Re: [Lurk] LURK: proposed charter for review

I think we are making a classic error here.

A standard should always have a use case driving it. But standards that are
designed to only meet the needs of one use case tend to end up in systems
that are more complicated and less powerful than those designed to provide
a particular function.

PKIX got the way it did because it is an agglomeration of point solutions
to narrow use cases.

On Sun, Jul 17, 2016 at 5:43 PM, Василий Долматов <vdolmatov@gmail.com>
wrote:

>
> > 17 июля 2016 г., в 16:06, Yaron Sheffer <yaronf.ietf@gmail.com>
> написал(а):
> >
> > Hi Thomas,
> >
> > For the record (and sorry for the late reply): I think it would be a
> terrible idea to standardize both approaches. I obviously prefer one of
> them, but I think it is way more important to have one single standard
> solution for the industry to use. If we form a working group, it will be up
> to the WG to decide which of the options solves best the use case in
> question.
> Ein Reiche, ein Volk, ein Fuehrer?
>
> Pfff..,,
>
> dol@
>
> >
> > Thanks,
> >    Yaron
> >
> > On 08/07/16 17:13, Fossati, Thomas (Nokia - GB) wrote:
> >> Hi Yaron,
> >>
> >> On 08/07/2016 14:22, "Lurk on behalf of Yaron Sheffer"
> >> <lurk-bounces@ietf.org on behalf of yaronf.ietf@gmail.com> wrote:
> >>> I agree that we should not preclude future extensions. But this can be
> >>> done with a variety of tools, including a simple protocol version
> >>> number.
> >> A version number works well in the context of a single interface, not if
> >> we want have multiple interfaces under the LURK umbrella (see below).
> >>
> >>> Do you think something more extensive is called for?
> >> At the moment we have two main competing solutions: yours (let's call it
> >> "cert-delegation") and Daniel's/Rich&Sam's (the "tls-signing-box").
> >>
> >> I'm not sure whether only one will survive, or both will be standardised
> >> as LURK interfaces?
> >>
> >> If the latter, I'd probably want to have a way to know whether my edge
> >> cache has to talk using the "cert-delegation" and/or "tls-signing-box"
> >> interface with content provider X without having to turn knobs here and
> >> there to make it happen :-)
> >>
> >> Cheers, t
> >>
> >
> > _______________________________________________
> > Lurk mailing list
> > Lurk@ietf.org
> > https://www.ietf.org/mailman/listinfo/lurk
>
>
> _______________________________________________
> Lurk mailing list
> Lurk@ietf.org
> https://www.ietf.org/mailman/listinfo/lurk
>
>