On 30 Dec 2015, at 23:06, Jari Arkko <jari(_dot_)arkko(_at_)piuha(_dot_)net>
wrote:
Hi Olle,
One thing I’m missing is a continued focus on IPv6.
If you were referring to the blog, I had to choose a set of issues to
highlight. But I think you are referring to actual IETF or industry focus, so
lets talk…
I know, it’s hard writing a blog and I do understand that you couldn’t cover
all. But since you asked for feedback :-)
I think there has to be some pressure and ways to handle IPv6 bugs in
current RFCs outside of the normal process.
We have submitted several drafts to the dispatch and sipcore working groups,
all died because of a lack of interest and discussion in the group. The bugs
we found in the SIP protocol still remains and needs to be fixed in order
for SIP implementations to handle dual stacks properly - but I can’t find a
way to get this published following the normal process.
So we have bugs related to IPv6, we have a group of authors and the
documents still are not going anywhere because a lack of interest.
I know that we have several bugs related to TLS handling in the SIP protocol
as well. The handling of IPv6 issues doesn’t make me very positive about
even bothering with writing any drafts about them.
This sounds familiar from many contexts, in standards, products, and open
source :-)
Everybody is focused on adding new features, and making sure the thing
actually works under all circumstances gets sadly less attention.
Don’t I know it from the Open Source projects I’m involved in :-)
Yes, I am pessimistic. Yes, I am pretty sure there are other protocols that
have bugs related to both IPv6 and TLS handling. I think the IETF needs to
find a way to handle these even in low-energy working groups like the
SIPcore group.
I don’t think we can entirely ignore our normal process. And by “normal
process” I don’t necessarily mean the formal part of our process, but rather
the fact that the core function of standards is that you have multiple people
that agree on something. One or even a couple of people can’t be the only
ones looking at something if we want to have a standard. That being said, the
existence of a bug is a bad thing. It says something about the state or
deployment of that particular part of the standard.
However, I am not pessimistic. We often highlight issues in the IETF and
ensure that chairs and participants understand the importance, and then we
pull through. Let me talk to my fellow ADs about what we could about this
case. Do you have a list of drafts that you would like to move forward?
I will forward the list separately - but in general, I think there has to be
some issues that ADs need some pressure to move on and help generate momentum.
Security is one, IPv6 is another that spans a lot of working groups. Maybe we
need ambassadors that help these ADs and groups to focus on these issues. To
me, it feels wrong that the IETF doesn’t have focus and energy to fix IPv6
issues - what message does it give to the implementors?
Happy New Year!
/O