The Infinite Loop

Earlier this year, while I was developing a mobile application for sending users personalized hydration reminders, I realized simply how horrible I was at naming variables. Of course, there were several parts of developing the app, such as overcomplicating the problems themselves, where I struggled, but I never expected nomenclature to be an issue. My ability to overcomplicate problems simply sky-rocketed. I would have variables such as numCups, numCupsNeeded, numGlasses, numGlassesOfWaterRequired - they were all different variations describing the same idea with very similar and “highly descriptive” names. It became so problematic that I struggled to differentiate between these variables, not knowing which one was a dynamic counter and which one was describing a threshold preference. My urgency to “make it work” led me to succumb to the natural tendencies of adding more useless code, often more complex than necessary, and practically un-debuggable.

I remember the look on my father’s face as he tried to decipher the tangled lines of code. With such bad naming abilities, even comments could no longer help make comprehension easier. Unhappy at the exponential way my programs were becoming more cluttered, I began reading Code Complete. As I read my way through, I came across several topics including the importance of having good design, and implementing it appropriately. Essentially, it was emphasizing the importance of planning precisely before starting work, as is the case with anything.

In an ideal society, I realize that we can experience such a smooth relationship between planning, implementing, and making revisions to planning. Technically, we can stay in an infinite loop of ideation, planning, implementation, and making improvements because we can always improve. However, it is an issue when the infinite loop curves prematurely, which is what I often fall into until I feel frustrated and completely leave design. I keep thinking and planning, perfecting it, adding so many details simply because I falsely believe it will be helpful. My thorough plans are not very flexible.

At some point, I am only thinking and not doing, leading to a lack of motivation because of no tangible results. I evidently want to avoid this with my Original Work project. For that reason, I decided to truly understand the Encoder-Decoder architectures that are used in machine translation. The similarities between text summarization and machine translation, I believe are significant. A text summarization is really just an English to English translation, in a much more compact way. Several other questions remain unanswered. I understand that there are pretrained models that are often used, however if I am using a model that is trained on language-to-language translation, how would I use it for text summarization? How do I simply extract “the point” of the cumulative phrases? This infinite loop of asking questions, answering them but asking even more is a welcome one, never leading to a lack of motivation.

In addition to Original Work related challenges, I had the opportunity of participating in an asynchronous video interview. Although the idea seemed bizarre, having to talk to a green dot beside my camera, and a future listener I cannot see, being cut off exactly at 2 minutes, I was surprised at myself. After practicing not only interview-speaking and interview-question-answering in ISM, but simply developing the skill to speak with clarity yet remaining conversational, I was able to control my nervousness. A few months ago, the idea of an interview would result in sleeplessness or waking nightmares. It still does. A few months ago, my mind would be racing with unchecked energy, jumbling my words and twisting my tongue, preventing me from thinking and speaking. This time, however, I was able to think during the interview with a clear mind, trust myself, and feel confident. The most beautiful part was I did not realize this until my interview was over. The practice we have done inside and outside of ISM, speaking and putting ourselves in non-ideal situations let me handle the interview in a much more stress-free manner.