Leading tech companies have formed a collective that seeks to create a new lexicon that companies and developers can use to replace problematic language when building new technology.
Organisations including Red Hat, Cisco, IBM, VMware, and the Linux Foundation have established the Inclusive Naming Initiative following unilateral decisions earlier this year by some companies to scrap certain naming conventions that had been deemed exclusionary or insensitive.
GitHub, for example, retired the ‘master’ label for its default branch structure in June in order to address concerns about the term's association with slavery. Several other companies followed, including Twitter and Linux in July, stripping out potentially racially insensitive terms.
The initiative aims to coordinate this effort across the wider industry, rather than leave it to individual companies to replace what are commonly-used terms with a mix and match of alternatives. The fundamental aim is to remove all harmful and unclear language and replace it with an agreed-upon set of neutral terms.
“We knew from the beginning that this would be a complicated endeavour. On a technical level, change has to be made in hundreds of discrete communities, representing thousands of different projects across as many code repositories,” said Red Hat’s senior vice president and CTO Chris Wright. “Care has to be taken to prevent application or API breakage, maintain backward compatibility, and communicate the changes to users and customers.
“As big of a lift as that is, the technical change is only part of the challenge. We all come to this work with a different perspective, a different cultural grounding, and different levels of understanding of how certain words can impact others. Our hope is to start the discussion, provide context and answer questions, and inspire change across the open source ecosystem.”
The drive to change the language used in tech is a recent phenomenon, brought on by the death of George Floyd earlier this year and the subsequent mass resurgence of the Black Lives Matter movement in the US. As such, companies have been motivated to be more conscious about the working environment they create and promote by re-examining much of the terminology that’s been used by default for decades.
The project, which will begin next year, has devised an early list of words it considers falls into the bracket of harmful language, including whitelist and blacklist, and master and slave. While concrete alternatives will be agreed on by the initiative in due course, some early proposed alternatives include allowlist/denylist for whitelist/blacklist, and primary/secondary for master/slave.
The Inclusive Naming Initiative has said that “while master, in and of itself, is potentially neutral, the propensity in which it is associated with the term slave in computing makes master on its own guilty by association”.
The group recommends that companies adopt a replacement immediately, which may include main, origin, source or control plane.
Companies involved in the project will define the best practices and replacements with the wider tech community from the start of next year, which includes a discussion at a KubeCon community meeting. From the third quarter of 2021, the initiative will showcase case studies and success stories as well as big wins at KubeCon 2021. Towards the end of next year, the Naming Initiative will then develop a certification programme if there is enough demand.
