Hi Med, > On Apr 14, 2025, at 10:48 PM, mohamed.boucad...@orange.com wrote: > > Hi Kent, > > We do already have the following in 8407: > > Identifiers SHOULD NOT carry any special semantics that identify data > modeling properties. Only YANG statements and YANG extension > statements are designed to convey machine-readable data modeling > properties.
You self-pushed for this and many other changes in 8407 bis. Consensus was weak. Now you are asserting said items without acknowledging that it is quiet common to use, e.g., "-type" (951 cases) and "-list" (241 cases). > I think we need to be consistent among what we are producing. A noble goal. Did you see my Technical Erratum idea? Now do whether lists should be wrapped by containers - possibly even less consistency there! ;) > I don't see any problem we are solving by adding -grouping to grouping names. I wrote about the problem below. That said, I don't believe anyone ever tried to understand my motivation. I'm not saying it's great, but that it was done for a reason, and no one cared to ask me what it was. > Cheers, > Med Kent // contributor > >> -----Message d'origine----- >> De : Kent Watsen <k...@watsen.net> >> Envoyé : mardi 15 avril 2025 02:36 >> À : Andy Bierman <a...@yumaworks.com> >> Cc : Michael Richardson <m...@sandelman.ca>; netmod@ietf.org >> Objet : [netmod] Re: Suggestion for rfc8407bis: Don't prefix >> identifiers with their type >> >> >> >> >>> On Apr 11, 2025, at 11:18 AM, Andy Bierman <a...@yumaworks.com> >> wrote: >>> >>> This 'bad' practice (foo-grouping) has been used in RFCs as >> recently as RFC 9640. >> >> No one seemed to care in the years the WG was working on these >> documents. >> >> Those documents are just recently published. How about filing a >> Technical Erratum to convert them all? The data model would be >> unaffected... >> >> >>> In general, avoiding redundancy is a good idea, but naming >> conventions >>> for different types of identifiers are quite common. >> >> Perhaps use "-g" instead of "-grouping"? >> >> The goal for the YANG to be readable. I created this convention in >> order to make it more readable, because otherwise it became >> confusing when "foo" could be a a substring found in many >> identifiers (module names, groupings, containers, etc.). I had >> issues trying to navigate the modules before, which resolved after >> introducing the typing convention. >> >> I personally think there is bike-shedding going on here, and the >> 8407bis guidance is overreaching. Strange how no one asked me why >> I did this, to seek for a solution that addresses the issue I ran >> into. >> >> >> Kent // contributor
_______________________________________________ netmod mailing list -- netmod@ietf.org To unsubscribe send an email to netmod-le...@ietf.org
