Re: [mdnsext] dnssdext charter

[Ralph Droms <[e-mail]>]

On Aug 27, 2013, at 5:05 AM 8/27/13, Kerry Lynn <[e-mail]> wrote:

> Ralph,
> 
> First, I'd propose that we put the proposed charter version(s) on github as
> Carsten has done: https://github.com/6lo/charter.  Would you like me to do
> that this evening?

Although I think github is more mechanism than we need, I can't find a good alternative for posting a current charter so, yes, please put the proposed charter on github.  I've attached the initial draft charter from the BoF request and a revised charter with additional WG logistical info.

> 
> Second, I believe the charter must include the base case that was added
> in the last draft to make it explicit that any solution MUST NOT strand the
> millions of mDNS servers (printers and the like) that have already been
> deployed.
> 
> I agree with Peter that we probably don't need to include multiple homenet
> scenarios in the charter, especially since we have no requirements from that
> WG yet.
> 
> I would propose that we order the scenarios according to administrative
> complexity, from base case to enterprise.  (It will be interesting to define the
> differences of enterprise dnssdext vs. fully managed DNS-SD.)

Please send text with a concrete proposal?

> 
> Lastly, are we set on the designation "dnssdext", and will it be in the Apps
> area?

As far as I know, we're going with dnssdext.  Other suggestions can be considered.  I expect it will be in INT area.

- Ralph

> 
> Regards, -K-
> 
> 
> On Tue, Aug 27, 2013 at 4:50 AM, peter van der Stok <[e-mail]> wrote:
> Ralph,
> 
> I would suggest to keep the 4 scenarios in the charter.
> In the draft lynn-mdnsnext-requirements a mapping from req scenarios to charter scenarios can be done.
> Remembering the discussions during the Bof, the req doc. may end up with more scenarios than its current 6.
> Possibly they can be classified and then mapped to the 4 charter scenarios.
> 
> Peter
> 
> 
> 
> Ralph Droms (rdroms) schreef op 2013-08-26 19:07:
> 
> I'm revising the draft dnssdext charter according to the discussion
> during the BoF in Berlin.  One issue that occurs to me that we didn't
> explicitly discuss during the BoF is the list of the deployment
> scenarios to be considered by the WG.  The draft charter includes a
> list of four scenarios:
> 
> a) Commercial enterprise networks
> b) Academic/educational/university campus networks
> c) Multi-link home networks, such as those envisaged by the
> HOMENET WG
> d) Multi-link/single subnet (mesh) networks, such as those
> described by the ZigBee Alliance Z-IP specification
> 
> while draft-lynn-mdnsext-requirements includes a list of six scenarios:
> 
> (A) Personal Area networks, e.g., one laptop and one printer.
> This is the simplest example of an mDNS network.
> 
> (B) Home networks, consisting of:
> 
> * Single exit router: the network may have multiple upstream
> providers or networks, but all outgoing and incoming trafic goes
> through a single router.
> 
> * One level depth: all links on the network are connected to the
> same default router.
> 
> * Single administrative domain: all nodes under the same admin
> entity.
> 
> (C) Like B but may have a tree of links behind the single exit
> router.  However, the forwarding nodes are almost self-configured
> and do not require routing protocol administrators.
> 
> (D) Enterprise networks, consisting of:
> 
> * Any depth of the forwarding tree, under a single administrative
> domain.  The large majority of the forwarding and security
> devices are configured.
> 
> (E) Higher Education networks, consisting of:
> 
> * Any depth of the forwarding tree, core network under a central
> administrative domain but leaf networks under multiple
> administrative entities.  The large majority of the forwarding
> and security devices are configured.
> 
> (F) Mesh networks such as RPL/6LoWPAN, multi-link but single prefix
> networks.
> 
> The list of scenarios from draft-lynn-mdnsext-requirements was the
> basis for discussion of requirements during the BoF.
> 
> We likely need to coordinate the list of requirements in the charter
> with the list in the draft-lynn-mdnsext-requirements.  The two lists
> are actually not that far apart; the requirements docs includes (A)
> which is not in the draft charter, and (B) and (C) could perhaps be
> combined into one scenario, matching c) from the charter.
> 
> I'm looking for consensus about how to proceed:
> 
> * Modify the charter to align with draft-lynn-mdnsext-requirements
> * Modify draft-lynn-mdnsext-requirements to align with the charter
> * Replace the specific list of scenarios from the charter with a
> pointer to the requirements document
> * Modify both draft-lynn-mdnsext-requirements and the charter to bring
> them into alignment
> 
> Note that I'm deferring consideration of specific edits to the
> scenarios, such as s/tree of links/arbitrary topology/ in (C) from
> draft-lynn-mdnsext-requirements.
> 
> - Ralph
> 
> _______________________________________________
> mdnsext mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/mdnsext
> _______________________________________________
> mdnsext mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/mdnsext
> 
> _______________________________________________
> mdnsext mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/mdnsext