A running list of ideas for future blog posts and improvements. To read more about the motivation behind this, click here
Post Ideas
Discuss how I made shokunin
Title: If you build it, will they come?
Subtitle: notes on building this data science blog via a meta-review of notes other people have made while building their blogs
Discuss the how and why of power calculations
Title: I have the power
Title: Notes on how you too can (by the power of simulation) … have (the) power!!!
Picture of he-man
Look into Quarto 1.4 and manuscripts as the document driving posts
Explain analysis engineering (originally mentioned here, now done here)
Perform and Discuss MTA re-analysis
Title: To OMNY or not to OMNY
Subtitle: On using poisson distributions to save money
Inspiration: Which is Better? OMNY or 30-Day Unlimited MetroCard?
Idea: Driven by my benefits-election period, I try to figure out whether to use OMNY or a 30-Day Unlimited MetroCard
Improvements vs inspiration:
I update prices to reflect new MTA pricing
I account for inflation to reflect how time has passed
I use R (rather than python) and Quarto (rather than a jupyter notebook)
Explain converting thesis to Quarto from RMarkdown
Discuss my approach to learning
Discuss and explain RMDD concept by using MTA re-analysis
Title: RMDD on MTA analysis
Subtitle: Using RMDD to turn the MTA analysis into a standalone package
Inspiration: RMarkdown Driven Development
Idea: I think RMDD is a brilliant concept and I’m curious to see if I can use it to bridge the gap between interactive analysis/reporting and fully reproducible, production-like code.
Discuss signing up for Kent Beck’s blog and using TDD at work
Discuss converting MTA package into a shiny app
Title: Figure out whether OMNY or the monthly unlimited is better for you
Subtitle: Converting a package into a shiny app
Idea: I haven’t used shiny in a long time, but I think converting a relatively simple package into a web application will help resurrect old skills and build new ones.
Discuss PostgreSQL API
Discuss improving speed of R code
Title: That’s slower than I thought!
Subtitle: On doing one successful thing too many times
Starting Quote: The way you’d do something once is rarely the best way to do it 1000x - Quote from David Robinson
Inspiring article: Making R Code Faster: A Case Study
Idea: Driven by a situation at work where I’d written code that did something correctly once and then just attempted to map that code. Although it took a while. I came back to the repo ran it and was shocked that it ran much quicker than before. Turns out a senior colleague had changed one line (shown below). This is an attempt to figure out why it was so fast.
Explain adding RSS feed and submitting blog to Rbloggers
Discuss Udemy’s 100 days of code course
Idea:
Explain why you started the course
Periodically update post with list of repositories where you’ve made cool things
Make post about being careful about what you put in warnings
Title: The condition of man… is a condition of war of everyone against everyone
Subtitle: On taking care about how you define conditions
Idea: I was recently working on unit testing an API package and I could not get a unit test to pass despite modifying the function that was being tested and being absolutely sure the error message I expected was the one that was being thrown. Turns out that although the messages “looked” the same to the human eye, there was some regex weirdness happening.
Discuss cool projects I’ve build through Udemy course
Turn OTIL presentations into short blog posts
Discuss cool presentations I’ve made
Blog Improvements
Add last updated functionality
Add projects section
Add thesis and PostgreSQL API
Add thesis to github
Add API to github
Add presentation section
Add code-and-tell presentations
Add datapalooza presentations
Add RSS feed and submit blog to Rbloggers