Re: [nfsv4] Milestone draft for working group review

[David Noveck <[e-mail]>]

> A simple list of criteria for entertaining an idea, wacky or
> otherwise, might be:

This seems to me more like a list of criteria for pursuing an idea.

We should be open to entertaining ideas that don't (yet) meet these,
as long as there is an identifiable idea to consider.

- Consensus around a problem statement

We certainly need a problem statement and it wold be helpful if
it clearly enough expressed that everybody understands it the same way.

I wouldn't like to have to poll people as to whether they agrre to any
particular view so i would stay awat from the word  "consensus" .  Of
course, if somebody says he doesn't understand what the problem is,
we should deal with that.

> - One or more clearly expressed proposed solutions

Yes.

> - Someone (or ones) identified as the driver

I agree but I thhink we need more clarity on what we are asking of\the
driver(s).

> - One or more WG members identified as reviewers or area experts

This is helpful.  We have had too many cases in which documents are
not fully reviewed unti a mad rush at last call.  We should be paying more
attention to getting early review of documents.

> With a full set of these, we have an idea of what is to be addressed
> and how; whether it is appropriate work for the WG to consider and
> help with; and whose specific elbow grease will be applied to get
> it published.

That is certainly enough for consideration as a potential milestone.

It might be past that point.  For example, you might consider this as a list
of criteria for making this a working group item.  So the question, is,
given these criuteria have been met, what else do you need to make
something
a working group item?

> At that point, the content amounts to a short I-D anyway.

At least a short one.

> Hence "Requires I-D" would be a category of potentially interesting
> ideas that are temporarily lacking one or more of the above criteria,
> though we expect all will eventually be provided in some form.

I don't see having this as a separate category.  You need the things
mentioned above and an I-D is a good way of providing them, but
having an I-d, does not, in itself, provide these things

> Essentially this is a list containing I-D placeholders.

It would contain those but it might also include I-D pointers, if I-D
an exists, but it isn't quite ready for some reason.

> You can change the name, if "Requires" has an overly demanding tone.
> Would "I-D Requested" be more inviting?

I'd handle it a different way.  The list/table would have a place for the
name
of the I-D or the designation "Not done (yet)".

I don't want to get into a required/requested/would-be-nice discussion.
The point is that your colleagues in the working group need this
information,
since they will want to know about your idea, and an I-D is one good way to
provide it.


On Thu, Aug 24, 2017 at 10:55 AM, Chuck Lever <[e-mail]>
wrote:

>
> > On Aug 22, 2017, at 8:15 AM, David Noveck <[e-mail]> wrote:
> >
> > > Our usual criterion for considering "wacky" ideas is a personal
> > > I-D.
> >
> > Personal I-D's also discuss non-wacky ideas,  The wackiness of
> > any particular idea is a subjective judgment that different people will
> > make differently and that will change over time, with working
> > group discussion.
> >
> > > I'm OK with placing these ideas in a "requires I-D"
> > > category.
> >
> > I don't anticipate that the list I produce will be organized accoring
> > to categories of that sort.  In any case, I expect that it will contain:
> >       • Ideas for which there is an I-D and those for which there is no
> I-D (yet).
> >       • Ideas that I consider "wacky",but I'll try to filter out those
> that the entire working group is liable to find wacky.  In any case,
> subsequent workig group discussion could result in elimination of most of
> those.
> > > I'd like to see these items and any clarifications discussed
> > > on-list
> >
> > I'll make sure to copy the list on any discussion.
> >
> > > first
> >
> > Before what?,
> >
> > > or (preferably) in I-D form.
> >
> > That's certainly preferabe but my ability to get other people to write
> > documents, or even read them, is quite limited.
> >
> > > Perhaps that's a
> > > high bar,
> >
> > It certainly is.
> >
> > > but the WG deserves a clear problem statement and
> > > explanation of a proposed solution, so that it's members can
> > > evaluate new ideas fairly.
> >
> > The question is at what point they actually need that.  I think it
> > is appropriate for the working group to insist on an I-D before embarking
> > on a working-group document and lately it has been doing that.
> >
> > With regard to expectations for inclusion on the potential milestone
> > list, I don't think it makes sense to make the bar that high.  Doing that
> > would filter out too many "wacky" ideas before the working group has
> > had an opportunity to decide that they were not so wacky after all
>
> My opinion is based on recent experience with trying to assess the
> content of Tom's IETF 99 "Next Steps" slides. Appealing topics,
> all! But as I discussed these with various other members of the
> Working Group, it became clear that each of us saw something mildly
> or even vastly different in each bullet point. An I-D provides a
> locus for consensus about each idea.
>
> The other important piece of information that an I-D immediately
> documents is it identifies stakeholders and those who are willing
> to pursue the publication of the content.
>
> A simple list of criteria for entertaining an idea, wacky or
> otherwise, might be:
>
> - Consensus around a problem statement
>
> - One or more clearly expressed proposed solutions
>
> - Someone (or ones) identified as the driver
>
> - One or more WG members identified as reviewers or area experts
>
> With a full set of these, we have an idea of what is to be addressed
> and how; whether it is appropriate work for the WG to consider and
> help with; and whose specific elbow grease will be applied to get
> it published.
>
> At that point, the content amounts to a short I-D anyway.
>
> Hence "Requires I-D" would be a category of potentially interesting
> ideas that are temporarily lacking one or more of the above criteria,
> though we expect all will eventually be provided in some form.
> Essentially this is a list containing I-D placeholders.
>
> You can change the name, if "Requires" has an overly demanding tone.
> Would "I-D Requested" be more inviting?
>
>
> --
> Chuck Lever
>
>
>
>