Delivered as good news on the 04th of this month, (well technically almost the 5th since the UTC timings gave only half an hour of a difference between the dates) Google Summer of Code (short for GSoC) selection @ The R project for statistical computing for this year’s edition is something that I have been pinning for since the second half of 2019, the time when I came to know about GSoC’s very existence.
Considering factors that possibly led to my success, I completed all of the tests laid out by my mentor Toby Dylan Hocking and prepared a suitable proposal early on, each with a few changes as suitably recommended. I also happened to find a co-mentor for my project - Neeraj Dhanraj Bokde, who happens to be the co-author of the GuessCompx package, which I dug out of CRAN’s list in search of an ideal reference package (it deals with the same subject of complexity testing). Regarding GuessCompx, I would like to thank the main package architect Marc Agenis for his share of help in making me understand the functionality of his package through a set of shared responses between us during January. I sure am glad I approached him (back in 2019), along with my GSoC mentors.
Apart from these factors, there is some hard work involved from my side, which includes learning R since it’s not in my semester courses, so I had to devote a bit of time each day, which was difficult because my 5th semester subjects were relatively more though than the ones in the previous semesters. Nevertheless, I scored a ten again (making it three times in a row - a 10/10 GPA for all the CS-subject semesters up so far, although it doesn’t matter much to me - I’m just glad I made the effort both ways!). If you’ve been following my posts, you’ve most likely seen ones made for my reference during December 2019 (CN, Accounts, Java) - the time with end semester exams on, coupled with preparation for GSoC tests, learning R, and contributing to a mix of programming (SO) and game communities, which could give you an idea how much time I devote to stuff I like even though it sounds overbooked.
Regarding R, I only learned certain aspects of it, i.e. primarily the topics which I thought I would need to cover for my package (thanks to LinkedIn-Learning and rdocumentation for that), and there remains the vast sea beneath the language left for me to uncover.
That kinda sums up my entry. But the selection is one side of the story, and passing it successfully is another - which is my current priority, now that I have been selected in my first go.
My future blog posts until the end of the official timeline will feature content on my R package testComplexity, typically followed by some posts with a naming convention of testComplexity : xyz, where xyz would mostly pertain to the functions in my package, which follow a syntax in Camel Caps. (eg: asymptoticTimings)
| Anirban | 05/10/2020 |