Speaker Details

Michele Costantino Soccio

Michele is a Principal Engineer at Amadeus, where he addresses the software and database scalability challenges of the Airlines Revenue Accounting solution. Since joining Amadeus in 2010, he has specialized in distributed systems and in transitioning complex systems from legacy stacks to modern, more efficient architectures. Prior to this, he was a researcher in multimedia video quality assessment and a pianist.

Michele is passionate about both computer and human languages and is an avid reader and classical music enthusiast

Life after C++
Conference (INTERMEDIATE level)
At 45 years old, C++ has become a target. A multitude of “successor languages” have emerged with one goal: to replace it in our hearts and our codebases. These languages are ready to take away those programming nuances and annoyances that have made sure we, C++ lovers and developers, don't spend too much time on actual work.
Rust, Zig, Hylo, even Swift, and more. Each language has its own approach to becoming The One that with one final memory leak will make us forget about exceptions, data races, and cryptic template error messages.
With a bit of history to set the context, and code examples to spark curiosity, we'll explore the most promising “successor languages”, their perspectives and which problems they want to solve. We'll see how and why industry giants like Google, Adobe, and even Apple are actively involved in moving beyond C++. We'll see why memory safety isn't everything and why compatibility will be crucial. After all, one does not simply leave behind decades of great software for the sake of a nicer syntax.
I've attempted to break free from C++ for over 10 years. Now, there's a light at the end of the tunnel. A life after C++ where, for better or for worse, there will still be plenty of C++.
Delta Storage: a compression that makes the difference
Byte Size (BEGINNER level)
Billion of events make large databases. And with large databases come large headaches.
When you have many records that only differ for a few fields, you need a compression method that could make the difference. Literally.
So goodbye GZIP and welcome VCDIFF!
We'll see how to efficiently compress any kind of data taking advantage of their similarity while being completely format agnostic. You'll see how powerful the VCDIFF compression is with some live coding and some live drawing where we'll try to beat even the JPEG compression.