better than reading the news.

                      Favorited—I'll be coming back to absorb more, as my aging semi-fluency in engineering physics and SQL doesn't help much with the notation I last saw in the 1980s.

                      • Tade0 12 hours ago

                        > Deciding to delegate to a future version of me that knows more math

                        Relatable. Huge part of my decision on what degree to pursue was a list of problems (mostly linear algebra) I needed to solve, but didn't have the guidance (and internet connection) to.

                        • tcshit a day ago

                          Nice writeup! I was hoping to see a photo of the fractal on your wall.. Nice link to Knuth video that I somehow have missed.

                          • leephillips 21 hours ago

                            Isn’t that it on the left in the last image?

                            • tcshit 14 hours ago

                              Yeah, maybe it is. It would be cool to make it much bigger, frame it and put it on the wall. Or create a mosaic tiled artwork, similar to Knuth’s dragon curve wall.

                              • chriskw 13 hours ago

                                Yeah, it's in the last image and in the thumbnail at very top (which I realize now is really hard to spot on mobile), intentionally not in the spotlight to leave space for the twist at the end.

                                https://chriskw.xyz/images/fractal/thumbnail.jpg

                                I think it would work perfectly as a mosaic eventually, but for the time being I'm perfectly content with the "rustic" 8x11 graph paper sized one taped to the wall. Currently planning to put up a slice of the orthotopeflower as a companion piece once I find matches for the colored pencils I used back then.

                          • Scene_Cast2 a day ago

                            I wonder if something similar can be applied to get a dither pattern with built-in level of detail adjustment.

                            • kragen 10 hours ago

                              This is beautiful. Thank you.

                              • 867-5309 11 hours ago

                                well, that escalated beautifully

                                • cies 14 hours ago

                                  I had this one up the wall (giant print) at a place I worked:

                                  https://raw.githubusercontent.com/cies/haskell-fractal/refs/... [17MB, sorry Github]

                                  It contains the Haskell code that produced it: https://github.com/cies/haskell-fractal

                                  Especially the `sharpen` function was interesting to come up with (I used some now-offline tool to do curve fitting for me): https://github.com/cies/haskell-fractal/blob/master/fractal....

                                  Fun little project. :)

                                  • cess11 a day ago

                                    Nice writeup. The Heighway dragon of Jurassic Park fame is pretty neat too.

                                    https://en.m.wikipedia.org/wiki/Dragon_curve

                                    • entropicdrifter a day ago

                                      Kinda looks like a propeller

                                      • shermantanktop a day ago

                                        Things with four arms that all curve the same way unfortunately tend to look swastika-ish.

                                        • leni536 a day ago

                                          The the arms of the author's "wallflower" fractal don't seem to curve, as opposed to the other, similar fractal (quadratic von Koch island). Which can be explained by each iteration adding a mirroring.

                                          • winnit 11 hours ago

                                            The unfortunate thing here is that the swastika was appropriated by a genocidal regime. The symbol still has a totally different life in India and Japan.

                                        • bdamm a day ago

                                          That was fun.

                                          • matt3210 16 hours ago

                                            Now make a tiling game engine that uses these!

                                            • mathfailure 7 hours ago

                                              Too much math.