Re: [mmox] Creating walled gardens considered harmful

[Lisa Dusseault <[e-mail]>]

Hi Morgaine,

As one of the people who highlighted the opportunity to break off
smaller pieces of work, I can add that I would also be concerned about
those areas working too independently.  When several efforts are all
IETF WG efforts, there's lots of opportunity for cross-over and
coordination, and we've done this many times before.  When there are
non-IETF efforts too, that gets even harder but not impossible.

As with so many things, successful coordination depends on individuals
--  WG chairs, editors, ADs but emphasizing participants who cross
over between efforts.

Lisa

2009/3/25 Morgaine <[e-mail]>:
> What an excellent BoF!
>
> I believe that MMOX can validly claim a complete victory from the day's
> events.
>
> With so many excellent formal presentations and an astounding number of very
> insightful contributions from the floor, ably mediated by very fair and
> effective chairing, I think we were blessed to succeed, and we did.
> Congratulations to everybody. :-)
>
> On the substance, this will probably take many weeks of analysis, but I
> would like to highlight here just one of the many insightful contributions
> that were made.  (Unfortunately I cannot pin down exactly who said it.)
>
> Splitting the work of MMOX into various implementation-oriented groups would
> be an excellent way of making progress, but it carries a severe danger:  it
> could create several non-interoperating walled gardens, if the groups work
> independently without a common view and purpose.
>
> While that might be better than no interop at all, it would be an
> inappropriate result for a single MMOX workgroup to endorse
> non-interoperation.  If we are to remain as one group, I would propose that
> there be an oversight team appointed as well, to ensure that sufficient
> extensibility be built into each subgroup to avoid total segmentation into
> walled gardens.
>
> The requirements for this need not be onerous, as it could be achieved
> through simple avoidance of all-or-nothing protocols.  As a specific example
> arising from the Weblin presentation, an isolated ecosystem of Second
> Life-like worlds should still be able to provide avatar data for use in
> Weblin worlds, despite there being no full interop.  This could be achieved
> quite simply by good decoupling of services and avoidance of service
> bundling.
>
> If we are consciously creating separate gardens, they should not be walled
> off.  Let us work towards achieving some degree of interop even among our
> subgroups, so that as a unified MMOX we can claim victory at the end as well
> as at the start of this process.
>
>
> Morgaine.
>
>
> _______________________________________________
> mmox mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/mmox
>
>