[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: RE: PLCS TC - the next edition ????
|
Consensus in SC4 is that you will need to use the
freshly-minted configuration control tools to generate a second edition AP
module, based on the correct (not necessarily latest) versions of the
modules.
As regards balloting, we have a couple of choices.
New edition requires a NWI/CD ballot to deliver a TS - 3 months, run by
Secretariat. Minor revision calls for an ISO FDIS ballot = 2 months plus
preparation time. There is a precedent for an SC4 resolution to submit
agreed TCs to modules for publication as second editions, although there is a
risk of delay unless the proposal is submitted as an advance resolution for a
meeting, or for a postal ballot.
In short, you will need a vote of some sort to do a new
edition.
Howard From: Rob Bodington [mailto:rob.bodington@eurostep.com] Sent: 21 February 2007 06:31 To: Mason, Howard (UK) Cc: plcs-tog@lists.oasis-open.org Subject: RE: PLCS TC - the next edition ???? Hi Howard I am attempting to understand what is required to produce a TC for AP239 to fix some select omissions.
You indicate in the mail from July last year that there are no TCs on modules permitted. I am not sure how this will work. Does this mean that modular APs cannot have TCs? Does this mean that we need to produce a 2nd edition of at least the AP module? (That is where the express is)
If we do, does that mean that we need to use the latest editions of the modules?
The most straightforward approach for us would be modify the EXPRESS in the AP module as a TC. At some later date, these changes could then be rolled into the relevant modules.
The reason for asking is the TOG needs to make a policy decision here. We are busy writing templates and it would be good to do it once. Regards -------------------------------------------
From: Mason, Howard (UK)
[mailto:howard.mason@baesystems.com]
We agreed not to go through the administrative hassle of managing TCs on modules.
Howard
From: Rob Bodington
[mailto:rob.bodington@eurostep.com] Thanks Howard One thought …. The changes are to the EXPRESS, so are therefore to the modules …. I heard that we were not allowed to do a TC on a module, it had to be a 2nd edition. Is that correct?
Regards -------------------------------------------
-----Original
Message-----
I attach for your information the relevant extract from the SC4 Handbook on Change Management, which highlights the relevant options.
Either Amendment or Revision (to get a new edition) require the normal ballot cycle. TCs can be issued by the Working Groups. Minor revisions, resulting in a complete new document, require an FDIS ballot.
In response to Rob's questions:
This raise several questions about the TC process. 1) How long does it take to produce a TC? HGM> drafting time plus a couple of weeks to approve and process in ISO 2) What do we need to produce? A complete new AP package? HGM> Not for a TC - you might wish to generate a monor revision if htere are many changes 3) Does it require a ballot process? HGM> Not for a TC - just agreement in the WG 4) How many can we produce? I.e should we wait until we have a number of issues addressed then release the TC, or release a TC everytime we address an issue? HGM Wait- there is normally a maximum of two TC before you issue a new edition - sometimes can stretch to three. 5) Can we make major modifications in the TC? E.g. bring the Risk module, address the config management of tasks, or should we just address clear errors like missing select extensions HGM> This sounds like a new edition - we can hardly call it a minor revision.. Just fix errors in a TC.
From: Rob Bodington
[mailto:rob.bodington@eurostep.com]
Hi I was wondering what the process for producing a Technical Corrieganda PLCS would be.
As we are developing the capabilities, we are finding some selects that should have been extended. In a couple of cases, this is preventing us from completing the capabilities, or forcing us to represent something in a less optimal way.
I was thinking that we need a formal process for dealing with modifications to the AP that takes into account the short term requirements to make fixes to the AP for implementations and the requirement to ensure that these fixes are rolled back into the ISO document.
My proposal is that we: 1) Record the issues against AP239 in a single issue log in dexlib – i.e. dexlib/docs/issues/ap239_issues.xml
2) Raise the issue as SEDS so that they are registered against ISO 10303-239, recoding the seds number in the issue log
3) Copy the EXPRESS models for the AP to DEXLIb. Have two files. One a concatenation of all module ARMs, and the other being the long form derived from the ARMs. Fix the EXPRESS according to the issue.
4) Modify the capabilities to use the modified EXPRESS and include a note in the capability referring to the issue and SEDs and to say that it deviates from the standard, but that the issue has been raised against the standard.
5) At some point in time, raise a TC against PLCS.
This raise several questions about the TC process. 1) How long does it take to produce a TC? 2) What do we need to produce? A complete new AP package? 3) Does it require a ballot process? 4) How many can we produce? I.e should we wait until we have a number of issues addressed then release the TC, or release a TC everytime we address an issue? 5) Can we make major modifications in the TC? E.g. bring the Risk module, address the config management of tasks, or should we just address clear errors like missing select extensions
I realise that this opens a potential Pandora’s box of changes that we might be tempted to make, but given where we are with PLCS, I think that we should limit the TC to fixing errors.
What does everyone think?
In the mean time, I will create: 1) The issue log at: dexlib/docs/issues/ap239_issues.xml 2) The EXPRESS
files:
Regards -------------------------------------------
|
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]