The open source community has made its mark on the software development space through its one-for-all, all-for-one group think and pace of innovation that no single enterprise could ever hope to match. This has allowed open source-based platforms to spread quickly through the market and become the basis for much of the advancements across the software-defined ecosystem.
Tidelift, which provides support services for open source projects, noted in a recent survey that 44% of organizations were likely to use more open source software during the pandemic and ensuring economic slowdown. Perhaps more shocking was that just 2% said they expect less open source software usage.
The survey also found that finances was the biggest reason for firms to increase their use of open source software.
“This has always been one of the enticing benefits of open source — that developers move more quickly because they can choose from billions of lines of code that have already been written versus having to start from scratch,” the report stated.
But like just about everything in 2020, that pace and mindset has been impacted by outside forces. This can be seen by what has been a slackening in the pace of some of that development as well as a re-thing as to how the ecosystem views some of the language it uses in pushing that development.
Kubernetes Slow DownThat slowdown was highlighted by the Kubernetes project moving from its standard quarterly release cycle this year due to police protest that were tied to the murder of George Floyd. Instead of pushing the 1.19 release out at mid-year, the Kubernetes project delayed the release by two months.
Taylor Dolezal, lead on the Kubernetes 1.19 Release, in an email to SDxCentral explained that those protests, along with broader support for the Black Lives Movement and the overarching COVID-19 pandemic had to be factored into how the project moved forward.
“I think it was just a lot to deal with, and I liked that we were able to come together and agree on hey, let’s take a little bit more time in this release and not burn ourselves out. Because being open source it’s definitely community focused, and it was nice to not push those people too much or stress them out,” Dolezal said.
At the time of the 1.19 release, Dolezal explained that the subsequent 1.20 release would be the last one scheduled for the year and that he expected the project to return to its three-month iteration cycle in 2021. That 1.20 release is indeed set to be the final 2020 release, however, the Kubernetes project is reconsidering a return to its quarterly release cycle.
As part of the recent virtual KubeCon + CloudNativeCon North America 2020 event, one of the keynotes focused on reaching out to the community for feedback on how it should approach ongoing development.
Stephen Augustus, who is a senior open source engineer at VMware and was co-chair of the KubeCon event, took that route as part of his update on the Kubernetes project.
“The Kubecon keynote was essentially to call everyone to request feedback on changing the release cycle,” Augustus said in an interview with SDxCentral. He explained that the current model was to stick with the three-month release cycle for at least the first new release (1.21) of 2021, but that might change by mid-year to accommodate the proposal set out in his keynote.
“I definitely want to take at least at least a month to gather feedback,” Augustus said. “We've got some great feedback already, but we want to make sure that we do something that is sustainable for everyone who's consuming Kubernetes, which is really the ultimate goal.”
Open Source Community for ChangeAnother topic that the open source community is tackling is in how it uses terms to describe software terms. The Cloud Native Computing Foundation (CNCF) is heading up that effort as part of its Inclusive Naming Initiative.
The initiative was announced last month and is focused on replacing harmful language with neutral terms whenever possible.
“This includes but is not limited to language linked to racism, sexism, homophobia, transphobia, ableism, or discrimination against any underrepresented group,” CNCF noted in a statement. “Beyond this, the group is making strides to improve the clarity of codebases and documentation by replacing idioms, metaphors, and slang specific to the English language.”
Priyanka Sharma, GM of CNCF, explained that this includes terms like “master/slave” and “white list/black list,” among others. She noted that the events of 2020 “really made us think that if not now, then when.”
“It’s an incremental process that has been going on for a very long time,” Sharma said. “We entered into a particular moment of time, which the catalyst was the events that occurred in 2020.”
The initiative will see CNCF provide the framework for the processes to be used to replace harmful and unclear language with an agreed upon set of neutral terms. It has substantial backing from a number of CNCF members, including Cisco, IBM, Red Hat, and VMware.
Sharma noted that while the initiative does have broad support, it’s really about using the influence of the open source community to drive the change upward within an organization.
“Open source has really evolved over the past 10-plus years to become a mainstay of technologies and where a lot of innovation is happening,” Sharma said. “And so anything that we work on, we can actually collaborate with a larger ecosystem. And we have a language of communication already amongst us so anything that we're doing can spread wide and become a real solution as opposed to a one-off effort here and there. So I think the maturity and growth of open source and cloud native are also factoring into this awesome result.”