Terry and Adrian,
Can't we simple update the mailing list charter as Terry explained he
see the purpose of the list, and then go on with the discussion.
/Loa
On 2017-01-17 11:33, Terry Manderson wrote:
Hi Adrian,
Avoidance of discussion happens everywhere on many topics in the IETF, and as we
know silence does not equal consensus. We also all know why that occurs. Having the
list specifically for this topic is to ensure that sane discussion does happen, and
happens in a location that easy for anyone (esp the IESG) to find at some point down
the track when (invariably) an ask comes forward about a having BOF. So yes, the
interested parties do need to self-organize to get that "agreed' definition.
Please don't read any weight into the term "agreed IETF definition". This
mailing list does not get to say this is THE definition for the IETF. the wording is to
scopes the discussion such that the definition needs to be in scope for the IETF. Could
it have been said better, yep. Mea culpa. (Although I really thought how the IETF forms
consensus is well known. Will ensure to be far more explicit in future)
Where it goes from here depends on what happens on the mailing list and that is
simply no different to most of the other bodies of work that comes into the
IETF.
Cheers
Terry
(Dammit, the single malt is on the top shelf!)
On 17/01/2017, 1:10 PM, "Adrian Farrel" <adrian(_at_)olddog(_dot_)co(_dot_)uk>
wrote:
Hello Terry,
This all seems fine and dandy, but there is a disconnect between what you say there "discussing a
proposed definition" and "presenting a better formed definition to the IETF" (admirable
intentions for a mailing list), and "determining an agreed IETF definition" as indicated by some of
the participants on the list as well as in the revised mailing list announcement.
I hope I am not needlessly picking at words, but it seems that this
disconnect could be the source of some future uncomfortable moments.
As Lou mentions, the topic seems to have avoided discussion on the TEAS list where it
was invited to take place. That need not be a bad thing if the proponents need to
self-organise a bit. And I see no harm in providing an archived mailing list under IETF
"note well" terms for that self-organisation to take place. It's just about
setting expectations of where the results of the organisation need to be taken.
Cheers,
Adrian
(The benefits of a low bar include being able to reach your drink while in
a sedentary position)
> I granted approval of this list, as a list, so that the definition of
network slicing
> could be discussed in an IETF context.
>
> Really, it boils down to the (on list) discussion of what is a proposed
network
> slicing definition that could see the IETF doing work on. So really about
presenting
> a better formed definition to the IETF, for the IETF to consider at some
future
> point in time. As we know with most "I have a problem that I think the
IETF
> should work on" proposals we tend to ask for the problem to be defined in
a way
> that does communicate the depth and breadth of the issue or the idea
before a
> BOF is considered. This is where I see network slicing now. Showing that
this very
> amorphous concept has the hope of some agreed shape and also that there
are
> sufficient bodies to form that shape, whatever it is.
>
> As a mailing list (and JUSTa mailing list!) the work for the interested
parties on
> that mailing list is to try to put words together that is actually
meaningful in the
> IETF context. To be brutally honest I have doubts that this is possible
from what I
> read to date but I do commit (as AD) to allowing discussion to occur as
I'm neither
> the magistrate of taste nor the gate of interest.
>
> Cheers
> Terry
>
> On 14/01/2017, 3:37 AM, "ietf on behalf of Adrian Farrel"
<ietf-bounces(_at_)ietf(_dot_)org
> on behalf of adrian(_at_)olddog(_dot_)co(_dot_)uk> wrote:
>
> Thanks Stewart and Alex.
>
> > This list is intended for discussion of network slicing to determine
> > an agreed IETF definition of the term Network Slicing, problems and
> > gaps to be covered with an aim to facilitate interoperation across
> > different operator and vendor solutions. The list also determines
(and
> > assimilates) which elements of the slicing problems are already
> > covered by existing IETF designs or work in progress.
>
> It's good to discuss stuff.
>
> How will agreement of "an IETF definition" be measured?
> Or maybe you mean to attempt to agree a definition among the people
> subscribed to the list and propose that as a definition for use by the
IETF?
> But still, who on the list will call consensus?
>
> Why is this something to be petty about?
> Because I need to know whether this is a list I have to join and
monitor in case I
> don't agree the definition, or whether that definition will come up for
IETF review
> in the normal way.
>
> Perhaps the AD who granted this list with this charter could speak up?
>
> Adrian
>
>
--
Loa Andersson email:
loa(_at_)mail01(_dot_)huawei(_dot_)com
Senior MPLS Expert loa(_at_)pi(_dot_)nu
Huawei Technologies (consultant) phone: +46 739 81 21 64