Google has launched an open source version of one of its libraries. The Internet giant uses the Library to control some of his own core products. The library ensures that data can be collected without revealing personally identifiable information inside or outside the companies.

Miguel Guevara, product manager at Google’s Privacy and Data Protection Office, states that privacy protections are important to maintain users’ and customers’ trust, TechCrunch writes. He talks about ‘differentially-private data analysis’, which he calls a principled approach with which organisations can learn from the vast majority of their data. At the same time, they ensure that the data of an individual cannot be re-identified.

Functions in library

The library that is now available as an open source version focuses on functions that are often difficult to build from scratch. The library also contains many of the standard static functions a developer needs. That’s like count, sum and mean.

In addition, the library contains an extra library for rigorous testing. This is important because it is difficult to build in ‘differential privacy’ properly. It also includes a PostgreSQL extension, as well as a number of recipes to help developers get started.

The library is available on GitHub under an Apache license and is written in C++. Finally, the library has been designed in such a way that it can be extended to include other functions, such as additional mechanisms, collection function and privacy budget management.

Google dives into open source

Google has made quite a few open source projects this year. For example, in August it created its voice engine for Live Transcribe open source, so developers can use it to transcribe long conversations. The tool makes it possible to convert audio into text in real time. It does this using algorithms and machine learning.

In February, the company also made its tool Clusterfuzz open source available. This allows developers to automatically test applications during the development process for possible bugs and possible vulnerabilities.