— Douglas Adams, The Salmon of Doubt

Well, yesterday was the deadline for student applications to the 2018 Google Summer of Code (GSoC), and the mentors at the JGraphT project are definitely stopping to take a long breath after the last few whirlwind days of students finalizing their proposal drafts.

We ended up with 110 finalized proposals submitted by students from all over the world. As you might expect, many of those were “spammy” (with last-minute applicants just pasting in their CV’s and regurgitated bits and pieces of our ideas page), but even after filtering those out, we’re left with 24 good-quality proposals to sort through. One student even designed a custom logo just for the proposal itself. ☺️

Since this is our first time participating in GSoC, we’re likely to be limited to mentoring 2 or 3 students at most. As a result, we’ll have no choice but to reject many good proposals, which is equally painful for us and the applying students. We’ve learned a lot about the process over the last few weeks, and we hope to be able to apply that experience next year as a returning organization with a bigger number of student slots allocated.

But for this year, we’re going to try to make the most of what we’re allowed; so over the next few weeks we’ll be reviewing proposals, making the final cut, and preparing to kick off a great summer with the selected students (who’ll be announced April 23).

If you are a student who has put in a significant amount of time and effort into the proposal drafting process, but you still don’t end up making the cut, believe me when I say we wish things could be different. But no matter what, we hope that:

• you already learned a lot from researching and drafting your proposal
• you gained valuable open source experience by working on the warmup problem and then submitting a pull request to our code base and getting it reviewed (and eventually merged)
• you’ll stay engaged and continue contributing when you have the time
• we’ll have a better chance of including you in the program next year!

In fact, we did already receive high-quality pull requests from many applicants; we’ve managed to merge a few of those and are in the process of reviewing the others. Here are some of the bigger ones:

• Chordality inspector (implemented via both MCS and lexical BFS iterators)
• Read/write synchronized graph wrapper
• Alpha centrality metric
• Palmer’s algorithm for constructing a Hamiltonian cycle in a graph meeting Ore’s condition
• Random regular graph generator
• Bipartite graph 2-coloring algorithm
• Triangle freedom graph test
• Dinic maximum flow algorithm

And there were plenty more in the way of smaller code improvements, wiki cleanups, etc. We’re very excited and grateful to have such enthusiastic participation happening!

A big thanks to my fellow mentors, Joris Kinable and Dimitrios Michail, for all their organizational and review effort to get us this far. We’re also pleased to announce that we’re being joined by a backup mentor (Ilya Razenshteyn) to help us support our students throughout the summer.