May 29, 2013

Google Summer of Code 2013 | My Story

Quite recently, I came up with a post on how I started contributing to open source, and explained that the only organization that I was following didn't make it through to the selected mentoring organizations of Google Summer of Code. That was a difficult situation for me. Being a Geology student and when every classmate of mine had at least two interns to boast of, I hadn't even bothered to apply properly to professors or companies. The GSoC was my last hope. However, everything seemed grim.

The organizations started showing up in the list by 12.30 AM, and there was already a mail in the e-cidadania mailing list saying they hadn't made it. I decided to just go to sleep, and wondered that my exams being less than 20 days away, how on Earth would I get into something as difficult as this. Worse, I had nothing productive to do this summer and I wondered if it was high time I start learning Ruby on Rails.

Next morning, I got up afresh and opened the list. After being in hardcore Django development for the last two years, I instinctively did a keyword search for 'django'. Among the results were some organizations which were too tough to get into (Honeynet Project and Shogun for example). On searching for Python, the first thing that caught my eye was Ankur India, and the tag 'Bengali'.

The good thing about Ankur was that their projects didn't involve submission of patches. They were more concerned about applications that would help in l10n/i18n. The next few days I spent in researching for their project ideas, and made a sample proposal, which was not well received(for various reasons that I came to know about later). In fact, at that time I feared that my GSoC might have ended before it even started!

I started asking questions on Quora and recklessly promoted them(I had recently received 2500 credits from a total stranger as goodwill for explaining the idea of an application that I created). The only clue I had received was a tool called Deckard. I asked around a bit more, contacting creators of the already existing tools in l10n, and one of them, Nicolas, also the lead of Deckard, turned out to be a mentor of Ankur(he had joined just that week)! He helped me out a lot, and with drastic changes to the proposal, I came up with something that was in a good condition to be submitted. I could see light at the end of the tunnel.

By that time, I wanted to get in the Google Summer of Code at any cost, and it had become an obsession. I really love coding, and this was my first chance to do something substantial, and officially get an internship at the same time! I couldn't let it go- and there was no way I was going to.

Meanwhile, a good friend of mine got me interested in a PHP based project, ATutor(under Inclusive Design Institute). Although I had not contributed directly to any PHP app in the the last few months, I did all the data porting through PHP (although it's slower)... There was a small problem with it though. ATutor preferred communications in the IRC, and the developers were based in Canada, so the time gap made communication difficult. Another issue was that the irc (as well as the git) port is blocked in IIT Roorkee, and due to some reason, the webchat of OFTC would log out automatically after a few minutes of inactivity. However, I did manage to get a bug assigned to myself after maybe five attempts and submitted a patch soon(that's always a very crucial part). The project idea that I was interested in was a largely research based project, and as I had followed the ajaxification of PhpMyAdmin last year (until they dropped it as a GSoC project), I added the ajaxification as a part of the project, which I believe made it stronger. I got a very detailed feedback, and worked on improving it before the deadline.

Each of these proposals took over 24 hours to perfect, and after their submissions, I decided to finally concentrate on my end semester exams. I did submit another last moment proposal to ATutor (halfheartedly) in the dying minutes of the student application window, but that was mostly for fun (that's my idea of fun in the middle of my end semesters, btw)!

Everything seemed quiet for some time. One fine day, I received a mail from Alexey Novak, my mentor for the Atutor project saying he thought my proposal was strong, but that I should increase my presence at the IRC. We chatted that evening (6 PM for me would be 8.30 AM for him, owing to the time zone), regarding the project idea so that we were clear on points. It seemed I might get through, but I could not be sure...

... until the morning of 23rd of May. I got a mail from Greg Gay, the head of the ATutor project, with the header "GSoC Duplicate" (with carbon copies to the Ankur coordinator and the mentors for both projects.) Here's what it said.
Hi Shaumik,

You have potentially been selected by the IDI to work on the ATutor project and by Ankur India to work on one of their projects. We have decided to let you choose which project you prefer.

Can you let us know as soon as you can which project you want to work on. CC everyone when you reply please.

For a few minutes, I just kept staring at my screen wondering if I was reading it right. Not before I read it five times could I digest the fact that both my proposals were good enough to be selected for something like the Google Summer of Code!

Then came the hard part. I had to choose one. All the posts I had read on the GSoC came flashing into my mind reminding of the very basics of what a GSoC should be. That is when I decided that with my experience of developing applications for IE 6 (some professors still use it, don't blame me!), I would be able to work more efficiently in ATutor.

I knew from one of the seniors that GSoC sends emails with either two subjects "Congratulations... " or "Thank you ..." depending on whether your proposal got selected. I got the "Thank You ..." mail and my heart stopped. Was it even possible that I didn't make it? Did the IDI admins forget to select me? Was it a glitch in the system?

After a quick glance through the email, the term "Ankur" caught the eye. This mail was for the Ankur proposal that I had submitted. Soon, I got the mail which made me happy.

With the mail asking me which project to choose, I was (almost) officially informed that I had made it but the fact that I get pessimistic when it comes to good things like this, I didn't write this blog until the results were announced officially in the Google Melange site

When the list of organizations were announced back in early April, I had no idea if I would even be considered for any project. Almost two months later, it feels that I have come a long way.
And so, the search for an internship was over! All's well that ends well.

Liked this post? Have any suggestions? Just let me know. Feel free to comment below!

0 responses:

Post a Comment