h a l f b a k e r y
No serviceable parts inside.

meta:

account: browse anonymously, or get an account and write.

 user: pass:
register,

# magnitude mathematique

Is there natural stepping of numbers as magnitude changes?
 (+1) [vote for, against]

Mathematics has the rock solid unit that doesn't change in series.
1 in 1 2 3 4 .....1325432 1325433 .... But mathematics is a language, it's properties of the unit is singular, it abides by logical rules. There is nothing a about number has that says stop this is the limit.

In the real world, units are multifaceted, multidimensional so interactions are physically limited by type and structure of the subunit. There are set ranges that certain sized subunits can be viable.

I proposed taking a census of all things and trying to examine them as discrete subunits. This census would range from subatomic particles, atoms, repeating crystal structures, micro-organisms, fauna and flora, geological structures, planet/ star sizes , galaxy sizes.

Does the universe go 1 2 3 .. or 1 2 7 9 12 13 14 15 21 ...* Logically 5 exists but under construction rules of the universe not a magnitude subunit available.

The gaps might indicate the real world recursive physical patterns**. The standard movie goes linearly from atom to web of galaxies but I not too sure that is the case.

Maybe pounds, shillings and pence was on the right track.

* Totally made up, to illustrate the idea. ** patterns that come as the small is repeatedly stacked to the large, Or as with life, the subunit pattern that is stuck between the very small and very large recursive patterns.

 — wjt, Jan 13 2021

Fibonacci series in nature. [RayfordSteele, Jan 14 2021]

The Tukey Ladder of Powers http://onlinestatbo...rmations/tukey.html
For [pertinax] this is what I was crudely trying to describe - i.e. different degrees of curvature can suggest some underlying x^n relationship. Outside of this ladder of powers is the exponential relationship n^x which can look like an excessive version as n gets larger. [zen_tom, Jan 14 2021]

Battle Beneath the Earth https://www.imdb.co...87/?ref_=fn_al_tt_1
Prior Art [8th of 7, Jan 14 2021]

Regards maths content on YouTube, this channel has been really illuminating on a number of topics. Gives great perspective and context, whilst not shying away from (sometimes) quite technical detail. [zen_tom, Jan 14 2021]

E series of preferred numbers https://en.wikipedi...f_preferred_numbers
pick your level of precision, get a series of numbers [lurch, Jan 14 2021]

Explaining Benford's Law http://www.dspguide.com/ch34.htm

It's only a "Law" when it is applied to the right constraints... https://www.youtube...watch?v=etx0k1nLn78
[RayfordSteele, Jan 19 2021]

Well done, we're glad that's finally properly sorted out. Your Nobel Prize is in the post.
 — 8th of 7, Jan 13 2021

 There are lots of transformations you can do that will convert from observed data to a linear pattern. For things that grow monotonically with some degree of acceleration, this might crudely go from n^x, through a "ladder" of x^n forms (where n goes from some positive integer through fractions down to large negative integers) with logx in the middle.

 Then there are oscillating series, divergent or otherwise, and really interesting plays on complex series where rather than the number line being on a line, it gets expanded into the plane, from where you get really special spaces like the Riemann zeta function which both spirals chaotically, but also precisely spells out the locations of all the prime numbers, themselves which have the property that their nth member tends to lie close to series defined by n log(n).

 Other series of natural interest include the binomial series (itself generated or reducible from Euler's Gamma Function, which itself is the analog of the factorial function applied to non-integers) which governs discrete combinatoric affairs* and by extension a great deal of classic probability. Then there's Fibonacci series, and triangular series, each of which appear in nature in different places.

 Much of mathematics is astoundingly good at describing things we observe in nature, often doing so decades or centuries before anyone is able to make any actual observations or even conceive of them. Much of our understanding of nature today was explored mathematically long before it was ever observed.

 Regards the idea of looking at the structure of things at different levels of scaling - that's been very closely studied under the name of "scale invariance" and it's degree or strength (i.e. the amount of scaling you have to do to repeat a structure) is usually described as being of some non-integer dimension. So if a 2d curve is somehow self-similar (i.e. it exhibits sub-units that look like macro-units) then it will have an effective dimension of 2.x lying somewhere between 2 and 3. Different objects, systems and patterns exhibit different levels of this dimension, so the coastline of the UK is estimated to be something like 1.2 other objects might be more, or less squiggly.

* re combinatoric affairs - I'd suggest this is where the secret sauce of nested structure lies - one atom alone isn't very interesting, but when there exists a network of atoms all influencing one another through their network of interactions, you get the opportunity to create an interesting unit. When multiple networks of these molecular units interact through their networks, again a transcendent unit is created. This network effect where small, simple units interact across some network of relationships seems to be a common theme in nature, and again, there's a fair amount of interesting mathematics that covers this phenomena. Unfortunately, the numbers involved in calculating the possible connections (permutations) of a collection of n objects explodes into incalculably large values beyond relatively small collections, even say 20, after rounding produces something like 2.4 million billion permutations, and the more you add, the faster the numbers get bigger - few computers can deal with such calculations, even given time and lots of air-conditioning.
 — zen_tom, Jan 14 2021

 // Much of mathematics is astoundingly good at describing things we observe in nature, //

 ... except that due to neurophysiological structures, humans have an innate bias to "see" patterns - faces are the classic case - in what are actually truly random configurations, leading to the risk of self-fulfilling prophecies.

[zen], what you've written makes so much more sense than [wjt]'s original post- particularly about self-similarity persisting across micro to macro scales, as the Mandelbrot set demonstrates - that it's a pity that the two texts can't be exchanged ...
 — 8th of 7, Jan 14 2021

 Someone should spend some time with my favorite math channel gal, Vi Hart. Her explanations of interesting math things are quite good.

Nature is less random than you think. Leaf growth patterns are a function of simple growth hormone maximal locations, which by their usage and repetition make for some fascinating mathematical patterns. See link.
 — RayfordSteele, Jan 14 2021

I would just like to go on record to say that if it were possible to mentally download your understanding of the language of mathematics without having to go through all of the work it took for you to attain it... I would be good with that.
 — 2 fries shy of a happy meal, Jan 14 2021

 I understand patterns in nature have been found that have mathematical precis series or formulae like shell spirals and population interactions.

 We all came from the expansion, how all the tiny space-time subunits unfolded/congealed to what we have now must also have a skeleton mathematical framework. Definitely not a even constant continuum.

 And I don't think it is about repetition all the way through as combination effects can twist making a scale 'invariance' disappear. If the scalar invariance is more about lowest/stable mass/energy ledges as scale increases , then this has been thought about, and I concede. Looking out into the world, there seems a lot of empty space with a lot of embroidery about the edges in blobs.

 This more a general overview of whether matter/organization clumps, as scale increases, hence trying to shine a light the ultimate skeletal theory.

 [kdf] This poor written concept, did get a well written, informed text from [zen_tom].

Experimental measurements are the constraints of the mathematical language.
 — wjt, Jan 14 2021

 //Looking out into the world, there seems a lot of empty space with a lot of embroidery about the edges in blobs.//

Stop using a doily as a COVID mask.
 — pertinax, Jan 14 2021

 //if it were possible to mentally download your understanding of the language of mathematics without having to go through all of the work it took for you to attain it//

Sounds great! But I think those two clauses are more intimately linked than you suspect... a bit like trying to get from America to Europe without crossing the ocean...
 — pocmloc, Jan 14 2021

 There are the gifted, for whom the logic and symbols of the mathematical language fall into place because the subconscious does all the heavy lifting.

Of course, work is still needed to really utilize given talents.
 — wjt, Jan 14 2021

 // You would rather see nonsense in reply to a good idea than the other way around? //

Yes, if for no other reason than familiarity; we post excellent ideas, others post nonsense replies.
 — 8th of 7, Jan 14 2021

Or just get your bearings straight, and take the short route.
 — pertinax, Jan 14 2021

 //this might crudely go from n^x, through a "ladder" of x^n forms//

Could you provide an example, [zen_tom]?
 — pertinax, Jan 14 2021

 // Tunnel through the Earth. //



The most effective place for a fixed link is the Bering Strait; a tunnel has been discussed for some time.
 — 8th of 7, Jan 14 2021

I'm a Vi Hart fan as well, and if people are looking for other good maths content, I'd recommend 3Blue1Brown who is uniquely good at explaining horribly *horribly* tricky material in an intuitive, geometric way that in multiple cases has helped me understand (or feel as though I understand) things that hitherto been completely alien. I'm still rewatching his linear algebra series for maybe the 3rd or fourth time, and still finding it full of useful insight.
 — zen_tom, Jan 14 2021

You still cross beneath the ocean
 — pocmloc, Jan 14 2021

 On most of your planet, the antipode to any given position on land is quite likely to be ocean - easy enough to calculate with a bit of basic spherical geometry. So, wherever you move on land, you're likely to be crossing ocean, albeit with the thickness of the planet in between.

Unless, of course, you subscribe to the view that it's "turtles all the way down" ...
 — 8th of 7, Jan 14 2021

[wjt], if I'm understanding you correctly, then something you might be interested in looking at is called the "E series of preferred numbers". It's how resistors get their weird but consistent values.
 — lurch, Jan 14 2021

Thank you for the link, [zen].
 — pertinax, Jan 14 2021

I read the date of posting as "March". I'm going to pretend the post let someone make a time machine, and they went back in time to edit the post and keep anyone else from discovering it.
 — Voice, Jan 14 2021

 Thanks [lurch].

A coastline is more of part of subunit . Tetonic plates I would think might be a subunit but inter planetary data might be needed on that one. Each of the subunits is going to be physically controlled by multitude of variables and physical factors but all of those dimensions emerge from the ultimate machine's underlying metal.
 — wjt, Jan 15 2021

 [zen_tom] Have you ever heard of Benford's Law? The idea that most numbers, in general, start with a 1.

 The pattern is scale invariant, and base invariant - even in binary, or any base you like, most numbers start with a 1! The wikipedia page does an excellent job of only throwing further confusion on the issue, and explains nothing.

I can _very_ strongly recommend reading chapter 34 of the DSP Guide [link], where the author explains Benford's Law and provides an excellent explanation. It really is a magic trick, and says a lot about our human ability to see patterns everywhere.
 — mitxela, Jan 19 2021

 // human ability to see patterns //

 Not so much an "ability" as a propensity; it's programmed into your neural nets at a very primitive* level. It exists even in the lowliest** vision-equipped creatures, as an essential system for survival, both recognizing food and avoiding predators.

 *Nationality can be a factor in assessing just what constitutes "primitive".

**Such as insects, flatworms, and P.E. teachers.
 — 8th of 7, Jan 19 2021

 Thanks [mit] one of the lines I used to use at work when people kept banging on about Machine Learning was that before we started fiddling about with all that, we should start with some basic signal processing (it turns out, for most practical purposes, it's not simple at all) and bar none, all the positive "AI" projects I've ever worked on ended up being DSP.

I had come across Benford's law before as a tool for fraud detection - or for acting as a marker to help decide whether something was "natural" or not - but not seen the DSP guide before - that looks exceedingly useful - I looked it up on Amazon and they're pricing it accordingly - so an online link is quite literally worth a great deal more than its weight in gold. Thankyou.
 — zen_tom, Jan 19 2021

Benford's law is only useful as a predictor when the statistical sampling is likely to be straddling either side of a power of ten. It came up as a challenge to voting totals in our election and had to be shot down by math gurus for why it was erroneously assumed by the plaintiffs.
 — RayfordSteele, Jan 19 2021

 It's got some interesting applications in switched gain electrometer amplifiers for mass spectrometry, too, where the decade gain ranges don't quite overlap perfectly, or there's the problem of calibrating a SEM against a basic Faraday bucket detector.

The solution turned out to be to have a switched-gain amp and a logarithmic amp in parallel, and gate the A/D converter between them, but that's another story ...
 — 8th of 7, Jan 20 2021

 I don't think Benford's law applies. To my thinking, Benford's rule looks inside the number set. I am trying to look outside, at the gap to other sets. Then again, anything can be re-functioned so possibly, all the different data sets could be made to relate by a single unit and run through Benford's Law.

[Lurch] had my thinking right. Does nature have a baseline preferred building structure and therefore preferred numbers? Do these transform as magnitude of complexity and structure increases?
 — wjt, Jan 23 2021

 [annotate]

back: main index