in HW: had to "figure out" what part of the code to change to play with gravitational laws other than Fg∝1/r2
Also aren't we lucky to have a 1/r2-law for gravity!
Physics aside: the extra credit part of the HW
Gravity doesn't actually pull things toward one another, but massive objects actually warp space-time around themselves.
Physics aside: the extra credit part of the HW
Physics aside: the extra credit part of the HW
Comments on reading for HW
Spans "start" and current computational physics efforts
Comments on reading for HW
Spans "start" and current computational physics efforts
Moore's "Law"
Comments on reading for HW
Spans "start" and current computational physics efforts
Moore's "Law"
Comments on reading for HW
Spans "start" and current computational physics efforts
Moore's "Law"
Comments on reading for HW
Spans "start" and current computational physics efforts
Moore's "Law"
Computational power
Comments on reading for HW
Spans "start" and current computational physics efforts
Moore's "Law"
Computational power
Comments on reading for HW
Spans "start" and current computational physics efforts
Moore's "Law"
Computational power - problem of scales of time and space
ex: star/galaxy sizes is factor ~ 1012 ("terra"), we can do ~1010 now
Comments on reading for HW
Spans "start" and current computational physics efforts
Moore's "Law"
Computational power
What is dark enegy?
Short answer: we don't know!
... longer answer:
Onto viz!
A note on Viz Terminology
Visualization
Information Visualization
Scientific Visualization
Simulation
Illustration
Scientific Illustration
Terminology
Science Viz
Time evolving spatial data viz
3D rendering
VR
Info Viz
bqplot
linked dashboards
My background
ytini.com
Naiman et al. 2017, Borkiewicz et al. 2018
My background
ytini.com
Naiman et al. 2017, Borkiewicz et al. 2018
My background
ytini.com
Naiman et al. 2017, Borkiewicz et al. 2018
My background
ytini.com
Naiman et al. 2017, Borkiewicz et al. 2018
How this week is going to look
Intro lecture/activity on a viz concept
Short programming activity
Use ideas-of-the-day to visualize your scientific data
Timed activity! (~2 minutes)
On a piece of paper or in notes on your computer:
What are the most memorable movies you saw over the last year?
Do you prefer cats or dogs?
How would you quantify your experience in visualization?
How many hours do you spend on school work each week?
Breakout Groups:
Group #1:
Break into groups of 2-3 folks:
(1) Share one thing you learned last week that you're excited to apply to data visualization this week.
(2) visualize the results from the data you wrote down previously - you can use your hands, a piece of paper, your computer or anything else you can come up with!
Breakout Groups:
Group #2: Combine your group with another group
Present your viz to the other group.
The other group will try to guess what your visualization is trying to convey.
Discuss how well the group did with their guesses & why you think that is.
What things did they like about your visualization? What things would they change?
There are no hard and fast "right" or "wrong" answers at this point, so don't worry!
Class outline: Syllabus
Computational Physics Week
Day 1: Introduction, syllabus, examples, and some basics about Astro, Physics, Programming
Day 2: Gravity, calculating 2-body orbits, more programming
Day 3: Numerical and analytical solutions of orbits for 2-body problem
Day 4: Multi-body problem in 2D
Day 5: Multi-body problem in 3D
Data Visualization Week
Day 1: Intro to data viz, simple 2D movies, interactivity
Day 2: Info viz and more on interactivity in 2D
Day 3: Plotting in 3D in Python interactively
Day 4: Graphic concepts, web-viz, 3D geometries
Day 5: 3D movies online, finalization of projects, viz party!
Class Mission
While you are already a consumer of visualizations, your
perspective should change to that of a producer of visualizations.
You should be comfortable reading AND writing imagery.
Overview - Themes and Goals
What are the components of an effective visualization of quantitative data?
What tools and ecosystems are available for visualizing data?
What systems can be put in place to generate visualizations rapidly and with high-fidelity representation?
"But the one thing you can't chart / Is how you feel in your heart"
We can't visualize everything
We can't visualize everything
By Vanessa Ezekowitz CC BY-SA 3.0, via Wikimedia Commons
Your brain does interpolation
There are 12 dots, can you count them all at the same time?
Your brain does interpolation
Step 1: Look at the cross
Step 2: Close left eye, keep looking at the cross
Step 3: Slowly move your head toward & away from screen until dot disappears
Your brain does interpolation
Step 1: Look at the cross
Step 2: Close left eye, keep looking at the cross
Step 3: Slowly move your head toward & away from screen until dot disappears
... and sometimes it gets it wrong!
Who are you visualizing for?
For yourself?
For a peer?
For someone else?
Tenet 1:
"Visualizing data" is not a strict subset of "making an image."
Tenet 1:
"Visualizing data" is not a strict subset of "making an image."
It involves:
Collection of the data
Organization of that data
Representation of that data
Tenet 2:
We tell lies to visualize, but we must be honest.
"The Principle of Proportional Ink" - callingbull**.org
"Spurious Correlations" - tylervigen.com
Sensors read the current "fill" of the battery
Analog / digital conversion
Normalized with respect to expected "full"
This is then scaled to a percentage
The battery image is filled from left to right
The image is then rasterized and displayed
Some fixed maximum amount of damage
Each time damage is taken, decrement
Each time damage is reversed, increment
Display number of hearts as appropriate
2 out of 3 "points"
Honesty
Our choices must be:
Deliberate
Informed
Motivated
Justifiable
Election Maps
Mark Newman of the University of Michigan has created visualizations of the
election maps from several of the most recent elections. For more information
and context, see his page http://www-personal.umich.edu/~mejn/election/2008/ .
