Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: [802SEC] Proposal for Editions



All

Just to throw in my general agreement:

1) Consolidations should be made when requested by the WG Chair. The reality is that you couldn't get more than 1 or 2 a year with a 4 month request and time to review the document. It should not be limited to revisions only. For extremely active WGs (802.11 would be an example of this), 2 consolidations a year would be about right to assist the balloters.

2) I think the WG Chair should appoint a review committee to review the consolidation. Does this need to be in the SA policy versus 802 or WG operations manual? I am open to suggestions here. One reason for the review committee is that there can be questions in the roll-up that may need to be answered. Also, the review committee may find errata or corrigendum that need to be handled quickly in the process of the review. They can start these projects quickly and finish them quickly as well to avoid problems for subsequent amendments. IMHO.

3) Distributing the standard amendment format and consolidation format would be nice, but then we would need a policy for which document is the "official" standard. I suspect that the answer is that the amendment format is official as the consolidation has not been voted on by the Sponsor.

4) We should clarify that consolidations enter the Get IEEE 802 program 6 months after publication. Thus, if the SA does 2 consolidations a year for an 802 WG (say for 802.3, 802.1 and 802.11), they would always have something to sell. :)

Flame away, it has been too calm on the list anyway.

James Gilb

On 10/06/2011 10:25 AM, Geoff Thompson wrote:
Pat-

As I said earlier, it is perfectly appropriate for the circulated ballot
draft to only include those clauses of the consolidated edition that are
within the scope of the ballot.
I believe that addresses your concern.

GeoffT

On 610//11 9:16 AM, Pat Thaler wrote:
In many cases, our base standards are very large and any given
amendment changes only a small portion of them. I'd find balloting a
consolidated redline to be very burdensome and not likely to focus
attention on the changes. I speak from recent experience balloting the
changes in the current 802.3-Rev project. I wouldn't want to go
through that pain on every amendment ballot.

Automated diffs sometimes don't produce very workable documents so
getting a good redline isn't always insignificant.

But having a consolidated document with the prior amendments is very
helpful when reviewing drafts. This is particularly true when the
amendments have touched large portions of the base text. It got pretty
difficult to review 802.1Q amendments during the period when the
provider bridging amendments hadn't been consolidated.

Regards,
Pat

-----Original Message-----
From: ***** IEEE 802 Executive Committee List *****
[mailto:STDS-802-SEC@ieee.org] On Behalf Of Grow, Bob
Sent: Wednesday, October 05, 2011 9:04 AM
To: STDS-802-SEC@LISTSERV.IEEE.ORG
Subject: Re: [802SEC] Proposal for Editions

As some may recall, I have been advocating publishing all amendments
and corrigenda as consolidations (editions) for some time. In my
proposals, I have recommended that WGs be allowed to decide if an
amendment/corrigendum is balloted as a redline or as change
instructions (as is currently done).

Publishing of redlines I agree is a valuable product and something
that IEEE-SA is now selectively offereing. Since these can be done as
diffs, I don't think it is a significant workload for editorial staff,
and probably less than the work of publishing a change instruction
amendment and then later having to merge the amendment for a revision.

Personally, for 802 I think a bundle of redline and clean versions
would be an appropriate product to serve the needs of implementers.

I agree balloting as redlines also is often valuable. A speed increase
or new PHY type for 802.3 is not likely to show as much benefit from a
redline ballot as would energy efficient Ethernet which is much better
reviewed in context (as you point out for much 802.1 work). In a very
large document though, a complete redline is difficult to find
scattered changes, so only supplying changed clauses or a changed
clause list might be necessary for a redline ballot.

The only point where I find using consolidated redlines problematic is
when amendment projects are very parallel (e.g., in ballot at the same
time).

--Bob

-----Original Message-----
From: ***** IEEE 802 Executive Committee List *****
[mailto:STDS-802-SEC@ieee.org] On Behalf Of Tony Jeffree
Sent: Wednesday, October 05, 2011 7:33 AM
To: STDS-802-SEC@LISTSERV.IEEE.ORG
Subject: Re: [802SEC] Proposal for Editions

I would agree with that too.

Regards,
Tony


On 5 October 2011 15:18, Geoff Thompson<thompson@ieee.org> wrote:

All-

I would focus on a specific aspect of this.
When a standard has a non-trivial number of approved parts it is
extremely
difficult for voters to evaluate additional drafts.
Having to consider a large number of documents when voting on drafts
is a
significant disservice to our balloting reviewers.
So, again, just having consolidations a base text for revisions doesn't
deal with the critical issue in 802.

Regards,
Geoff


On 510//11 5:44 AM, Tony Jeffree wrote:

I should have added to this...in WGs where a number of amendments to a
single base standard are being processed simultaneously, as has been
the
case in 802.1 for some while, the availability of an up-to-date
consolidation of all published material considerably aids the
development
of
further amendments. So for me, focusing the generation of
consolidations
solely on the need for base text for a revision only deals with half
(or
maybe much less) of the problem.

Regards,
Tony


On 5 October 2011 11:27, Tony Jeffree<tony@jeffree.co.uk> wrote:



Bob -

I agree with much of what you have said here. However, given feedback
received from implementers in 802.1, I don't believe that *just*
publishing
amendments/corrigenda as consolidations serves our readership
either. The
complexity of our standards, and the sometimes intricate way that an
amendment inserts itself into the base, means that for the
implementer to
have a clear picture of what an amendment does to the base document,
he/she
needs to see the deltas; in order to have a clear picture of what the
final
end result is, he/she needs to see the consolidation.

So I believe that publishing amendments/corrigenda as deltas to the
base
standard is necessary IN ADDITION TO publishing the consolidation, and
preferably publishing the consolidation at the same time or soon
after.
We
are currently processing several amendments to Q that are following
this
model; I am hoping that we will be able to publish a consolidation
very
soon
after we are done with the individual amendments. I know that this
creates
an additional load on the editing staff, but on the other hand, it
actually
delivers what the implementers need.

Regards,
Tony



On 27 September 2011 16:28, Grow, Bob<bob.grow@intel.com> wrote:



Paul:

While a reasonable IEEE-SA wide policy, it does little to help the
most
active 802 standards.

1. We have already learned that we have to create consolidations in
some
WGs simply to manage the continuing amendment of the standard. (This
policy
will not change that. If this were a rule rather than a pubs policy,
then
we would be in violation of the statement that a consolidation shall
only be
prepared for a revision. But if it is read only referring to pubs
staff
then there isn't an issue, only no help to our needs.)

2. Four months is a significant problem. In my 802.3 experience it
has
been difficult to find a period of 1 year in which to do a revision.
The 3
year, 3 amendment rule has to be satisfied and some of our standards
will
have a half dozen or more approved amendments/corrigenda and a few
new
amendment projects in process when we try to slot a revision into the
flow
of new projects. While we can typically give a 4 month heads-up for
when we
plan a revision, it will typically be triggered when the last
amendment
to
be included in the revision is approved. (No new amendments are
expected to
complete within a year or so.) At that point, the latest amendment
needs to
be prepared for publication and then consolidated into the revision
draft,
then maintenance changes need to be consolidated into the draft in
preparation to go to WG ballot. Either volunteers have to merge the
approved draft into the staff prepared revision draft (with
publication
changes possibly being missed),!
or we have to find a longer gap into which the revision can be
slotted,
or staff has to be willing to accelerate the consolidation of a
virtually
complete amdnement/corrigenda.

Fortunately, publication staff has been willing to work with us in
recognition of these needs, but the policy certainly doesn't specify
what we
need.

On the other hand, if all amendments and corrigenda were published as
consolidations (editions) pubs staff would not have to handle the
amendment/corrigenda twice (publish and then consolidate), our
balloters
and
users of the standard would not be faced with trying to make sense
of a
standard composed of a big set of documents with stacked changes and
order
specific changes, and we would be making fewer errors by consistently
having
a solid single base standard (not a base standard with separately
published
amendments and corrigenda that are part of the standard).

--Bob

-----Original Message-----
From: ***** IEEE 802 Executive Committee List ***** [mailto:
STDS-802-SEC@ieee.org] On Behalf Of Paul Nikolich
Sent: Tuesday, September 27, 2011 6:48 AM
To: STDS-802-SEC@LISTSERV.IEEE.ORG
Subject: [802SEC] Proposal for Editions

Dear EC members,

Attached is a proposal from the SA regarding a guideline for
consolidations (also known as editions or roll-ups). I've reviewed it
and
it looks like a reasonable process. The one gap I would like
filled is
the
time for the SA to respond to a WG Chair 'request for
consolidation' be
defined (one week, perhaps).

Please review the guideline and provide feedback to the EC
reflector and
Karen McCabe. Thank you.

Regards,

--Paul

----------
This email is sent from the 802 Executive Committee email reflector.
This
list is maintained by Listserv.

----------
This email is sent from the 802 Executive Committee email reflector.
This
list is maintained by Listserv.





----------
This email is sent from the 802 Executive Committee email reflector.
This
list is maintained by Listserv.



----------
This email is sent from the 802 Executive Committee email reflector.
This
list is maintained by Listserv.

----------
This email is sent from the 802 Executive Committee email reflector.
This list is maintained by Listserv.

----------
This email is sent from the 802 Executive Committee email reflector.
This list is maintained by Listserv.

----------
This email is sent from the 802 Executive Committee email reflector.
This list is maintained by Listserv.


----------
This email is sent from the 802 Executive Committee email reflector.
This list is maintained by Listserv.


----------
This email is sent from the 802 Executive Committee email reflector.  This list is maintained by Listserv.