still admiring J

On Friday, March 23, 2018, kirby urner <[email protected]> wrote:

>
> On Fri, Mar 23, 2018 at 1:33 AM, Wes Turner <[email protected]> wrote:
>
>>
>> xarray.Dataset is n-dimensional
>> https://xarray.pydata.org/en/stable/
>>
>> From a tweet a few days ago https://twitter.com/westurner/
>> status/973058715149578240 :
>>
>>
> ​Excellent pointers!
>
> I anticipate a steady stream of new math teachers (sort of like I was some
> decades back, teaching high school level in Jersey City, first job outta
> college),  and getting to where they're hungry to include more coding, and
> need appropriately mathy topics.
>
> Lots out there already, a steadily growing stash since the early days, of
> Logo and OLPC (One Laptop per Child), SugarLabs...  The Learning to Code
> movement (code.org etc.) is but the latest in a series of chapters.
>
> BTW, one of my code school mentors just showed me this one:
> http://www.codebymath.com/index.php/welcome/lesson_menu
>
> My own contributions to this genre go back to what I called the "Numeracy
> Series".  That's closer to when I first came across Python. I'd already
> played with "quadrays" in Visual Foxpro.[0]
>
> http://4dsolutions.net/ocn/numeracy0.html  (that's part 1 of 4, early
> 2000s)
>
> I was into mixing Python with POV-Ray, the free ray-tracer, to get simple
> geometric figures.
>
> These days mathy high school level topics might include...
>
> -----
>
> *1*  Crypto: a gateway to group theory.  Since my day, we've added more to
> the "fun reading" shelf, such as with Cryptonomicon (Stephenson, historical
> science fiction) and The Theory that Would Not Die (i.e. Bayesianism) by
> McGrayne.  As clear from the cover (https://flic.kr/p/22z1vAy) she's
> covering a lot of the same territory: Bletchley Park, Alan Turing, enigma
> machines.  You've got prime numbers, RSA, Elliptic Curves and hashlib SHA
> stuff.  Introduce blockchain.  A great way to access recent history.
>

"Show HN: An educational blockchain implementation in Python"
https://news.ycombinator.com/item?id=15945490
https://github.com/julienr/ipynb_playground/blob/master/bitc
oin/dumbcoin/dumbcoin.ipynb



https://cryptography.io/en/latest/
recommends
https://www.crypto101.io/ (PDF)

https://github.com/pyca/cryptography/blob/master/src/cryptography/hazmat/
primitives/ciphers/algorithms.py


https://en.wikipedia.org/wiki/Fermat%27s_little_theorem
https://en.wikipedia.org/wiki/RSA_(cryptosystem)

"How secure is 256 bit security?" by @3blue1brown
https://youtu.be/S9JGmA5_unY

https://en.wikipedia.org/wiki/EdDSA#Ed25519


> *2* Geometry: and in my case a genre of "mental geometry".  My topic is
> the somewhat off-beat solid geometry of the geodesic dome dude (another
> movie coming up [2]), and my modality is often Python.  One of the slides
> shows a Tetrahedron class with our formula for its volume, a public sandbox
> at REPL.IT. https://repl.it/@kurner/tetravolumes
>
> I don't remember if I've already shared this link ( https://goo.gl/zoVYF1
> ) to my slide deck (Google Slides) for a recent talk at Systems Science
> building, Portland State [1].
>
> That's probably still too off beat for most, whereas Fractals (dynamical
> systems, strange attractors, Wolfram automata), if counted as "geometry"
> (certainly visual) already have a strong following and both connect back to
> deeper math.  I've done work around those topics too and attest to Python's
> applicability.
>

Fractals... Mandelbrot
https://en.wikipedia.org/wiki/Chaos:_Making_a_New_Science

IDK what's a good Python package for fractals?


>
>
> *3*  Data Science: which is where the nD-array ("xray", "tensor") and
> Linear Algebra come in, and Calculus (which they're learning anyway).  When
> not connect Calc to Stats with a thicker pipeline right from the start?
> So what if the Gaussian Distribution doesn't have a closed form integral
> or whatever (https://goo.gl/9XyzN5 )?  Not an issue.  We just wanna make
> sense of the curves in some lessons, and both the CDF (the integral) and
> derivative of the PDF (normal curve) are worth talking about visually, and
> coding in Python.[3]
>

http://docs.sympy.org/latest/modules/stats.html#sympy.stats.Normal

http://docs.sympy.org/latest/modules/integrals/integrals.html


>
>
> -----
>
>
>
In sum:  if you're a math teacher chomping at the bit for something fresh,
> and not already over-exploited by the test-maker economy (you still have a
> free hand), then we have riches in:
>
> Crypto (web browser TLS, eCommerce, Supermarket Math [4]);
>
> Geometry, including off-beat whole-number volumed polyhedrons from New
> England Transcendentalist literature (v 2.0);
>
> Machine Learning / Data Science. All accessible to high schoolers.
>

https://www.kaggle.com/learn <https://www.kaggle.com/learn/overview>

https://jakevdp.github.io/PythonDataScienceHandbook/

https://developers.google.com/machine-learning/crash-course/ml-intro



> Kirby
>
> [0]  Quadrays are a weird coordinate system featuring a "caltrop" of basis
> rays to tetrahedron corner points (1,0,0,0) (0,1,0,0) (0,0,1,0) (0,0,0,1)
> from origin (0,0,0,0) -- but there's nothing "4D" in the sense of
> non-visualizable hyper-dimensional, just ordinary space mapping, 1-to-1
> with XYZ if you stick to the unique canonical representation.
> http://mathforum.org/library/view/6236.html
>
> [1]  http://worldgame.blogspot.com/2018/03/systems-science.html
>
> [2]  https://youtu.be/oyLWPWydvyo
>
> [3]  one of my Jupyter Notebooks about the bell curve:
> https://github.com/4dsolutions/Python5/blob/master/BellCurve.ipynb
>
> ​
> ​[4]  http://wikieducator.org/Supermarket_Math
>
>
>
>

_______________________________________________
Edu-sig mailing list
[email protected]
https://mail.python.org/mailman/listinfo/edu-sig