Hi Stan, hi all,
Am 19.04.2016 um 17:30 schrieb Stan Ratliff <ratliffstan@xxxxxxxxx>:
Lotte,
On Tue, Apr 19, 2016 at 11:03 AM, Lotte Steenbrink
<lotte.steenbrink@xxxxxxxxxxxx <mailto:lotte.steenbrink@xxxxxxxxxxxx>> wrote:
Hi all,
Am 19.04.2016 um 02:49 schrieb Stan Ratliff <ratliffstan@xxxxxxxxx
<mailto:ratliffstan@xxxxxxxxx>>:
Hello AODVv2 authors/editors:
OK, I can finally give some more details about the future of the draft.
Long story short:
1. Alvaro has given Justin and i a 6-week window to re-charter the WG, or
he's going to shut it down.
2. His current thinking is that we (the chairs) CANNOT proceed with the
re-charter unless/until AODVv2 is "disposed of". Disposed of means one of
(a) rejecting the doc, releasing it to the authors to pursue as an
AD-sponsored draft if they so desire. (b) Making the draft Experimental
instead of standards track, which lowers the bar, and issuing WGLC in that
mode,or (c) a *SUCCESSFUL* WGLC on Standards Track.
3. I've been in discussion with Justin. His current thinking is that we
should WGLC the doc, no later than 4/20 (Wednesday). It's something of an
open question whether to WGLC the doc as Standards Track, or Experimental.
4. There's another open question as to whether Alvaro will allow us to
start re-charter work while the WGLC is in progress. If his response is
"no", *IT WILL CHANGE THE CALCULUS*. The calculus *WILL* head towards
releasing the document - at this point, it's either AODVv2, or the whole
MANET WG.
5. If the WGLC is called, there will be specific criteria coming out of the
WGLC, which will determine the fate of the doc. If there is a "no
consensus" coming out of WGLC, I'd expect the document to be released back
to the authors.
I need your inputs on the notion of making the draft Experimental. It may
well save all of the work - as Justin stated, the "experiment" would be to
"get the protocol out into the public domain, to see if there is a
significant desire to adopt, and maybe to standardize in the future.“
Honestly? I’m not against it. Like Stan, my patience is just about gone and I
want to get this out of the door. Additionally, I'd be pretty effing angry if
multiple years of hard work went down the drain because we decided to be
stubborn and refuse to publish AODVv2 as anything but Standards Track.
The way I see it, AODVv2 is going to make standards track only with a lot of
good will (and there doesn’t seem to be much left of that in the WG):
- Even though two independent, interoperable implementations apparently
aren’t a requirement anymore, the one we do have was written for a bachelor’s
thesis and is neither feature complete nor currently up to date. (DYMO on
TinyOS[1] doesn’t count, it's 8 years old.)
- There are no records of „real world“ experience with AODVv2. I don’t know
of any AODVv2 deployments on actual hardware in a real/realistic environment
(arranging a few nodes on a lab table doesn’t count). Simulations that do
exist are several years old (and/or using AODV, not v2), and even if they
weren't, I wouldn’t count them as hard evidence and I doubt the IETF will.
Thomas took a stab at this with his “Can you tell us more about metrics used
in real AODVv2 deployments, wink wink” comments, and I don’t fault him for
that. If I were in his position, with his agenda (and maybe also the grudge
to match ;) ), I’d try to make that point as clear as I could.
The only thing that inspires confidence for me right now is the Model
Checking work done by Ms. Yousefi et al., but that isn’t very hands-on
either, is it.
But at any rate - I've been saying for quite a while that "we're running
out of runway". That time has arrived. There are no more extensions, no
more "just one more IETF". In some form or fashion, this has been going on
for 10 years. It ends now, or the WG ends.
I need your thoughts, and I need them quickly. We should also be prepared
to post the document no later than 1200 Wednesday, to start a WGLC.
Whoa. Okay. I can do that, but please bear in mind that the amount of changes
we’ve made so far isn’t that big. I’ll go through this weeks’ E-Mails today
and tomorrow and make appropriate changes where I can (feel free to bombard
my inbox with instructions), but Uni has started and unfortunately I’ve got
some coursework to do too. (Also, bear in mind that we’ll go into WGLC with
the multiplexer issue still unfixed because Chris hasn’t approved anything
yet. Alternatively, I can just implement what I’ve suggested to him a month
ago and see if somebody complains about my fix.)
I'd say go ahead and put your last suggestion to Chris into the doc. Let's
see if we can wrap-up a version of the doc ready for review to begin the
WGLC. I still want to aim for 1200 Eastern US time tomorrow for the post.
Looks like Justin is leaning towards Experimental.
Regards,
Stan
Best regards,
Lotte
[1]
http://tinyos.stanford.edu/tinyos-wiki/index.php/Tymo#TYMO:_DYMO_on_TinyOS ;
<http://tinyos.stanford.edu/tinyos-wiki/index.php/Tymo#TYMO:_DYMO_on_TinyOS>
Regards,
Stan
