got several quick replies!
a few points:
1/ there seem 2 separable issues:
- whether AB's suggestion is something worth considering doing
- whether one believes it is feasible to do it.
Lets not mix the two.
2/ there seems an implicit assumption that, if a history record is written,
it'd only document one view (hence all the concerns about not being able to
agree)?
Maybe it is best to document different opinions if different opinions exist?
Some time down the road IETF can look back, not only see which opinion has a
better match to the reality (or even none of them), but also (hopefully!)
understand better where our thought has gone right/wrong.
3/ no question that doing this would take time and effort, just like anything
else that is worth doing. So the real question is: would this be worth trying?
Lixia
PS: would very much like to, but probably wont be able to keep up with this
discussion (and will miss next IETF to chat with people in person)
On Feb 3, 2013, at 10:28 AM, Sam Hartman <hartmans-ietf(_at_)mit(_dot_)edu>
wrote:
"Lixia" == Lixia Zhang <lixia(_at_)cs(_dot_)ucla(_dot_)edu> writes:
Lixia> 5218 is a general document; I believe what AB suggested is a
Lixia> historical record specifically for each WG: what you started
Lixia> with, what you went through, how you ended, what you have
Lixia> learned, both principles and lessons.
I'm not sure I've ever been involved with a WG where you could have
gotten consensus on any of the above enough to publish it. Nor can I
think of many WGs that have the excess energy to do this work. Even
getting consensus on a summary of where you ended up is quite
tricky. Because of who I'm responding to I'm imagining getting consensus
on a summary of where the RRG ended up on the discussion of well
whatever it is that they were discussing regarding id/loc split and all
that stuff.
WGs aren't particularly easier in this regard.
You could probably get consensus on which set of documents a WG
published in most cases, but if you added any context, say a discussion
about whether those documents are still good ideas, it becomes a lot
harder.
This has value. I've supported certain aspects of the ISD proposals, of
some things coming out of newtrk, etc. One aspect I've always disagreed
with is proposals of this type that proposed to be generally applicable
across the IETF.
I agree that writing down where we were, where we are, what happened in
between can be valuable. However it is quite expensive, and we need to
balance that.