On Wed, Aug 5, 2020 at 9:20 AM tom petch <[email protected]> wrote:
> Looking at an I-D with endless submodules, I am left thinking > > Costs > > greater in size > more complex to understand > harder to review > more likely to have undetected errors > > How about "confusion due to false sense of modularity"? People often think submodules work like modules (but they don't). There think a submodule should "compile" when validated on its own (that used to be true in YANG 1.0 but not in YANG 1.1). They think mix-and-match include submodule revisions can be selected by the server vendor just like importing modules (but that does not work). > Benefits > > > > What am I missing? I have only ever seen one such module in the IETF that > I can recall and it is draft-ietf-idr-bgp-model and AFAICT it would be > smaller, simpler, easier to understand if it were done with grouping > instead of submodule. > > Tom Petch > > Andy > _______________________________________________ > netmod mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/netmod >
_______________________________________________ netmod mailing list [email protected] https://www.ietf.org/mailman/listinfo/netmod
