Document technical Pros/Cons with Grin Soft Forks

[lehnberg]

This issue tracks an asynchronous discussion about the technical pros/cons related to various approaches for Grin soft forks, with the purpose on converging on a list (or not) of changes that ought to be targeted for v5.0.0.

The post will be updated continuously as the discussion evolves.

Related discussion

• https://github.com/mimblewimble/grin-pm/blob/master/notes/20200414-meeting-development.md#5-becoming-more-soft-fork-ready
• https://forum.grin.mw/t/network-upgrades-hard-forks-on-grin-v5-0-0-and-beyond/7231
• https://github.com/mimblewimble/grin-pm/blob/master/notes/20200428-meeting-development.md#31-soft-fork-discussion-to-be-initiated-in-a-permanent-place

---

Changelog:

• May 04: First draft

[antiochp]

IIRC The key takeaway around making transaction kernels more softfork friendly was in introducing a vector for arbitrary on-chain data.

To support arbitrary future kernel variants we would, by definition, need to allow arbitrary data.