Re: [Netmod-ver-dt] prelim IETF 107 slides for version selection

["Sterne, Jason (Nokia - CA/Ottawa)" <[e-mail]>]

Hi Joe,

I like your point about the filename issue, but that's something for yang-module-versioning (not these slides).

Our version-selection slides do have an issue about numbering but it is about the schema-set name.

I added the bullet you mention to the last slide.

Jason


From: Joe Clarke (jclarke) <[e-mail]>
Sent: Friday, March 27, 2020 9:50 AM
To: Sterne, Jason (Nokia - CA/Ottawa) <[e-mail]>
Cc: Rob Wilton (rwilton) <[e-mail]>; [e-mail]
Subject: Re: [Netmod-ver-dt] prelim IETF 107 slides for version selection

Since I had the issues part, a few perhaps aids there.

For the file naming issue, I think showing an example of:

ietf-interfaces@2018-02-20

vs.

ietf-interfaces@1.3.0<mailto:ietf-interfaces@1.3.0>

Is one way to illustrate the challenge.  But smart parsers will deal with this easily.  I think the bigger risk is something like this:

vendor-example@2020-03-29

vs.

vendor-example@2020-02-31

Where the latter is the second vendor release of 2020 with build ID 31 (something silly, but doable with revision-labels).  Parsers that blindly assume something that looks like a date is may have other issues.  Sure, we could add verbiage that a revision-label can’t look like a date or a semver, but perhaps it’s easier to have:

vendor-example#2020-02-31

===

On the last slide, based on our discussion the other day, should we add a bullet that says, “DT prefers this to remain as one document in netconf”?

Joe


On Mar 26, 2020, at 18:23, Sterne, Jason (Nokia - CA/Ottawa) <[e-mail]<mailto:[e-mail]>> wrote:

Thx. See inline.
Jason

From: Rob Wilton (rwilton) <[e-mail]<mailto:[e-mail]>>
Sent: Tuesday, March 17, 2020 7:50 AM
To: Sterne, Jason (Nokia - CA/Ottawa) <[e-mail]<mailto:[e-mail]>>; [e-mail]<mailto:[e-mail]>
Subject: RE: prelim IETF 107 slides for version selection

Hi Jason,

Slides generally look good to me.

A few minor comments:

On slide 2:

·        Perhaps change “some on the client” to “minimal for the client”[>>JTS: ] changed

On slide 3:

·        I wasn’t sure that the second bullet completely scans.  Should it just be “It can also be used to allow clients to select between YANG schema defined by different organizations”?[>>JTS: ] Yes - changed

Between slides 4 and 5:

·        I think that it might be helpful to have an extra slide after 4, that covers the “selectable” leaf-list, and perhaps default too, both in configuration and <operational>
[>>JTS: ] Added new slide 5 - please take a look.

·        Having a diagram to show the relationship between packages, datastore-schema, schema-sets, might be helpful, but only if someone has the time to do them.
[>>JTS: ] I'll try to get some cycles to do that.  Basically that schema-sets are comprised of packages (or rather packages for each datastore).  But what do you mean by datastore-schema? That term is mentioned once in the version selection draft. Is it defined somewhere else?

Thanks,
Rob


From: Netmod-ver-dt <[e-mail]<mailto:[e-mail]>> On Behalf Of Sterne, Jason (Nokia - CA/Ottawa)
Sent: 13 March 2020 20:37
To: [e-mail]<mailto:[e-mail]>
Subject: [Netmod-ver-dt] prelim IETF 107 slides for version selection

Hi guys,

I put together some slides on the changes to version selection since the previous IETF in case someone needs to present these next week while I'm away (seems unlikely though).

Joe - can you take ownership of this preso for next week and add your "Issues"?

If we don't present before I'm back (Monday March 23) then I can help polish them further but I wanted to at least get this part to the DT.

Rgds,
Jason
<ietf-107-netmod-yang-ver-selection-update-v2.pptx>_______________________________________________
Netmod-ver-dt mailing list
[e-mail]<mailto:[e-mail]>
https://www.ietf.org/mailman/listinfo/netmod-ver-dt