partenen, vorarlberg, austria

Crossing the moors

When I started this blog, I purposely set myself two goals, with the expectation I would eventually decide between them and pursue only one. Initially, I felt equally interested in Machine Learning and Blockchain and the decision to start a series dedicated to Statistics, as a preparatory step for Machine Learning, arose from short-term, instead of long-term, circumstances.

As time passed, and I wrote the first posts, I became disillusioned with the choice. I felt inclined to read about Blockchain rather than about ML, intrigued by the novelty, the evident problems and the faster rate of change. At the same time, ML began to appear too large a field, with many possible and lengthy routes.

At least I had constrained my first task: publish solutions for a series of HackerRank problems, and use them as motivators for more theoretical posts. But the reality of it was brutal: although I solved all the problems privately in 2 weeks, I took at least one week to write a post about each solution, requiring several months to complete the series. Indeed, I have still only written posts for about half the problems, and this rhythm is not sustainable in the long-term.

I tried to frame my situation in the adventure theme: writing about the HackerRank series was the first test in the Adventure World, and felt very much like traversing a bog, or a wet moor, with the occasional mound. I am now standing on such a prominence, and taking a pause to scan the landscape around. I see still half the fen in front of me, and after it a barren and distant horizon where I cannot yet distinguish settlements that would provide assistance or training.

All the while, I have been very conscious of the difficulties inherent in writing:

  1. finding the source material, which involves reading, understanding and practicing
  2. organizing that material in a coherent, understandable post, at regular and frequent intervals
  3. given the time constraints imposed by full-time job and a very young child.

Truth is, reality is always harder than even your most careful expectations.

I often had to battle sleep, in what felt very much like the world evoked by Kazuo Ichiguro in his magically sensitive and subtle book The Buried Giant. (If you intend to read it, you may as well skip the following paragraphs now).

I felt like under the spell of the dragon who weaved a persistent pervading fog to drain the land of memories and perhaps its people of their strength. My first test in the adventure world was to acknowledge and persist against the breath of the sleeping Dragon, this porous wall that saps all energy, this demoralizing presence that pushes back against the advance of the Hero, that sows doubt in his heart and keeps the treasure for ever in the distance. Every step in this land, every post, seemed harder than it should normally be, weighed down by simple fatigue.

The first task, similar to crossing a moor, was not hard: there were no difficult fights, and solving the problems amounted mostly to a pleasing stroll across somewhat irregular terrain. I was already familiar with the majority of the concepts, the implementations were easy, but I had to do it in relatively new language for me. There were also several points where I could hook some more in-depth discussions, somewhat like leaving the path briefly to explore denser parts of the terrain. But time is inexorable, and it became clear to me any sort of real progress would take a long time.

The whole situation has changed now: I have the chance to do my research in Blockchain during my office hours, which are longer and of prime quality. I believe it makes sense for me to focus this blog on Blockchain, and on posts focusing at a higher level of discussion (less code, more ideas). Indeed, that seemed like the most rational option from the start, given my professional background. I will still need to write my posts at night, when everyone else is sleeping, meaning the pressure of time will not be fully lifted, but I hope that I will be able to write more fluently and more authoritatively.

That said, I still need to get into this new world, and it will take some time until I can generate ideas and contents for new posts. Until that happens, I will carry on to finish the HackerRank series. But that is a matter for another post.

Leave a Reply