[manet] Re: [EXTERNAL] Re: draft-dearlove-manet-olsrv2 -responsive-03

I see Chris’s point because these technologies are so detailed that it is impossible to capture all details in a single document that stands unchanged for all time without future amendments. I do agree with Abdussalam that a new protocol should strive to be as complete as possible in its first edition, but there comes a time when it needs to be declared “done” even though future amendments may already be in the works.

Thank you - Fred

From: Christopher Dearlove <[email protected]>
Sent: Thursday, March 19, 2026 3:29 PM
To: Abdussalam Baryun <[email protected]>
Cc: [email protected] List <[email protected]>
Subject: [EXTERNAL] [manet] Re: draft-dearlove-manet-olsrv2-responsive-03

EXT email: be mindful of links/attachments.


One document is simply neither possible not desirable, for reasons noted.

Yes, there could be a little bit of combination. Not much, but a bit. But it’s not hard when checking an RFC to see the notes about what updates it. And it’s standard practice.

And it wasn’t AODVv2/DYMO that was the original motivation for separating NHDP as a protocol in its own right, not jut a part of OLSR. Several people wanted it, among them being the authors of SMF. And that separation exists - it’s baked into lot of places (IANA registries for example). Trying to revere that would be serious work, might cause people some problems, and addresses a non-problem.

Unfortunately the implementation I wrote - which has everything in it, even RFC 7859 and work beyond that - is not publicly available. I know of one other possibility (not written by me) that I’ll check, but don’t expect anything from.


On 19 Mar 2026, at 07:04, Abdussalam Baryun <[email protected]<mailto:[email protected]>> wrote:



On Tue, Mar 17, 2026 at 11:40 AM Christopher Dearlove <[email protected]<mailto:[email protected]>> wrote:
And as I said, but also saying more:

(a) Multiple drafts is standard operating practice in RFCs. I used OSPF as an example, but I doubt there are many widely used non-trivial protocols that are defined in single documents.

one doc is easier for the internet community to follow up this standard and easier to use.


(b) Having thought more, I don’t think it is practical. It was a WG requirement that NHDP was separated from OLSRv2 so the firmer could be used separate from the latter. Once that is done, we have a lot of documents (jitter, time TLV, baseline security) that apply to both and thus to avoid repetition need themselves to be separate. And documents like 7859 (number from memory) that I an example of security that is separate for technical and political reasons (the method used may date - or already be dated).

Yes that separation was ok for NHDP_doc and packetbb_doc, because there was an approach by this WG to have same manet_interface/manet_packet for both OLSR and AODV, however, IMHO the old participants failed to complete the AODV or failed to make the separation having its main benefit for other protocols.


(c) Even if there were some value in doing what combining is possible, there’s no one I can see with the expertise and inclination (or funding) to do it.

IMHO we in IETF MANET WG will not need now OLSRv2_expertise but we mostly will need an implementation that combines all updates for rfc7181, I tried to get an implementation that was from the open source community but that implementation did not work. So I will now start a new search for an implementation that holds all updates or even separate, or I will need to implement it separately starting at 7181 doc.

best regards
AB

_______________________________________________
manet mailing list -- [email protected]
To unsubscribe send an email to [email protected]