Topics

Yocto Project Future Direction(s)


Richard Purdie
 

Hi,

The YP TSC has been discussing the topic of future development
directions for a while. We're written up a summary of those onto the
wiki:

https://wiki.yoctoproject.org/wiki/Future_Directions

These have been shared and discussed with the YP members. Most of these
topics are resource constrained, we believe them to be valuable but we
don't have the right people with the time to spend to make them happen.

Our aim is that if/as/when there are resources available we'd move
forward in these areas. We wanted to try and combine together the TSC's
thoughts, the current status and tribal knowledge on these topic areas
into one place.

It is an evolving document. If anyone does want to discuss any of these
areas further or contribute, please do!

Cheers,

Richard
(on behalf of the YP TSC)


Rich Persaud
 

On Jul 28, 2020, at 10:32, Richard Purdie <richard.purdie@...> wrote:

Hi,

The YP TSC has been discussing the topic of future development
directions for a while. We're written up a summary of those onto the
wiki:

https://wiki.yoctoproject.org/wiki/Future_Directions

Thanks for making this available!  

These have been shared and discussed with the YP members. Most of these
topics are resource constrained, we believe them to be valuable but we
don't have the right people with the time to spend to make them happen.

Our aim is that if/as/when there are resources available we'd move
forward in these areas. We wanted to try and combine together the TSC's
thoughts, the current status and tribal knowledge on these topic areas
into one place.

It is an evolving document. If anyone does want to discuss any of these
areas further or contribute, please do!

Would backward compatibility (e.g. via buildtools-extended-tarball) be covered by the "Other future topics" section on multiple toolchains?  

Is it worth adding hash equivalency to one of the roadmap sections, or is that an implict capability required by several roadmap items?

On Layer setup/config, it may be useful to work with other communities by proposing an enhancement to upstream git, applying the experience from multiple OE approaches as case studies and sources of requirements. While it would take longer, it would reduce the risk of "now we have N+1 problems".

For QA automation, Code Submission and Usability topics, there was good discussion at LPC 2019 (https://lwn.net/Articles/799134/) and a subsequent mailing list,  https://lore.kernel.org/workflows/.  There are upcoming safety, security & testing tracks for LPC 2020, https://www.linuxplumbersconf.org/event/7/page/80-accepted-microconferences, for collaboration with upstream efforts.

Rich