Re: [office] "In-passing errors" and my sense of errata

[robert_weir@us.ibm.com]

There are four kinds of defect comments, based on how we treat them:

1) Comments received during public review periods -- for these, we are 
obligated under OASIS rules to process the comments, and note our 
resolutions on our mailing list, before the draft can be approved.  Where 
the comment is reporting a defect in the standard we will typically make 
the change in the draft and send it for further review.   No need for 
errata, no need for rationale.  We just change the text directly, since it 
is still a draft.  Since the text is still a draft, we are free, if we 
wish, to fix other defects uncovered while fixing the reported ones.

2) Comments received the public on an approved ODF standard -- for these 
we have no obligations under OASIS rules.  However, the TC has agreed to 
process and log all public comments in a spreadsheet, and to process them. 
 Our options for any defect are:  a) Include a fix in one or more errata 
documents, b) Fix only in ODF 1.2, c) Refer to the ODF-Next Requirements 
SC, or d) Do nothing.  Since ODF 1.0 has been widely translated I would 
like to avoid making widespread changes to the text.  However, technical 
errors can and should be addressed via errata.

3) Comments received via TC members.  Similar to #2.   If it is an 
editorial error, then I'd recommend fixing it in ODF 1.2 only.  If it is a 
technical defect, then raise it to the list as an item for an errata 
document.  We probably need some mechanism for tracking these, to ensure 
they don't get lost.  We could use the same mechanism as #2 to track them? 
 Or maybe we can have a wiki page for member-submitted defect reports? 
Whatever we do, we need a tight loop between the reported of the defect 
and Patrick.

4) Formal defect reports from JTC1/SC34.  This is really the turd on our 
plate right now.  It is striding two different processes and really 
doesn't follow the rules correctly for either one of them.  The best thing 
we can do is to get it off our plate ASAP.  This means an ODF 1.0 approved 
errata document that addresses the SC34 defect report, and only items in 
that defect report.  (The JTC1 process assumes that their corrigenda 
documents contain only fixes to defects that JTC1 reported).  We'll have 
an opportunity to issue additional ODF 1.0 errata documents to address 
additional comments in the future, so nothing is lost.  However, we 
committed to responding to the SC34 defect report, and we're already quite 
late on that.  So I'd ask for the TC to help us get through this process. 

I'm hoping that #4 is a one-time thing, and we can treat any future 
occurrences the same as #2.  As it is, I have some discomfort in that we 
are addressing the JTC1 defect report, out of turn, even though they are 
almost entirely trivial editorial comments, while we have stepped over (in 
chronological order) other public comments, some of which indicate more 
credible defects and ambiguities. This is not fair to the submittors, is 
not good PR for the TC and leads to a poorer errata report than if we had 
triaged all of the comments.  But since we committed to delivering #4, and 
Murata-san is expecting it and indeed widely complaining about it,  let's 
finish the job and get it off our plate.  But we should commit ourselves 
to taking a broader view of errata for the next ODF 1.0 errata document. 

Regards,

-Rob

"Dennis E. Hamilton" <dennis.hamilton@acm.org> wrote on 09/15/2008 
06:16:26 PM:

> [image removed] 
> 
> [office] "In-passing errors" and my sense of errata
> 
> Dennis E. Hamilton 
> 
> to:
> 
> ODF TC List
> 
> 09/15/2008 06:21 PM
> 
> Please respond to dennis.hamilton
> 
> Not wanting to go into repetitive appeals and re-argument, I replied to
> Patrick off-list.
> 
> I am still not interested in that. 
> 
> I am forwarding this to the list because it also provides a declaration 
on
> what you can count on me in terms of the principles that I will follow.
> That part is about how I conduct myself.  It is not an invitation to
> reconsider anything on the part of the TC.
> 
>  - Dennis
> 
> Dennis E. Hamilton
> ------------------
> NuovoDoc: Design for Document System Interoperability 
> mailto:Dennis.Hamilton@acm.org | gsm:+1-206.779.9430 
> http://NuovoDoc.com http://ODMA.info/dev/ http://nfoWorks.org 
> 
> 
> -----Original Message-----
> From: Dennis E. Hamilton [mailto:dennis.hamilton@acm.org] 
> Sent: Monday, September 15, 2008 14:25
> To: 'Patrick Durusau'
> Subject: RE: [office] Re: #6.8 - Errata Review - "In-passing errors"
> 
> I understand about the fear that somehow we would be off scouring the 
whole
> document. 
> 
> I don't suppose that and you'll notice that I have been very judicious 
in
> offering where I think it is appropriate to add another item related to 
an
> already-identified defect.
> 
> Although I hope you appreciate how weird this looks from the outside, I 
am
> certainly not going to get upset about how this gets worked out.
> 
> Meanwhile, when my explorations of the documents bring up oddities, I 
will
> capture them and probably mention them too.  I hate thinking that these 
be
> simply left for others to stumble on when also attempting to master the
> specifications.  Having done the work, I am inclined to want to save 
others
> from needing to do it too.  Also, reporting them would also be useful 
for
> informing the ODF 1.2 work.
> 
> I look forward to your document.  I will look it over by tomorrow at the
> latest, in case there is anything to resolve before next week's call. My
> hope is that enough people look and review the work to tidy up the 
errata so
> that we can approve another review on Monday.
> 
>  - Dennis
> 
> PS: From a comment that Rob Weir made, my sense is that any re-review of 
the
> errata must be with a markup of the one last reviewed.  Is that your
> understanding?
> 
> -----Original Message-----
> From: Patrick Durusau [mailto:patrick@durusau.net] 
> 
http://www.iso.org/iso/search.htm?qt=29500&published=on&active_tab=standards

> Sent: Monday, September 15, 2008 13:53
> To: dennis.hamilton@acm.org
> Cc: ODF office
> Subject: Re: [office] Re: #6.8 - Errata Review - "In-passing errors"
> 
> Dennis,
> 
> In theory I don't disagree with your position on errors in the text and 
> don't want you to have the impression that I do.
> 
> As a practical matter, however, errata have to be bounded by those 
> reported by someone, else how would we know where to stop? One position 
> would be that we should create "errata" that when applied would result 
> in ODF 1.2. That is certainly possible in theory but I would not want to 

> volunteer for the task.
> 
> The purpose of this errata is to fulfill our maintenance obligations 
> that OASIS undertook when it submitted ODF 1.0 2nd edition to ISO. There 

> are a variety of opinions about the process we have or have not followed 

> and I don't want to get diverted into that discussion here.
> 
> Suffice it to say that this set of errata was created to answer the 
> Japanese "defect" report which I referenced earlier today. And yes, 
> there were a number of errors in my latest draft that you caught and I 
> am very grateful for the hard work.
> 
> Yes, the ODF TC needs a vigorous discussion of errata processes, how 
> long we will even answer questions about prior versions, etc.
> 
> However, my impression is that while ODF is of a great deal of interest 
> to many parties, the TC doesn't have unlimited resources and so I 
> suspect that at some point we are going to decline to maintain earlier 
> versions of the ODF standard. That is just a guess on my part but I 
> think it is probably accurate.
> 
> Hope you are having a great day!
> 
> Patrick
> 
> PS: I have been cross-checking against your checks, plus against the 
> Japanese errata and both the ODF 1.0 and ISO 26300 text so I am hopeful 
> that the version I file tonight will be fairly clean.
> 
> Dennis E. Hamilton wrote:
> http://lists.oasis-open.org/archives/office/200809/msg00036.html
> > Patrick,
> >
> > I was not consulting the defect report.  I just happened to see, while
> > trying to match up the lines, that there was this other line that had
> > exactly the same problem. (I ran into that line by mistake while 
checking
> > the existing errata items and it took extra effort to be sure which 
were
> the
> > correct places to apply the existing errata.) 
> >
> > So I am pointing out this other place in the same table.  In one sense 
it
> is
> > the same defect, and I am not so literal about scope (since as a 
reviewer
> of
> > the errata as it is disseminated, I have no idea about that - the 
scope
> > restriction is not in evidence).
> [ ... ]
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe from this mail list, you must leave the OASIS TC that
> generates this mail.  Follow this link to all your TCs in OASIS at:
> https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php 
>