Pretty sober take on soft forks arguing we should do infrequent soft forks that batch features together rather than frequent soft forks of few features. It sites two primary reasons for batching:
more soft forks increase potential for chain splits, bugs, etc.
there isn't consensus on activation methods or what to do when they fail
sidenote Q: Is there some better UI to read those mailing list threads? Pipermail UI is pain for the last 20 years...
Not yet. Something I selfishly want to do on SN. Email replies also have a tree structure
Pretty sober take on soft forks arguing we should do infrequent soft forks that batch features together rather than frequent soft forks of few features. It sites two primary reasons for batching: