Re: [CCAMP] YANG prefix for Tunnel models (was RE: YANG prefix for flexi-grid (was RE: FW: Proposal for YANG model prefix naming))

[Tarek Saad <[e-mail]>]

Hi all,

Thanks for sharing your thoughts on this. Ideally, we can distinguish TE tunnel from other type of tunnels from the prefix – i.e. ‘tetnl’. However, if concern is long prefix then I am OK with ‘tnl’ too.

Regards,
Tarek

On 6/23/21, 8:09 AM, "Teas" <[e-mail]> wrote:


Hi Tom,

I agree with the argument against the hyphen, it makes sense. Regarding "tnl" vs "tn" I still believe "tn" could be too misleading.

Maybe "tnl" without hyphen could be a good compromise between length and comprehensibility?

BR
Daniele

> -----Original Message-----
> From: tom petch <[e-mail]>
> Sent: den 22 juni 2021 12:19
> To: Daniele Ceccarelli <[e-mail]>; Italo Busi
> <[e-mail]>; CCAMP <[e-mail]>; [e-mail]
> Subject: Re: [CCAMP] YANG prefix for Tunnel models (was RE: YANG prefix for
> flexi-grid (was RE: FW: Proposal for YANG model prefix naming))
>
> From: Daniele Ceccarelli <[e-mail]>
> Sent: 22 June 2021 10:20
> Hi all,
>
> If adding one character is not a big issue I would suggest to go for -tnl, as -tn
> might cause ambiguity with a lot of other items, e.g. transport network, transit
> node, termination node...while the only thing I can think of for "tnl" is tunnel.
> I would reject all the other options for the good reasons brought up by Tom.
>
> <tp>
>
> Bear in mind that a prefix can appear 10 or more times in the path of an
> augment so additional characters mount up.  The choice of the terse te: and
> tet:, whether by accident or design, was inspired.  This also argues against the
> hyphen in the prefix as there are likely to be lots of hyphens in identifiers coming
> after the prefix in a YANG path and having a hyphen before the colon of the
> prefix may confuse ie augment /nw:networks/nw:network/nw:node/tet:te
>                /tet:information-source-entry
>                /tet:connectivity-matrices
>                /tet:connectivity-matrix/tet:path-properties
>                /tet:path-route-objects/tet:path-route-object
>                /tet:type/tet:label/tet:label-hop/tet:te-label
>                /tet:technology:
>       is better than, hyphens inserted, augment
> /nw:networks/nw:network/nw:node/te-t:te
>                /te-t:information-source-entry
>                /te-t:connectivity-matrices
>                /te-t:connectivity-matrix/te-t:path-properties
>                /te-t:path-route-objects/te-t:path-route-object
>                /te-t:type/te-t:label/te-t:label-hop/te-t:te-label
>                /te-t:technology:
>
> Tom Petch
>
> BR
> Daniele
>
> > -----Original Message-----
> > From: CCAMP <[e-mail]> On Behalf Of tom petch
> > Sent: den 22 juni 2021 11:08
> > To: Italo Busi <[e-mail]>; CCAMP <[e-mail]>;
> > [e-mail]
> > Subject: Re: [CCAMP] YANG prefix for Tunnel models (was RE: YANG
> > prefix for flexi-grid (was RE: FW: Proposal for YANG model prefix
> > naming))
> >
> > From: CCAMP <[e-mail]> on behalf of Italo Busi
> > <[e-mail]>
> > Sent: 21 June 2021 19:52
> >
> > Let's continue the discussion about the prefix to be used for tunnel
> > models, adding also TEAS WG since this discussion impact also the
> > ietf-te model
> >
> > There have been some negative comments from Tom against the current
> > prefix (i.e., "te" which would also lead to "otn", "wson" and "flexg"
> > in CCAMP tunnel
> > models) which is broader than just the te-tunnel
> >
> > The letter 't' is not possible because it has been already used for
> > topology (i.e., "tet", "otnt", "wsont", "flexgt")
> >
> > The only option to abbreviate tunnel I have seen so far are tn or tnl
> > with or without the hyphen: "tetn", "te-tn", "tetnl" or "te-tnl"
> >
> > I have a slight preference for "te-tnl", which would lead to "otn-tnl", "wson-
> tnl"
> > and "flexg-tnl" in CCAMP, but I am open to other opinions or better
> > proposals
> >
> > I have only a strong preference to close this issue as quickly as
> > possible :)
> >
> > Any other opinion or better suggestions?
> >
> > <tp>
> > I am losing track of the context, which modules are involved, but
> > assuming it is WSON, OTN. FLEXG (MW?, DWDM?) then I would prefer
> > '-tn' to '-tnl'  as it is shorter  and to 'tn' as it is easier to
> > read. 'tl' 'to' can be misread as 't1' 't0' in some typeface. ('tx' I often use but
> usually as transport or transmit not tunnel).
> > 'tu' I reject because it is consonants that are wanted not vowels in
> > such abbreviations,
> >
> > There are plenty of other topo in TEAS but I do not know which if any
> > will have a matching tunnel.
> >
> > Tom Petch
> >
> > Italo
> >
> > > -----Original Message-----
> > > From: tom petch [mailto:[e-mail]]
> > > Sent: giovedì 1 aprile 2021 17:40
> > > To: Italo Busi <[e-mail]>; CCAMP <[e-mail]>
> > > Subject: Re: [CCAMP] YANG prefix for flexi-grid (was RE: FW:
> > > Proposal for YANG model prefix naming)
> > >
> > > From: CCAMP <[e-mail]> on behalf of Italo Busi
> > > <[e-mail]>
> > > Sent: 31 March 2021 16:49
> > >
> > > It seems that addressing this issue on a step by step has worked
> > > well for the topology models
> > >
> > > Let's then consider a second question/convention about the prefix to
> > > use for flexi-grid YANG model
> > >
> > > I think that "flexi-grid" (which would lead to "flexi-gt" for
> > > flexi-grid Topology) is a bit too long for a prefix
> > >
> > > I have not found any better option to shorten flexible other than
> > > flexi or flex. I think f is really too short to be meaningful and fx
> > > can be misunderstood as fixed as well
> > >
> > > Therefore, I think we can shorten it to either "flexig" or "flexg"
> > > (which would lead to "flexigt" or "flexgt" respectively for
> > > flexi-grid
> > > Topology)
> > >
> > > I have a slight preference for the latter option ("flexg" leading to
> > > "flexgt" for flexi-grid Topology) but I am open to other opinions or
> > > better proposals
> > >
> > > Any other opinion?
> > >
> > > <tp>
> > > I prefer flexg of the options you suggest.
> > >
> > > There is something un-English about flexig, it just sounds wrong,
> > > and flexigt is worse.
> > >
> > > (In passing, flex-g looks wrong to me with just one letter after the
> > > hyphen, it needs two).
> > >
> > > Tom Petch
> > >
> > > Thanks, Italo
> > >
> > > > -----Original Message-----
> > > > From: tom petch [mailto:[e-mail]]
> > > > Sent: lunedì 15 marzo 2021 13:00
> > > > To: Daniele Ceccarelli
> > > > <daniele.ceccarelli=[e-mail]>;
> > > > CCAMP <[e-mail]>
> > > > Subject: Re: [CCAMP] FW: Proposal for YANG model prefix naming
> > > >
> > > > From: CCAMP <[e-mail]> on behalf of Daniele
> > > > Ceccarelli <daniele.ceccarelli=[e-mail]>
> > > > Sent: 11 March 2021 16:38
> > > >
> > > > as previously anticipated during the CCAMP session today, we will
> > > > ask the RFC editor to update the YANG model prefix for the WSON
> > > > topology to
> > > "wsont".
> > > >
> > > > Many thanks for sharing your thoughts and participating to the discussion.
> > > >
> > > > <tp>
> > > > I await the minutes with interest!
> > > >
> > > > Meanwhile ccamp-flexigrid-yang is plain wrong.  Under IANA it
> > > > registers flexi- grid-topology whereas the when statements have
> > > > tet-flexig which I would characterise as ugly and uglier!  And I
> > > > seem to recall this is not my first post on the prefix in this I-D
> > > > which clearly is not ready for Last Call:-(
> > > >
> > > > Tom Petch
> > > >
> > > >
> > > > Daniele
> > > >
> > > > -----Original Message-----
> > > > From: CCAMP <[e-mail]> On Behalf Of Daniele
> > > > Ceccarelli
> > > > Sent: den 15 februari 2021 16:17
> > > > To: tom petch <[e-mail]>; Italo Busi
> > > > <[e-mail]>; 'CCAMP' <[e-mail]>;
> > > [e-mail]
> > > > Cc: [e-mail]
> > > > Subject: Re: [CCAMP] Proposal for YANG model prefix naming
> > > >
> > > > Thanks for your feedback Tom.
> > > >
> > > > Working group, other opinions?
> > > >
> > > > Thanks,
> > > > Daniele
> > > >
> > > > -----Original Message-----
> > > > From: tom petch <[e-mail]>
> > > > Sent: den 12 februari 2021 13:29
> > > > To: Italo Busi <[e-mail]>; Daniele Ceccarelli
> > > > <[e-mail]>; 'CCAMP' <[e-mail]>;
> > > > [e-mail]
> > > > Cc: [e-mail]
> > > > Subject: Re: [CCAMP] Proposal for YANG model prefix naming
> > > >
> > > > From: Italo Busi <[e-mail]>
> > > > Sent: 12 February 2021 09:02
> > > >
> > > > Thanks Daniele
> > > >
> > > > FYI: we are discussing these options also with TEAS experts:
> > > >
> > > > https://protect2.fireeye.com/v1/url?k=d3d01972-8c4b2077-d3d059e9-
> > > > 86959e472243-ddf7de26918206cd&q=1&e=586b5fcf-a971-4d25-81f3-
> > > > 4c64316f0395&u=https%3A%2F%2Fgithub.com%2Ftsaad-
> > > > dev%2Fte%2Fissues%2F125
> > > >
> > > > Let's see if we can get an agreement at least for these two drafts
> > > > which are in RFC queue:
> > > >
> > > > > draft-ietf-ccamp-layer0-types
> > > > > ietf-layer0-types (l0-types) --->(l0-types)
> > > >
> > > > I think we can agree with this prefix since all the proposals are aligned.
> > > >
> > > > > draft-ietf-ccamp-wson-yang
> > > > > ietf-wson-topology  (wson) --->(wson-topo)
> > > >
> > > > Here, I think we have two candidates: wson-topo and wsont
> > > >
> > > > I have a slight preference for wsont since it is aligned with tet
> > > > and nt prefix conventions used in RFC8795 and RFC8345 but I can
> > > > accept wson-topo (it could be seen as aligned with wson-tunnel or
> > > > wson-tnl prefix
> > > conventions).
> > > >
> > > > What do you think?
> > > >
> > > > <tp>
> > > > Since TEAS already has its tanks on the lawn, I would go for wsont.
> > > >
> > > > Adrian expressed a preference for ...topo but I think that wrong
> > > > as it gets too long IMHO as with ethtetopo or else eth-te-topo
> > > >
> > > > Tom Petch
> > > > Italo
> > > >
> > > > > -----Original Message-----
> > > > > From: Daniele Ceccarelli
> > > > > [mailto:[e-mail]]
> > > > > Sent: giovedì 4 febbraio 2021 09:14
> > > > > To: tom petch <[e-mail]>; 'CCAMP' <[e-mail]>;
> > > > > [e-mail]
> > > > > Cc: [e-mail]
> > > > > Subject: Re: [CCAMP] Proposal for YANG model prefix naming
> > > > >
> > > > > Tom, Adrian, all,
> > > > >
> > > > > The L0 types and the WSON topology drafts are now on hold. We
> > > > > can include them in the updated prefix naming.
> > > > > The RFC editor will delay processing these documents until the
> > > > > updated versions are available. The AD (John or Deborah) will
> > > > > need to approve the changes.
> > > > >
> > > > > This will also affect the IANA registries, they have been informed as well.
> > > > >
> > > > > BR
> > > > > Daniele
> > > > >
> > > > > -----Original Message-----
> > > > > From: CCAMP <[e-mail]> On Behalf Of tom petch
> > > > > Sent: den 3 februari 2021 17:44
> > > > > To: 'CCAMP' <[e-mail]>; [e-mail]
> > > > > Cc: [e-mail]
> > > > > Subject: Re: [CCAMP] Proposal for YANG model prefix naming
> > > > >
> > > > > From: Adrian Farrel <[e-mail]>
> > > > > Sent: 03 February 2021 12:12
> > > > >
> > > > > Hi Tom,
> > > > >
> > > > > >> Proposal for YANG model prefix naming.
> > > > > >>
> > > > > >> Radek and then Tom raised the issue of consistency in prefix
> > > > > >> naming based
> > > > > on
> > > > > >> the fact that the TE topology model uses 'tet' and the TE
> > > > > >> topology state model uses 'tet-s'
> > > > > >
> > > > > > I think that the starting point is a list of I-D/RFC and I see
> > > > > > some
> > > > > glitches in your list.
> > > > > >
> > > > > > draft-ietf-client-signal-yang probably should be
> > > > > draft-ietf-ccamp-client-signal-yang
> > > > >
> > > > > Yes, typo.
> > > > >
> > > > > > I see two flexigrid I-D but you only list one
> > > > >
> > > > > I'm at https://datatracker.ietf.org/wg/ccamp/documents/ looking
> > > > > at extant WG documents.
> > > > > Looks like draft-ietf-ccamp-flexigrid-media-channel-yang expired
> > > > > almost 6 months ago.
> > > > >
> > > > > > wson-yang and l1types have been approved by the IESG so I
> > > > > > regard those as
> > > > > fixed
> > > > > > points that it is now too late to change and which we should
> > > > > > build around
> > > > >
> > > > > Well, colour me confused.
> > > > > I thought this whole thing came up in debate of the WSON YANG model.
> > > > > If that debate is now closed, let's all move on and not worry
> > > > > about any of this any more.
> > > > >
> > > > > > I have seen more than one wson model
> > > > >
> > > > > There's an information model in draft-ietf-ccamp-wson-iv-info,
> > > > > but no data model.
> > > > > draft-ietf-ccamp-wson-tunnel-model expired almost 6 months ago
> > > > >
> > > > > > microwave seems to be missing
> > > > >
> > > > > draft-ietf-ccamp-mw-topo-yang expired almost 18 months ago
> > > > >
> > > > > <tp>
> > > > >
> > > > > Adrian,
> > > > >
> > > > > The progress of I-D in the routing area can be erratic.  The
> > > > > fact that the IETF has expired the I-D does not mean that it
> > > > > will not come back to life - a whole raft of I-D that were
> > > > > produced in a rush just before the IETF meeting have just
> > > > > expired 6 months later and some are now being resuscitated,
> > > > > others will be in future, others will not.  Some re-appear
> > > > years later when their time has come.
> > > > >
> > > > > To me, an expired draft says that someone was interested enough
> > > > > to put in a lot of work and even if that work is not current,
> > > > > then it would be a short- sighted naming convention, although
> > > > > well in keeping with the traditions of the IETF, not to cater for such work
> in future.
> > > > >
> > > > > For myself, I like names that start with the most important
> > > > > property and for me, that is WSON. OTN, RSVP and so on, and that
> > > > > is the basis on which I reviewed them, and not the fact that
> > > > > they are te - the rival proposal is for te to be the centre of
> > > > > the universe around which everything revolves, regardless of
> > > > > which WG
> > > > > - TEAS, CCAMP, ... -it
> > > may be in.
> > > > I am not a fan of this approach.
> > > > >
> > > > > Tom Petch
> > > > >
> > > > > CCAMP is currently working on plenty of YANG models, so it might
> > > > > be worth stepping back and getting the prefixes consistent
> > > > > across all of our
> > > > work.
> > > > > I'm not sure this is the most important thing on our list, and
> > > > > perhaps it would be better to discuss the colour of the bike
> > > > > shed, but to make sure that we do this just once, here is my attempt.
> > > > >
> > > > > My conclusion is that, although it would be nice to be
> > > > > consistent with using just a suffix of 't' to indicate
> > > > > 'topology', this becomes messy with some of the longer names,
> > > > > and it is clearer to always
> > use 'topo'
> > > > > (leaving the TE topology model as the odd one out).
> > > > >
> > > > > This proposal only extends to CCAMP YANG models, and I don't
> > > > > think this list can debate the wider scoping of prefixes, but I
> > > > > think it would extend well enough.
> > > > >
> > > > > The list shows...
> > > > > Draftname
> > > > > Modelname (currentprefix)--->(proposedprefix)
> > > > >
> > > > > draft-ietf-ccamp-client-signal-yang
> > > > > ietf-eth-tran-service (ethtsvc)--->(etht-svc)
> > > > > ietf-eth-tran-types
> > > > > (etht-types) ---
> > > > > >(etht-types) ietf-trans-client-service (clntsvc)
> > > > > >--->(tclnt-svc)
> > > > > >ietf-trans-client-
> > > > > svc-types (clntsvc-types) --->(tclnt-svc-types)
> > > > >
> > > > > draft-ietf-ccamp-dwdm-if-param-yang
> > > > > ietf-ext-xponder-wdm-if
> > > > > (ietf-ext-xponder-wdm-if)--->(ext-xponder-wdm-if)
> > > > >
> > > > > draft-ietf-ccamp-flexigrid-yang
> > > > > ietf-flexi-grid-topology (flexi-grid-topology)
> > > > > --->(flexi-grid-topo)
> > > > >
> > > > > draft-ietf-ccamp-l1csm-yang
> > > > > ietf-l1csm (l1csm) --->(l1csm)
> > > > >
> > > > > draft-ietf-ccamp-layer0-types
> > > > > ietf-layer0-types (l0-types) --->(l0-types)
> > > > >
> > > > > draft-ietf-ccamp-layer1-types
> > > > > ietf-layer1-types (l1-types) --->(l1types)
> > > > >
> > > > > draft-ietf-ccamp-optical-impairment-topology-yang
> > > > > ietf-optical-impairment-topology (optical-imp-topo)
> > > > > --->(optical-imp-topo)
> > > > >
> > > > > draft-ietf-ccamp-otn-topo-yang
> > > > > ietf-otn-topology (otntopo) --->(otn-topo)
> > > > >
> > > > > draft-ietf-ccamp-otn-tunnel-model ietf-otn-tunnel (otn-tunnel)
> > > > > --->(otn-tunnel)
> > > > >
> > > > > draft-ietf-ccamp-wson-yang
> > > > > ietf-wson-topology  (wson) --->(wson-topo)
> > > > >
> > > > >
> > > > > _______________________________________________
> > > > > CCAMP mailing list
> > > > > [e-mail]
> > > > > https://www.ietf.org/mailman/listinfo/ccamp
> > > > > =
> > > > >
> > > > >
> > > > > _______________________________________________
> > > > > CCAMP mailing list
> > > > > [e-mail]
> > > > > https://www.ietf.org/mailman/listinfo/ccamp
> > > > >
> > > >
> > > > _______________________________________________
> > > > CCAMP mailing list
> > > > [e-mail]
> > > > https://www.ietf.org/mailman/listinfo/ccamp
> > > >
> > > > _______________________________________________
> > > > CCAMP mailing list
> > > > [e-mail]
> > > > https://www.ietf.org/mailman/listinfo/ccamp
> > > >
> > >
> > > _______________________________________________
> > > CCAMP mailing list
> > > [e-mail]
> > > https://www.ietf.org/mailman/listinfo/ccamp
> >
> > _______________________________________________
> > CCAMP mailing list
> > [e-mail]
> > https://www.ietf.org/mailman/listinfo/ccamp
> >
> > _______________________________________________
> > CCAMP mailing list
> > [e-mail]
> > https://www.ietf.org/mailman/listinfo/ccamp

_______________________________________________
Teas mailing list
[e-mail]
https://www.ietf.org/mailman/listinfo/teas