Student Sachin and his experience in Google Code-in 2018

Notice about minors permission: Students have written those posts and sent the photos, and give parents permission to us to publish them. Still they are anonymous in the sense that we do not publish their authentic names, but his nicknames on the contest.
Sachin



This is my second attempt at GCI. In 2017, I was selected as a global finalist for a different organization.

This year, task 0 was to choose the right organization. There was probably no wrong answer here, as most of the 27 organisations are seasoned players. But something about Liquid Galaxy seemed mysterious and drew me in. Exploring Google Earth on a multitude of screens was certainly alluring. Being a 11th grader, I had dabbled a bit in Java and Python in the past, but how would that help me in making this fascinating product? Eventually, I got comfort by quickly solving a few archived tasks. I decided that I would rather have a challenge and learn new and interesting things rather than completing drab tasks only with the intent of winning. It didn’t hurt that Liquid Galaxy LAB’s headquarters is located in Spain as I have a special interest in Spanish. In nuce, Liquid Galaxy was the chosen one.

Now that Google Code-In is coming to end, here is a short compilation of some of the tasks I worked on that I found very interesting and enjoyable.

Hardest Tasks

Chrome Experiments with Liquid Galaxy
As the mentors had already mentioned in the title of the task, this was going to be the hardest task. Briefly, the task was to display any chrome experiment (a showcase of work by coders who are pushing the boundaries of web technology, creating beautiful, unique web experiences) on a liquid galaxy. Now, this is much easier said than done. After hours spent researching various ways to do this, and trying on multiple different experiments (Tron Tank, Field), I had to throw in the towel because of several factors including the time limit on the task and the fact that my exams were (and still are!) going on. However, I am still working on building this and am confident I will have it ready soon.

Installation script for LGxVoiceControl
This task was to, as the title suggests, create an installation script for LGxVoiceControl, a feature that allows the Google Assistant to communicate with a Liquid Galaxy. Although this did not seem like a very hard task, I was new to shell scripts. One of the steps involved in the installation is to restart the computer after making certain changes, and then continue the script. I was quite confused as to how to do this initially, but brainstormed and came up with an innovative solution towards the end.

Favourite/Enjoyable Tasks
Although I know it sounds cliche, I thoroughly enjoyed every single task I worked on throughout Google Code-In. This list of favourite tasks was really hard to come up with for me. Note: the tasks are not in order.

TensorFlow tasks
There were a couple of TensorFlow tasks, one of recognizing various monuments and another about discriminating between images of burnt, half-burnt and regular forests. The monument recognition task was actually the first task that I did in GCI this year, and was possibly that task that got me hooked to Liquid Galaxy. I really enjoyed working on both of these TF tasks. I had previously worked with TensorFlow and have applied to several other uses also, but it still amazes me whenever the model is able to correctly predict the identity of some input. In addition, I liked the Burnt Forests task because it allowed me to be creative in collecting the training images (since there weren’t any when I claimed the task) – I created a script that scraped images off Google Images for the same topic, which was also nice to implement.

Firebase tasks
There was a series of 4 Firebase tasks that I also loved working on. In my opinion, Firebase is a beautiful tool, and it allows for so much and is also really simple to implement. The string of Firebase tasks used many aspects of Firebase, including the Realtime Database and the authentication. Again, I had previously worked on Firebase, and so knew the nuances of it already (but the feeling you get when the realtime database syncs is still pretty amazing!). Part of these tasks, I also tried to customize my site to make it as user-friendly as possible.

Android (Design + Questionnaire + link to LG)
I identify myself as an Android developer, so of course, I enjoyed the Android tasks. These included the redesigning the user interface, implementing colour trails (this was particularly cool!), generating a pie-chart based on answers to a questionnaire, and also linking the HAPIS app with the Liquid Galaxy.

Dronecoria
As a drone enthusiast, I enjoyed the Dronecoria tasks. These were generating a path on Google Earth for the Drone to follow, which enabled me to learn about KML, and the TensorFlow tasks (explained earlier).


LG Installation + Aquarium
The first time I installed Liquid Galaxy on 3 computers, my mind was blown! I was awestruck by the immersive display that was right in front of me. I was also pleasantly surprised by how simple it was to install the Liquid Galaxy. All I had to do was fill in a couple of lines of code on a file included in the Google Earth package, and that was it! I will never look at a map the same way again! I also liked the Aquarium task, where I had to view an aquarium on the Liquid Galaxy. This task involved research, and helped me identify various other uses of the Liquid Galaxy.

New games for LG
Although this was a task that did not require as much time and dedication as some of the others, it was one that allowed me to be creative and come up with some fun games related to geography (a subject that I am passionate about!). I ended up proposing a game that I thoroughly enjoy playing (and am sure you will too), GeoGuessr.

Google Assistant
Learning about developing on the Google Assistant and DialogFlow was an enriching experience for me. I liked the task on the codelabs related to the Google Assistant, as well as integrating the Google Assistant with the Liquid Galaxy. Coincidentally, I was already going through them when the task came out!

Easiest tasks

Design tasks
Personally, I thought the design tasks were easier to complete. These include the t-shirt design, logo design and leaflet design tasks. However, they did let me be creative, and I am grateful for that!

POI sets
The POI set tasks served as a useful stagger to learn about KML and the Google Earth, but were simple and only required a short time commitment.

GCI 2018 is coming to an end, and so are my exams. What a pity, they could have nicely staggered so that I do justice to both. But then, who said life is fair?


After two years of participating in Google Code-In, I am hooked to open source programming. I have already bookmarked an endless list of open source codes to explore during my holidays. I guess that is the essence of GCI – to get teenagers interested in open source. But one thing is for sure, I absolutely LOVED participating in GCI this year with the Liquid Galaxy Project.






Our Team

'

PARTNERS