TLDR:

This blog is NOT:

This blog is:

  • a space for me to refine my thinking on machine learning principles and practices
  • an homage to An A-Z of ELT by Scott Thornbury

Why another blog?

Good question.

True, I already have two blogs: one on EdTech and another on using R for data visualization. However, the real question is if a Google search for “data science blog” returns nearly half a BILLION results1, why should I create one?

Surely any opinion piece I write will have been written better by someone more qualified, right?

As for coding tutorials, again, won’t everything I create have already been done by someone with a PhD in computer science2?

While the answer to both of the questions above is more likely than not “yes”, it doesn’t detract from the actual aim of this enterprise which is to provide a space for me to solidify my understanding of data science/ machine learning.

How?

By writing about what I have learned, I will identity the areas where I am weak while solidifying the areas where I’m proficient.

But why do I think it will work?

Standing on the Shoulders of Giants

People who studied in the colleges of science and engineering are usually familiar Dr. Richard Feynmam. While he is famous for his work on quantum mechanics, his explanation of the Challenger disaster will forever be seared in my mind.

Why do I mention him? Feynman’s “notebook technique” is a model for learning anything:

  1. Write down the concept you wish to master
  2. Explain the concept in comprehensible language
  3. Revise ideas/concepts which you can’t explain to a high standard
  4. Identify complex areas and think of how to simplify them further (i.e., create analogies)

Or, to quote one of the true champions of teaching writing William Zinsser, “[w]riting and thinking and learning were the same process” and “Clear writing is the logical arrangement of thought; a scientist who thinks clearly can write as well as the best writer.”3

Or, to put it yet another way, we can cite a quote attributed to John Dewey, “We do not learn from experience […] but from reflecting on an experience.”4

Or, in my own words, I am writing this blog to identify and shift concepts, ideas, and practices from the “Don’t Know” column into the “Do Know”.

Consequently, nothing original will be found on this blog because my purpose isn’t to discover some novel statistical technique nor propose a new theory; my aim is far humbler: increase my understanding of machine learning by clarifying my thinking through the process of writing. Shouldn’t be that hard, right :smirk:

But why the name?

A little more than a decade ago when I was an English language teacher (ELT) in Hanoi, Vietnam, I was introduced to the work of Scott Thornbury, who is a legend in ELT circles, and whose blog An A-Z of ELT was required reading for my colleagues and I.

But why should I use a similar name?

First, structure: there are 26 letters in the English alphabet meaning, at the minimum, I’m compelled to write 26 entries. How long will that take? That’s a good question that I’m not even going to attempt to answer.

Second, I just like the name and, honestly, what more reason do I need? :laughing:

Footnotes

  1. Google search May 14th, 2020 ↩

  2. That is an idea for a future project, “What is the ratio of data science coding tutorials written by holders of PhDs compared to non-PhDs?” ↩

  3. From the “Preface” to Writing to Learn ↩

  4. This quote is most likely an amalgamation of several ideas which Dewey professed, but I still like it :smiley: ↩