Open Source Software

  1. Facebook has open-sourced its PyText project, a machine learning library for natural language processing (NLP) intended to make it easier to put together both experimental projects and production systems.

    PyText, built with Facebook’s existing PyTorch library for machine learning and used internally by the company, was created to address how machine learning using neural networks (such as for NLP). Such libraries typically were “a trade-off between frameworks optimized for experimentation and those optimized for production,” they said in a post.

    To read this article in full, please click here

  2. Microsoft’s open source development tool is an important piece of the developer’s toolkit. Built using GitHub’s cross-platform Electron framework, Visual Studio Code is a full-featured development editor that supports a wide selection of languages and platforms, from the familiar C and C# to modern environments and languages like Go and Node.js, with parity between Windows, MacOS, and Linux releases.

    To read this article in full, please click here

    (Insider Story)
  3. Microsoft and Docker have jointly announced a new project to create “an open-source, cloud-agnostic specification for packaging and running distributed applications,” according to Microsoft’s press notes.

    A Cloud Native Application Bundle (CNAB), as it’s called, provides a standard way for developers to package and run containerized applications in many computing environments, from Docker on a workstation to Kubernetes in a cloud instance.

    CNAB’s specification describes “bundles,” or groups of resources that constitute an application. Bundles also describe how apps are to be installed, upgraded, or removed, and how they are to be moved between environments, even when the target environment isn’t online (e.g., an air-gapped system). Microsoft claims it will be possible to digitally sign and verify app bundles “even when the underlying technology doesn’t natively support it.” Bundles can be deployed within organizations or at large by way of existing distribution systems, such as Docker Hub and Docker Trusted Registry.

    To read this article in full, please click here

  4. A new survey of the Rust user community, conducted by the Rust language team, shows growing interest in the language and its use—but also user frustration with some Rust features that the project touts as advantages.

    The survey drew responses from almost 6,000 Rust users. Questions ranged from the length of one’s experience with Rust to opinions about platforms, workflow targets, and toolchains.

    Because Rust is a relatively new programming language, the vast majority of surveyed users (76.1%) have been using Rust for less than 1 or 2 years. How long did it take for them to feel competent with the language? Most users surveyed felt proficient in “less than a month” (33.8%) or “less than a year” (30.3%). Relatively few (7.1%) felt proficient in less than a week. A significant chunk (22%) of those surveyed still don’t feel productive with Rust.

    To read this article in full, please click here

Go to top