Hi Charlie and all,
Am 24.12.2014 um 18:57 schrieb Charlie Perkins
<charles.perkins@xxxxxxxxxxxxx <mailto:charles.perkins@xxxxxxxxxxxxx>>:
Hello Lotte and all,
I've put a new intermediate revision on Github. I'd like to
publish it to the IETF directories with your approval.
It includes extensive changes from "node" terminology
to "address" terminology wherever appropriate, and gets
rid of the "Ndx" terminology in favor of the lists that
include null elements.
I'm still not convinced by the list terminology, to be honest. (For
the record: I'm not convinced by using Ndx either, because it refers
to 5444 packet internals that shouldn't be dictated or described in
detail by AODVv2, in my opinion. I hope my RFC5444 suggestions are
able to reflect how I think the language should be... More on that
below.)
Also, I've fixed some typos, as you can see in
https://github.com/Lotterleben/AODVv2-Draft/commit/f1fcc1902b90733faf0270af853c2f9a0f691b9a
. I hope I didn't fix anything that wasn't broken...
And a few newbie questions:
* I've noticed the text you added to the Sequence Number
(SeqNum) part of the Terminology section. If I understood it right,
the terminology section is meant to be a quick reference, but
shouldn't really contain instructions for the implementor, which is
why I would have looked for the information in the text you added in
6.4. Sequence Numbers. Am I missing something here?
* in section 6.6, you added the following text:
Let "Cost(R)", where 'R' is the route for which the Cost is to be
evaluated; the route table entry for R includes the information about
the metric type for R.
... Maybe it's just me, but I can't quite figure out what that
sentence means.
Lotte, if you have text for the purpose of improving the
description for supplying protocol elements to RFC 5444
(e.g., OrigAddr with OrigSeqnum, etc.) please let me
know. I am willing to help finish it once I get the basic
idea of what you want to do. It would be nice to put this
into the above-mentioned revision.
As I said in our hangout, I'd like to run my changes by the MANET
Mailing List before adding them to the draft.
Anyway, I wrote a little script that lets you run rfcdiff over
branches, showing how *one* file differs on two branches. It's in the
root directory of your git, and if you run
./aodvdiff.sh master packet_sketch txt/draft-ietf-manet-aodvv2-05.txt
You will get the rfcdiff html output showing all my changes. (I'm
rubbish at bash scripting, so I'm sure the script could be improved
greatly, but it should do the trick)
What I've done up until now was:
* added a sketch of a RREQ as I thought it might be useful to the
appendix
* added subsections describing existing addresses and TLVs (The text
would have to be smoothed out a lot, but the idea should be clear,
hopefully)
* moved some text around
Next up is going through sections 7 and up and adjusting them to my
changes.
Additionally, I was thinking if we could solve our quibbles with the
“Interface-ness” of RFC5444 with an additional draft that describes
guidelines on how to build a RFC5444 packet/message builder that is
optimized to AODVv2's needs? This way, we can keep the description of
AODVv2 RFC5444 packets as generic as possible in the AODVv2
document, but still provide packets optimized for AODVv2 to everyone
that needs them.)
Cheers,
Lotte
It would be nice to do this as a Christmas present to [manet].
Regards,
Charlie P.
