3 ms·
Someone else may have said this but strictly speaking breakfast is something like a cone in a vector space, unless you want to explain to me how to eat negative
by ccppurcell 7mo ago
Someone else may have said this but strictly speaking breakfast is something like a cone in a vector space, unless you want to explain to me how to eat negative eggs.
- graypegg 7mo agoChicken maybe? It would represent an egg that did exist, but now doesn't.
- TeMPOraL 7mo agoAlso, in the other direction in space time, it's an egg that could have been, but now won't.
- 1659447091 7mo agoNow I want some chicken & waffles https://en.wikipedia.org/wiki/Chicken_and_waffles https://en.wikipedia.org/wiki/Chicken_and_waffles
- ccppurcell 7mo agoIn the breakfast space, I'm afraid, chicken is orthogonal to eggs.
- porphyra 7mo agoMaybe the axes are actually logarithmic so 0 is 1 egg, 1 is e eggs, -1 is 1/e eggs, etc.
- asdff 7mo agoEasy. Start eating two eggs a day. Then one day, eat three eggs taking the extra egg from tomorrows two spare eggs left in the carton.
- moron4hire 7mo agoWhat is menstruation but negative egg entering the body?
- anon_cow1111 7mo agoI think I will attempt to just eat the negative eggs because at least I recognize, and can define what both "negative" and "eggs" mean. Can't say the same for literally half the words in the OP's graph.
- DonHopkins 7mo agoThese are a Barycentric Coordinate System, so the weights on the targets are normalized, must sum to one, and must be positive. https://en.wikipedia.org/wiki/Barycentric_coordinate_system https://en.wikipedia.org/wiki/Barycentric_coordinate_system Barycentric coordinates are the local coordinate system inside a simplex. A simplicial complex is what you get when you glue multiple simplices together along shared k-faces for k = 0 … n -- vertices (0-faces), edges (1-faces), triangles (2-faces), tetrahedra (3-faces), and higher-dimensional faces -- to form a larger state space. https://en.wikipedia.org/wiki/Simplicial_complex https://en.wikipedia.org/wiki/Simplicial_complex It's not possible to have negative eggs, but you can apply the same machinery to many other things, like facial animation mesh blend shapes (Apple ARKit, Blender Blend Shapes and the FaceIt plugin, Unity SkinnedMeshRenderer, etc), where weights are often allowed to be overdriven >1 or even underdriven <0 for exaggerated or monstrous effects. Nouveau Art Pipeline Demo: Blend Shapes: https://youtu.be/phM8Wnzs_-g?t=104 https://youtu.be/phM8Wnzs_-g?t=104 (None of Epstein's spiritually close friends and shameless guru confidants were harmed or embarrassed in this demo, alas.) Eric Hedman - "Doppel" Character Modelling with Blendshapes for Animation (ARKit): https://www.youtube.com/watch?v=6L7jtgRD5rs https://www.youtube.com/watch?v=6L7jtgRD5rs (Eric "Irk" Hedman designed and created the character animations and objects in The Sims 1, and as you can see is extremely skilled and delightful to work with! Hire him if you need professional high quality creative artwork and animation, and can pay him in bananas: https://erichedman.artstation.com/projects/8wJDgw https://erichedman.artstation.com/projects/8wJDgw ) Faceit: Facial Expressions And Performance Capture (Blender): https://superhivemarket.com/products/faceit https://superhivemarket.com/products/faceit Unity SkinnedMeshRenderer: https://docs.unity3d.com/6000.3/Documentation/ScriptReference/SkinnedMeshRenderer.html https://docs.unity3d.com/6000.3/Documentation/ScriptReferenc... Apple ARKit Tracking and visualizing faces: https://developer.apple.com/documentation/ARKit/tracking-and-visualizing-faces https://developer.apple.com/documentation/ARKit/tracking-and... ARFaceAnchor.BlendShapeLocation: Identifiers for specific facial features, for use with coefficients describing the relative movements of those features. https://developer.apple.com/documentation/arkit/arfaceanchor/blendshapelocation https://developer.apple.com/documentation/arkit/arfaceanchor...
- ccppurcell 7mo agoI was just referring to one of the early sentences saying breakfast is a vector space. If you allow arbitrarily many (say) eggs so that two eggs and one egg are different breakfasts, you get a cone inside R^d. If you normalise and consider the ingredients as fractions of a whole (so that 1 egg and 2 eggs are both represented by the 1.0 egg breakfast) you get this simplex structure and the coordinate system you mentioned. But that's still not a vector space as there are not inverses in general. At best it can be embedded in R^d.
- CoastalCoder 7mo ago> unless you want to explain to me how to eat negative eggs. I'm sure there's a good joke in there involving trans-dimensional vegans, but I'm still on my first cup of coffee.
- BigTTYGothGF 7mo agoAre you telling me this dimensional vegan is trans.
- vee-kay 7mo ago[dead]
- tclancy 7mo agoThere’s a guy in overalls in a Brooklyn bistro who used to scramble them but that was very late-summer 2025 and you’ll get funny looks for asking about them.
- felineflock 7mo agoI think it is all normalized into positive vector space [0, 1] using min-max.
- getnormality 7mo agoThe vectors containing the weights of each ingredient live in the positive orthant (x >= 0). When you divide by x by sum(x), the resulting vector p has p >= 0 and sum(p) = 1. That new sum constraint is how you get the simplex.