Rank: Member Groups: Registered
Joined: 14/12/2012(UTC) Posts: 23 Location: Neverland
Was thanked: 15 time(s) in 9 post(s)
|
Originally Posted by: kilele ...surely other fellows can give far better suggestions, maybe RisingEagle or even Andrey Wow, Kilele, to put me in the same phrase as Andrey is waaay beyond what I deserve. I'm not even in the same solar system as he is. Be that as it may, how can I refuse such a flattering invitation. And thank you. I must admit I haven't given this issue a lot of thought, but one thing I have been considering on a conceptual level that might work is that vectors and physical units are both attributes we assign to scalars. Andrey has built a wonderful software infrastructure for tracking and calculating physical units, vis, filling in the oval to specify one's personal choice in physical units and applying automatic unit conversions. Can this same machinery be used to specify vector, matrix, or other complexity with additional symbols? So if I want to override meters per second with millimeters per second and set it up as a 5 dimensional vector I would fill in the oval as v5;mm/s. A matrix might be m3x2;mm/s. For scalar arrays that contain mixed dimension such as spacetime 4-D vectors (3 space, one time), a more complex encoding would be required, but shouldn't be that difficult to design. Instead of one oval, perhaps Andrey could set up a second oval of a different color for specifying any scalar array type specification separately from any physical unit: scalar/vector/matrix/tensor/complex-number. As for fonts, bolding, color, decorator markings, etc., used to represent the different scalar array types, that is rather an arbitrary choice as long as everybody likes it. Beyond that, Andrey would need the software to track allowable operations between any two scalar arrays, each of arbitrary type. Such rules might be user cusomizable as well. For example, some vector spaces have inner products and metrics, others do not. The user should be able to disallow such operations between two vectors from a vector space that does not. If I have other ideas, I'll submit again. I am greatly impressed by the continuing hard work and cleverness of the smath community including, but definitely not limited to, Radovan, Uni, Martin, Kilele, Web Surfer, Ber7, and of course Andrey. I commend you all. Those of you I have failed to mention, it is not your shortcoming, but my own poor memory. Edited by user 14 June 2013 09:01:47(UTC)
| Reason: Not specified
|
1 user thanked Rising Eagle for this useful post.
|
|