From robert_weir@us.ibm.com Mon May 28 09:49:06 2007
MIME-Version: 1.0
Content-type: multipart/alternative;
 boundary="Boundary_(ID_Q2rE5K+zGcJgzBgcjIgNow)"
Date: Mon, 28 May 2007 12:49:19 -0400
From: robert_weir@us.ibm.com
Subject: Re: Rob Weir's Comments: Page 01, Comment 08-13,
 "Electronc Annexes: Normative vs. Informative"
In-reply-to: <002f01c7a04c$798d71e0$6ca855a0$@com>
To: Rex Jaeschke <rex@RexJaeschke.com>
Cc: "'Dave Welsh'" <dave.welsh@microsoft.com>,
        "'Doug Mahugh'" <Doug.Mahugh@microsoft.com>,
        "'Frank Farance'" <frank@farance.com>,
        "'Jerry Smith'" <jerry.smith@disa.mil>,
        "'Jon Bosak'" <jon.bosak@sun.com>,
        "'Kevin Trainor'" <ktrainor@ligent.net>,
        "'Levine, Leonard F CIV DISA GES-E'" <Leonard.Levine@disa.mil>,
        "'Lynne A. Price'" <lprice@txstruct.com>,
        "'Mason, James David (MXM)'" <masonjd@y12.doe.gov>,
        "'Norm Walsh'" <ndw@nwalsh.com>, patrick@durusau.net,
        "'Richard Barber'" <Richard.Barber@mindjet.com>,
        "'Steve Carton'" <steve.carton@retrievalsystems.com>,
        "'Tim Schoechle'" <timothy@schoechle.org>
X-MIMETrack: Serialize by Router on WTFMAIL03/WTF/M/Lotus(Release
 6.5.6FP1|April 24, 2007) at 05/28/2007 12:49:20 PM,
	Serialize complete at 05/28/2007 12:49:20 PM
X-PMX-Version: 5.2.0.264296

This is a multipart message in MIME format.

--Boundary_(ID_Q2rE5K+zGcJgzBgcjIgNow)
Content-type: text/plain; charset=ISO-8859-1
Content-transfer-encoding: 7BIT

Hi Rex,

The readme file accompanying the ballot says, "A folder 
'C045515e_Electronic_inserts' contains the 6 annexes to the Office Open 
XML document," implying that those zip files are themselves the annexes.  
But in no case do I see the electronic annex itself labeled as informative 
or normative, which is required of ISO Directives, Part 2, section 5.2.6:

"Each annex shall be designated by a heading comprising the word ?Annex? 
followed by a capital letter designating its serial order, beginning with 
?A?, e.g. ?Annex A?. The annex heading shall be followed by the indication 
?(normative)? or ?(informative)?, and by the title, each on a separate 
line."

(We should probably raise another issue on that readme file.  We shouldn't 
be relying on a readme file to tell us what is what in the standard.  That 
should be all covered in Part I.)

Since the annexes contain the "definitive" version of the schema (as 
stated in the text of Part 2 and Part 4), I want to make sure that they 
are properly included.  Conformance to OOXML is defined syntactically, and 
these schemas are the definitive description of this syntax.  So we need 
to ensure that their status with respect to the overall standard is 
clearly stated, that they are properly published, and not inadvertently 
omitted as occurred in the original submission of OOXML to JTC1 NB's in 
January.

Again, no objection to combining similar comments where we have the same 
recommendation.

-Rob

___________________________

Rob Weir
Software Architect
Workplace, Portal and Collaboration Software
IBM Software Group

email: robert_weir@us.ibm.com
phone: 1-978-399-7122
blog: http://www.robweir.com/blog/

"Rex Jaeschke" <rex@RexJaeschke.com> wrote on 05/27/2007 06:47:47 AM:

> Rob states, "There is no explicit indication given as to whether this
> [electronic] annex is informative or normative. See ISO Directives, Part 
2,
> section 5.2.6".
> 
> For all parts, the way in which normative and informative text is
> designated, is described in Part 1, clause 7, "General Description".
> 
> The intent is that everything not explicitly marked otherwise, is 
normative,
> with things explicitly marked as informative, examples, notes, and such, 
are
> informative. Having had a quick look, it's not clear to me that we say 
"that
> everything not explicitly marked otherwise is normative". Perhaps that's 
the
> problem here w.r.t to the XML Schemas.
> 
> The annexes pertaining to RELAX NG are clearly marked as informative. In
> that informative text, we find out that an electronic file exists. Rob, 
are
> you suggesting that we need to say whether than electronic file itself 
is
> informative as well? My understanding is that if one deleted/ignored all
> informative material, what is left is the bit you need to implement the
> standard. That is, the only way you can find out about the electronic 
file
> is to read the informative annex, which suggests that that file is also
> informative. But, if V1 feels strongly enough that this needs 
clarification,
> then I can take a further look at this.
> 
> 
> Separately, if we choose to forward these comments, I suggest we combine
> them into less than six comments, and possibly only one. My concern here 
is
> that at the Ballot Resolution Meeting (BRM), each separate comment 
discussed
> will take a certain amount of time for administrative overhead, plus 
time to
> address the specific comment. As such, I think if six related comments 
can
> be resolved as only one (or possibly two), that would be welcomed by BRM
> attendees.
> 
> 
> Unless I am mistaken, this set of six comments appears twice, one
> immediately following the other. Or am I missing something?
> 
> Rex
> 
> 

--Boundary_(ID_Q2rE5K+zGcJgzBgcjIgNow)
Content-type: text/html; charset=ISO-8859-1
Content-transfer-encoding: 7BIT


<br><font size=2 face="sans-serif">Hi Rex,</font>
<br>
<br><font size=2 face="sans-serif">The readme file accompanying the ballot
says, &quot;A folder 'C045515e_Electronic_inserts' contains the 6 annexes
to the Office Open XML document,&quot; implying that those zip files are
themselves the annexes. &nbsp; &nbsp;But in no case do I see the electronic
annex itself labeled as informative or normative, which is required of
ISO Directives, Part 2, section 5.2.6:</font>
<br>
<br><font size=2 face="Arial">&quot;Each annex shall be designated by a
heading comprising the word &#8220;Annex&#8221; followed by a capital letter designating
its serial order, beginning with &#8220;A&#8221;, e.g. &#8220;Annex A&#8221;. The annex heading
shall be followed by the indication &#8220;(normative)&#8221; or &#8220;(informative)&#8221;,
and by the title, each on a separate line.&quot;</font>
<br>
<br><font size=2 face="Arial">(We should probably raise another issue on
that readme file. &nbsp;We shouldn't be relying on a readme file to tell
us what is what in the standard. &nbsp;That should be all covered in Part
I.)</font>
<br>
<br><font size=2 face="Arial">Since the annexes contain the &quot;definitive&quot;
version of the schema (as stated in the text of Part 2 and Part 4), I want
to make sure that they are properly included. &nbsp;Conformance to OOXML
is defined syntactically, and these schemas are the definitive description
of this syntax. &nbsp;So we need to ensure that their status with respect
to the overall standard is clearly stated, that they are properly published,
and not inadvertently omitted as occurred in the original submission of
OOXML to JTC1 NB's in January.</font>
<br>
<br><font size=2 face="Arial">Again, no objection to combining similar
comments where we have the same recommendation.</font>
<br>
<br><font size=2 face="Arial">-Rob</font>
<br>
<br><font size=2 face="sans-serif">___________________________<br>
<br>
Rob Weir<br>
Software Architect<br>
Workplace, Portal and Collaboration Software<br>
IBM Software Group<br>
<br>
email: robert_weir@us.ibm.com<br>
phone: 1-978-399-7122<br>
blog: http://www.robweir.com/blog/</font>
<br>
<br><tt><font size=2>&quot;Rex Jaeschke&quot; &lt;rex@RexJaeschke.com&gt;
wrote on 05/27/2007 06:47:47 AM:<br>
<br>
&gt; Rob states, &quot;There is no explicit indication given as to whether
this<br>
&gt; [electronic] annex is informative or normative. See ISO Directives,
Part 2,<br>
&gt; section 5.2.6&quot;.<br>
&gt; <br>
&gt; For all parts, the way in which normative and informative text is<br>
&gt; designated, is described in Part 1, clause 7, &quot;General Description&quot;.<br>
&gt; <br>
&gt; The intent is that everything not explicitly marked otherwise, is
normative,<br>
&gt; with things explicitly marked as informative, examples, notes, and
such, are<br>
&gt; informative. Having had a quick look, it's not clear to me that we
say &quot;that<br>
&gt; everything not explicitly marked otherwise is normative&quot;. Perhaps
that's the<br>
&gt; problem here w.r.t to the XML Schemas.<br>
&gt; <br>
&gt; The annexes pertaining to RELAX NG are clearly marked as informative.
In<br>
&gt; that informative text, we find out that an electronic file exists.
Rob, are<br>
&gt; you suggesting that we need to say whether than electronic file itself
is<br>
&gt; informative as well? My understanding is that if one deleted/ignored
all<br>
&gt; informative material, what is left is the bit you need to implement
the<br>
&gt; standard. That is, the only way you can find out about the electronic
file<br>
&gt; is to read the informative annex, which suggests that that file is
also<br>
&gt; informative. But, if V1 feels strongly enough that this needs clarification,<br>
&gt; then I can take a further look at this.<br>
&gt; <br>
&gt; <br>
&gt; Separately, if we choose to forward these comments, I suggest we combine<br>
&gt; them into less than six comments, and possibly only one. My concern
here is<br>
&gt; that at the Ballot Resolution Meeting (BRM), each separate comment
discussed<br>
&gt; will take a certain amount of time for administrative overhead, plus
time to<br>
&gt; address the specific comment. As such, I think if six related comments
can<br>
&gt; be resolved as only one (or possibly two), that would be welcomed
by BRM<br>
&gt; attendees.<br>
&gt; <br>
&gt; <br>
&gt; Unless I am mistaken, this set of six comments appears twice, one<br>
&gt; immediately following the other. Or am I missing something?<br>
&gt; <br>
&gt; Rex<br>
&gt; <br>
&gt; <br>
</font></tt>

--Boundary_(ID_Q2rE5K+zGcJgzBgcjIgNow)--

