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

Re: [802SEC] Provision of obsolete draft standards to members



Jon-

802.3 has always kept track of things "by project" rather than "by standard".
That makes it a little easier to keep track of things when doing this sort of thing.
Thus each Revision or Corrigenda can be tracked separately in the same manner
as a new standard or an amendment.
(more below)
On 510//12 11:30 AM, Jon Rosdahl wrote:
Just to be sure we are all talking to the same points.
 
When a WG has a draft of a document - D1.0 it sends it out for ballot.
It then takes feedback/comments and makes D2.0
this continues until we make it to Sponsor ballot ...say D5.0 (for sake of this argument).
The current convention in 802.3 is usually
    Draft 1.n:    Task Force ballot and recircs
    Draft 2.n:    WG Ballot and recircs
    Draft 3.n:    Sponsor Ballot and recircs
During the Sponsor ballot process there are several recircs
and lets say D9.0 makes it to RevCom.
 
This produces a Published-year document. (say Widget-2000).
 
When Widget-2000 got revised, it had several steps.
    Widget-2000-d1.0 which was balloted in the WG
    and several drafts created
    Then it was sent to Sponsor ballot as Widget-D7.0
  It gets published and is called Widget 2012.

802.3 gives this a separate (now two) letter project designation (that the IEEE doesn't use for a revision project)

 
Now the question is which documents should be available to the public
Which documents should be available to the members
and for how long.
 
As I understand from the thread:
     The CD contains all the published version history.

Correct in theory.  I believe that some of the older stuff is missing.  This more likely to be true when the published standard was not available in PDF.

      Each WG has saved the all the interim drafts, but they are generally available only to the chair.
     Many of the WG have some subset of the full set of documents available to the WG voters.
     Some Individuals have an archive that is better than others.
 
The "superseded" moniker I believes goes with those versions of the specification that was made public and are then replaced (superseded) with a newer version.

My rule has always been that I rolled the version number when more than one person has had access to the draft.  Numbers are cheap.  I haven't always made each version available in the archive if it was just a working version that was never distributed beyond an editor or few.

 
Now you can see why Adrian may have called some of the versions obsolete, as once the recirc has happened, and sent to RevCom, they are obsolete in that there is a published (final) version of the standard.

My point to Adrian is that any version that is circulated (to any balloting or approval group) is not "obsolete", it is "superseded".
In particular, the draft that was sent to RevCom and approved is never "obsolete" nor is it "superseded" by the published version.
The approval draft as voted rather than the published edition is definitive.

 
So now which documents are we thinking should be saved?
All on the CD? or just the published ones?

I think that just having a complete archive of published standards is good enough for almost all practical purposes

Just thinking out loud.
Jon

  

Best regards,
    Geoff

     
 
 
----- Original Message -----
Sent: Friday, October 05, 2012 11:04 AM
Subject: Re: [802SEC] Provision of obsolete draft standards to members

Adrian-

I feel that there are a couple of minor points to tidy up here.

I feel that "obsolete" isn't the best terminology to use in terms of precision

1)    Re, your statement:

> Obsolete (my usage) relates to drafts obsoleted by publication of the amendment/standard/revision they become.


Drafts aren't actually obsoleted by publication.  When push comes to shove, the last balloted draft as approved by the SASB is the definitive document. Therefore the last balloted draft needs to be kept available.  It's important to recognize that changes do creep in during the publication process.  We try very hard to make sure that the opportunity for changes is minimized and any changes are not of substance but the process is not perfect and some groups are better at this than others.  The differences between the SASB approved document and the publication are not balloted nor reviewed by the balloting group.  When a difference of substance is found, an errata (which is not a balloted document) is issued.

2)    Re, your statement:

> Obsolete does not refer to published amendments/standards/revisions that are obsoleted by a later revision.  Such

> documents are available on the members’ CD in the archive directory.  They are also (in the 802.11 case) available

> for purchase on Techstreet. 

The appropriate term for these is "superseded" rather than "obsoleted"

Thanks for starting a worthwhile discussion.

Best regards,

        Geoff



On 410//12 11:53 PM, Stephens, Adrian P wrote:

Many thanks all for responses on this.

Here’s what I learned:

Nobody objected to the principle of making obsolete drafts available to members.

That was the primary goal of my email,  and we’ll be using this (It’s Bruce’s call,  but I can see where this will go)

to make such drafts available to members.

Obsolete (my usage) relates to drafts obsoleted by publication of the amendment/standard/revision they become.

Obsolete does not refer to published amendments/standards/revisions that are obsoleted by a later revision.  Such

documents are available on the members’ CD in the archive directory.  They are also (in the 802.11 case) available

for purchase on Techstreet.

Access to obsoleted drafts should be “member access” – i.e., whatever rules the group uses to protect/share its current drafts.

(I had a comment on “everlasting access”.   This is defined to be a period p of  years where:

            p = Max(tsyat_i – n_i)

where

            tsyat = estimated remaining lifetime of member i (unsigned integer, default value (3x20)+10)

            n_i = current age of member i

            the maximum is performed over all members currently wanting or in the future likely to want to access old drafts)

Best Regards,

Adrian P STEPHENS

Tel: +44 1954 204 609 (office)
Tel: +44 7920 084 900 (mobile)
Skype: adrian_stephens

----------------------------------------------
Intel Corporation (UK) Limited
Registered No. 1134945 (England)
Registered Office: Pipers Way, Swindon SN3 1RJ
VAT No: 860 2173 47

From: Stephens, Adrian P
Sent: Thursday, October 04, 2012 4:54 PM
To: STDS-802-SEC@LISTSERV.IEEE.ORG
Subject: Provision of obsolete draft standards to members

Dear 802 SEC,

A question came up in 802.11 at the last session regarding access to obsolete standards of drafts.

(e.g. drafts of amendments that were subsequently published,  and then subsequently obsoleted by a revision).

The reasons to access such old documents might be:

<!--[if !supportLists]-->1.       <!--[endif]-->To avoid paying a fee for the published document (yes,  I know we have Get802 and currently a $5 fee).

<!--[if !supportLists]-->2.       <!--[endif]-->For historical interest,  such as documenting the development of a standard for a book or paper

<!--[if !supportLists]-->3.       <!--[endif]-->For legal reasons

802.11 removes access to such drafts as soon as they are obsoleted by publication of an amendment or revision.

There are perhaps three reasons to do this:

<!--[if !supportLists]-->1.       <!--[endif]-->The terms under which access to drafts is provided is to develop the amendment/revision.

<!--[if !supportLists]-->2.       <!--[endif]-->Looking at old drafts can be injurious to your interoperability

<!--[if !supportLists]-->3.       <!--[endif]-->May dissuade members from purchasing/acquiring/using the approved document

I have researched the IEEE-SA rules and can find no statement that says that access to drafts has to be

removed at any time.

Could I have comments from the SEC members on the following:

<!--[if !supportLists]-->1.       <!--[endif]-->Do you believe there are any IEEE-SA rules that prevent providing everlasting access to drafts?

<!--[if !supportLists]-->2.       <!--[endif]-->Do you see any practical pros and cons to this?

<!--[if !supportLists]-->3.       <!--[endif]-->Do you do this in your group?

Best Regards,

Adrian P STEPHENS

Tel: +44 1954 204 609 (office)
Tel: +44 7920 084 900 (mobile)
Skype: adrian_stephens

----------------------------------------------
Intel Corporation (UK) Limited
Registered No. 1134945 (England)
Registered Office: Pipers Way, Swindon SN3 1RJ
VAT No: 860 2173 47

---------- 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.