GSoC: How? This? That?

Google Summer of Code has been announced again, and I’m all filled with nostalgia, but it’s not the only things I’m full with (). With a motive of helping GSoC aspirants, I’ll be summarising (in somewhat details), how I prepared for GSoC 2018 and successfully completed my project under CCExtractor.

The post is timeline based and it is near as follows:

• December - January : Selecting the organization.
• January - February : Contributing to the organization and researching on the project.
• February - March : Writing the proposal.
• March - April : Contributing further to the organization.

December to January

Getting Started

I foggily remember I started to contribute to open source from December beginning only with no intention to give an attempt at GSoC since I was just a freshman. I participated in KWoC ‘17 and focused on learning the basics of contribution: how to write a good commit, what are the practices to be followed while sending Pull Requests and much more. One of the most important skills I acquired during KWoC was how to ask good questions. Below are some of the resources one should absolutely go through just with an aim to make the open-source community a little more educated (to be followed in the given order for beginners):

• Why Git? Teach me the basics!
• Explain me the Jargons, please?
• Should my first commit message be “first commit”?
• How to write a good Pull Request?
• I want to do my first contribution, NOW!

And here’s the most important (and I MEAN IT) resource: Catb’s Guide To Asking Smart Questions. It is one single resource that can make it or break it; it teaches the art of asking good questions and patience to look through existing answers. I’ll quote one of the organizations’ admin on how he reacted to someone asking “How to contribute”.

There is no better sign than that to put him on the blacklist!

Knowing how to frame good problems and solve them by yourself is half the work done. Period.

Filtering Organisations

If you note, I’ve used “filtering” and not “selecting”; it is not you who selects the organization, all you can do is limit the organizations you can be selected by .

I started filtering organizations around the last week of December and the process continued heavily till mid-January. Below is the methodology I used to shortlist the organization I would apply to.

i) Going through the list of last year’s organizations I picked those where I think I would be interested to work. I went through all the organizations thrice; for those which interested me, I looked deeper: looked on their past year’s projects and went through their work on a surface level.

ii) Once I had the interest-based list, I filtered out those with technologies which I did not know or was not willing to learn anytime soon. One important consideration to make here is that not all the technologies listed are used, and one should thoroughly go through current projects of the organization before filtering them out.

iii) At this step, I had around half a dozen organization and so should you. This step is where it starts to get a little tough. You should filter out a few more projects after going through their communication channels, projects, codebases and beginner resources.

iv) Join the communication channel for all the organization remaining in the list: solve their beginner issues, ask “smart” questions, send some PRs and just stay there for a while. The point where you start feeling like you don’t want to be a part of this community, leave the group. It can be due to a lot of reasons: inactive channel, poor reviews, poor documentation etc.

v) Ideally, you should be left with one or two organizations where you feel like home. In parallel, you should be in mid-January and ready for coding heavy work.

January to February

At The Org

Every organization has a different method for assessing the project proposal. Here is CCExtractor’s guide for GSoC 2018; every organization has similar pages and they explicitly tell how they will be judging the proposals. One point to keep in mind is that they only apply these criteria AFTER they deem your proposal to be good enough.

Generally speaking, the best way to get selected is to come under the radar of mentors and admins; this can be done by solving issues, asking “smart” questions and guiding others towards useful resources. It might come to your mind that pinging the mentor personally will help a lot but it does the opposite; stick to public channels for seeking help (in the beginning).

The Project

Alongside, you should be going through the ideas and projects the organizations (in your list) have. There will always be a good number of projects and you should select what interests you the most; nobody expects you to know how you wish to accomplish it at this moment. But it is important you are passionate to complete it, and in the process learn the required technologies.

Once you’ve finalized on the project (which should be totally based on interest), start figuring out what all the technologies you would need to work on it and start learning about them. This time should be optimized for researching on various methods and figuring out what will work the best in the production; finding the best method for an action will give you lead over other proposals for the same project.

You should keep a detailed record of your experiment with different technologies to present in the proposal. While researching for my GSoC project, I created a fully fledged Google Drive client as a prototype for methods I had suggested in my proposal. This helps a lot in convincing that you know what you are talking (or writing) about.

February To March

Writing The Proposal

I began writing the proposal around early February; it is advisable to start writing the proposal immediately after you’ve selected the project(s) you want to work on. There are many formats for proposal, and some organization even define how they want the proposal to be and what all they want you to include. If the organization is open to all formats, it is suggested to take inspiration from proposals that got selected in previous years. Here is my proposal as a reference which was slightly inspired by previous proposals (especially Saurabh Srivastava’s) submitted to my organization.

There is an usual friction in starting to write the proposal; I faced it and it took me three days to write the first page. This is not a good place to get stuck at, and should be avoided by writing the obvious parts of the proposal (introduction and contact details, for example) first. One should be very very patient while writing the project since it can (and should) take up a month.

You should read “How to write a kick-ass proposal for Google Summer of Code” by TEO where he explains in-depth how you should go about writing the proposal. Please note that proposals should only be limited by your imagination (and organizations’ specifications) and NO guide should be followed rigorously.

Contributions and other community involvement should be going abreast; one should avoid to have long inactivity in the channel, and never miss conversations about the project.

March To April

Submission

The proposal should be in final stages by the week of submission and should be given to your seniors and mentors for review. Do not share the proposal publicly, and only share links with commenting rights; there are always some cases of plagiarised proposals and you should be cautious.

A good sign of selection, at this time, is that your mentor should be talking to you in detail about the project and expecting a proposal from you on the same.

A word of advise - although you can submit at most three proposals, it is advisable to NOT submit more than once unless you are not well prepared, or unsure about your selection in your primary organization.

After Submitting

DO NOT become inactive in the community after submitting the proposal; keep solving issues, reviewing PRs, helping others and asking “smart” questions. Many organizations keep this time for judging the student’s commitment since now he has no pressure of writing the proposal and should be contributing directly to the organization.

This wraps up the technical and organizational part of the preparation. There will be many emotional challenges you’ll be facing while submitting the proposal: you’ll lose hope at times, your patience might get over, you might grow jealousy towards your competitors and much more. It is highly important to keep yourself emotionally healthy during this phase; talk to your family, friends, and seniors for help and motivation.

I’m highly grateful to Himanshu Mishra and Naresh Ramesh for being my mentors during my preparations.