Re: [manet] I-D Action: draft-ietf-manet-dlep-lid-extension-01.txt

Sorry for the top-post, damn Outlook.

One thing I have considered is whether adding some worked examples to the text has value.  I have had to explain LIDs a couple of times to people who haven’t immediately realized that the extension solves their problem.  It’s always an issue with IETF drafts that they drill down to the raw protocols, often not explaining the problem space well (or perhaps it’s just my writing style).

My thoughts are, perhaps a section covering: “So you have a Layer 3 modem”, and “So you have a LTE/WiFi AP” and you want DLEP.  Obviously I hate writing text, I’m an engineer, but if people think it has value, I could find the time.

Opinions?

Rick

From: HvFK [mailto:hvfk62@gmail.com]
Sent: 05 February 2018 16:32
To: Rick Taylor; hrogge@gmail.com
Cc: manet@ietf.org
Subject: AW: [manet] I-D Action: draft-ietf-manet-dlep-lid-extension-01.txt

Hi Rick, Henning, all

I think with both the Link-Identifier-Data-Item and Link-Identifier-Length-Data-Item we have some good flexibility. I personally do not see a need for the flags. I am also happy to remove it.
Best Regards,  Georg
________________________________
Von: Rick Taylor<mailto:rick@tropicalstormsoftware.com>
Gesendet: ‎05.‎02.‎2018 14:14
An: hrogge@gmail.com<mailto:hrogge@gmail.com>
Cc: manet@ietf.org<mailto:manet@ietf.org>
Betreff: Re: [manet] I-D Action: draft-ietf-manet-dlep-lid-extension-01.txt
On Mon, 2018-02-05 at 13:00 +0100, Henning Rogge wrote:
> On Mon, Feb 5, 2018 at 12:56 PM, Rick Taylor
> <rick@tropicalstormsoftware.com<mailto:rick@tropicalstormsoftware.com>> wrote:
> > Hi All,
> >
> > On Mon, 2018-02-05 at 11:01 +0100, Henning Rogge wrote:
> > > Hi,
> > >
> > > I think I like the new text two, its easy to read and introduce
> > > the
> > > new concept quickly.
> >
> > Thank you (and Georg as well) very much!
> >
> > >
> > > The only thing I am concerned about is the "empty" flags field...
> > > do
> > > we already have an usecase for this?
> >
> > I must admit, I copy and pasted the text from elsewhere, and then
> > realized I had no flags.
> >
> > However, one possible flag could used to distinguish between Links
> > to
> > Destinations with an IP address (addressable destinations/ Layer 3
> > radio net) and Links to  Infrastructure (no IP address, just a
> > subnet).
> >  Although this extension doesn't need to differentiate, it might be
> > useful for the router to know the difference when used with other
> > extensions (hop control perhaps).
>
> Wouldn't it be easier to look at the prefix length of the IP?
>
> > > If not, we could remove it and introduce it (if necessary) in an
> > > extension... we could discover the existence of an (optional)
> > > flags
> > > field by comparing the length of the link-id TLV with the (known)
> > > value of the link-id-length TLV.
> >
> > Yes, if a flags field was needed later, a 'Link Identifiers Flags'
> > extension could introduce it legitimately.
>
> So what do you think, just drop the flags field?

I'm happy to drop them if no-one can think of a use for them.

Rick
_______________________________________________
manet mailing list
manet@ietf.org<mailto:manet@ietf.org>
https://www.ietf.org/mailman/listinfo/manet