Hi Martin,

it is fascinating to see all those areas of your work in review. I'm very grateful that you took the block-encryption topic under your wings. Now I am as eager as you to see it become a natural part of Sculpt, similar to how we enjoy the NIC router today.

My plans so far for 2021

• Finish making NIC drivers restartable in Sculpt

• Establish new API for network sessions (NIC, Uplink) inspired by the

modern Block-Request APIs in order to simplify the use of network sessions and eliminate corner cases

• Graphical front end and simpler integration in general for the

Consistent Block Encrypter in Sculpt

• Write a comprehensive CBE documentation

• Revive the Spunky project and bring it into Genode mainline

• Enhance Ada/SPARK runtime support and use Ada/SPARK for further

fundamental code

I find all those points worthwhile to pursue. Regarding the Ada/SPARK direction, I sense that we'll need to ways to infect more Genodians with it. Personally, after a brief stint, I haven't managed to keep up with it unfortunately.

Independently from the roadmap discussion, I would appreciate the brainstorming of areas where Ada/SPARK would be naturally applicable in Genode. E.g., with CBE we use Ada/SPARK as a core of a VFS plugin. Maybe this could be generalized such that this language becomes a go-to choice for implementing other parts of the storage stack (file system, cache)?

Cheers Norman