W3C home > Mailing lists > Public > public-webcgm-wg@w3.org > August 2007

results of 1.0 errata scan

From: Lofton Henderson <lofton@rockynet.com>
Date: Tue, 07 Aug 2007 17:29:30 -0600
Message-Id: <5.1.0.14.2.20070807101245.02f32cb0@rockynet.com>
To: WebCGM WG <public-webcgm-wg@w3.org>

WebCGM WG,

Current working draft of the 1.0 errata page [1],
[1] 
http://www.w3.org/Graphics/WebCGM/WG/2007/errata-10/webcgm10-errata-20070621.html 
.

In addition to the five placeholder items there, I have found these 
additional references to needed 1.0 errata.

I'm just identifying the potential ones now, not fully specifying 
them.  Note also that #4 actually turned out to be a 2.0 item, a close look 
showing that 1.0 is unaffected.  (So there are four potential new 1.0 items.)

These would pretty much complete the scan and 1.0 errata list, unless 
someone can find others squirreled away obscurely in our archives of the 
past 5 years.  Any more contributions out there?

=====begin item #1 =====

>Date: Tue, 27 Aug 2002 12:18:34 -0600
>From: Lofton Henderson <lofton@rockynet.com>
>Subject: [cgmopen-members] WebCGM editorial error
>
>CGM Open members,
>I have found a big editorial bug in WebCGM 2nd Release. Look at 3.1.2.4, 
>and look at the bullet list in 3.2.1.1 (see reference [1]). They directly 
>contradict each other now. Recall we had a big issue about this, and even 
>had a special teleconference. As I recall, after considerable disagreement 
>and discussion, we finally settled for what is described in 3.1.2.4 (for 
>object behaviors other than view_context).
>The contradictory specifications (i.e., the bullet list) in 3.1.2.4 should 
>have been deleted, but by editorial mistake, they were not (my fault).

That describes the problem and the solution (replace the bullet list in 
3.2.1.1 with a reference to 3.1.2.4).  [This does not affect 2.0.]

=====end item =====
=====begin item #2 =====

>Date: Fri, 06 Dec 2002 07:28:33 -0700
>To: Lynne Rosenthal <lynne.rosenthal@nist.gov>,
>"Cruikshank, David W" <david.w.cruikshank@boeing.com>
>From: Lofton Henderson <lofton@rockynet.com>
>Subject: Re: Figure element question
>
>At 09:04 AM 12/6/2002 -0500, Lynne Rosenthal wrote:
>Dave
>That is a good question. I think what we meant was that the limit was on 
>graphical primitives.
>That was clearly the intent. There is no logical reason to include 
>attribute elements. The purpose of the limit is to make predictable the 
>resource requirements of the viewer, and attribute elements have 
>negligible bearing on that.
>However, the way it is written, it would apply to any elements - and since 
>attributes are an element - this means 128 elements, graphical, 
>attributes, or any others.
>Technically, yes.
>However, as a practical matter I would treat it as an ambiguity (open to 
>interpretation), therefore subject to an Erratum. If we had an active 
>errata process, we could correct it on the RHS of the PPF in CGM:1999.
>The practical solution is for the working profiles (ATA, WebCGM, etc) to 
>correct themselves. I.e., to clarify in the left-hand column of the PPF 
>that the limit is "128 graphical primitive elements".

The problem being that the PPF does not explicitly specify "graphical 
*primitive* elements" in the content-count limit of closed figure, the 
solution is to correct the PPF.  [This probably affects 2.0 as well.]

=====end item =====
=====begin item #3 =====

There is apparently a 1.0 erratum in this thread, these two samples of 
which should give plenty of details:

http://lists.oasis-open.org/archives/cgmo-webcgm/200407/msg00007.html
http://lists.oasis-open.org/archives/cgmo-webcgm/200407/msg00009.html

It is NOT about the WebCGM 'name' APS attribute.  It is about the old 1.0 
text in 3.1.4,
http://www.w3.org/TR/2001/REC-WebCGM-20011217/REC-03-CGM-IC.html#webcgm_3_1_4" 
.
The first paragraph was totally confused and wrong.  The options look 
like:  either delete first paragraph (since 2.0 DOM apparently solves the 
issue); or else, work up something that does the job in 1.0 but is correct.

a couple excerpts of above-linked email messages...

>Date: Mon, 05 Jul 2004 18:02:35 -0600
>To: <dieter@itedo.com>
>From: Lofton Henderson <lofton@rockynet.com>
>Subject: RE: [cgmo-webcgm] DOM question
>
>At 08:13 PM 7/5/2004 +0200, Dieter Weidenbrueck wrote:
>> >  BTW, the wording of the spec should be changed; the 'name'
>> >  attribute in xhtml has been deprecated... it has been replace with
>> >  XML:ID.  Therefore, to address a specific 'object', a user should be
>> >  referencing the unique 'id'.
>>one more defect for Lofton
>
>Whew!  This is a mess.  I think we misread and/or misunderstood the OBJECT 
>and PARAMETER element definitions.  The NAME attribute is indeed allowed 
>on OBJECT in HTML 4.01, but it refers specifically to "--submit as part of 
>form--".
>
>[...snip lotsa' detail...]
>>One question: Should we add some words about grandfathering of the
>>"name" attribute of the object tag to ensure upward compatibility?
>
>IMO, the 'name' specification in WebCGM 1.0 is an Erratum.  As I noted 
>above, that referencing text is wrong in several ways.
>
>That said, I'd like to know how vendors would like to treat it, [...]

another mail excerpt...

>From: "Dieter Weidenbrueck" <dieter@itedo.com>
>To: "Lofton Henderson" <lofton@rockynet.com>
>Date: Tue, 6 Jul 2004 08:32:45 +0200
>Subject: RE: [cgmo-webcgm] DOM question
>
> >That said, I'd like to know how vendors would like to treat it, given that
> >the mistake has been sitting there for a few years.  Approaches
> >could range
> >from fix it (Erratum fix means "retroactive" in worst case), to
> >"grandfather", to somewhere in between, like a non-normative note
> >explaining the former glitch in the specification and the fact
> >that there's
> >probably some legacy content floating around.
>
>I vote for an erratum plus a non-normative note.
>
>Reason:
>- if we fix it "retroactively" it provides for the correct way to
>   use WebCGM 1.0 in the future. If we wouldn't do it this way, it
>   would still be "legal" as per WebCGM REC to do the wrong thing.
>- There should be a non-normative note explaining the situation,
>   and suggesting that vendors continue to support the "name" if
>   they have done so up to now.


=====end item =====
=====begin item #4 =====

This is actually a trivial potential 2.0 erratum (or not), 1.0 is unaffected:
http://lists.oasis-open.org/archives/cgmo-webcgm/200510/msg00070.html .

>Date: Mon, 10 Oct 2005 11:01:14 -0600
>To: "Cruikshank, David W" <david.w.cruikshank@boeing.com>,
>         "CGM Open WebCGM TC" <cgmo-webcgm@lists.oasis-open.org>
>From: Lofton Henderson <lofton@rockynet.com>
>Subject: Re: [cgmo-webcgm] Review of WebCGM 2.0 Profile
>
>At 12:56 AM 10/5/2005 -0700, Cruikshank, David W wrote:
>[...]
>
>>T22.2  Application Data  it's interesting that the model profile says 
>>"The use of this element shall not be restricted." and we prohibit it?
>
>Yes, interesting.  That statement was not in the MP of CGM:1992 Amd.2 (the 
>original normative PPF, to which WebCGM 1.0 was written).  It somehow got 
>added when the PPF (and rules for profiles) were rolled into the CGM:1999 
>republication.  I don't know what it means.  And I can't find an erratum 
>which indicates that the change should have been made (to CGM:1999 PPF).
>
>I suggest we ignore it.  Alternately, we could put a note indicating the 
>above, or that the specification in the CGM:1999 PPF is suspected to be an 
>error.

It actually looks like an error in the CGM:1999 PPF, but (if we decide not 
to ignore it), we could put a note in the 2.0 PPF, via an erratum, that 
indicates the CGM:1999 PPF for the MP column is suspected to be in error.

=====end item =====
=====begin item #5 =====

In April '06 the TC had a discussion about drawing model, which resulted 
for 2.0 in some simple informative clarifications and quantitative 
explanations.

In WebCGM 2.0 it became a new subsection (2.2.2) in a heavily revised 
section 2.2:
http://www.w3.org/TR/2007/REC-webcgm20-20070130/WebCGM20-Concepts.html#webcgm_2_2_2

This section did not exist at all in section 2.2 of 1.0:
http://www.w3.org/TR/2001/REC-WebCGM-20011217/REC-02-CGM-Concepts.html#webcgm_2_2

Our april '06 dialog indicated that we were thinking to add it 
retroactively to 1.0 as an erratum.

=====end item =====

Regards,
-Lofton

=====begin item #N =====
...template...
=====end item =====
Received on Tuesday, 7 August 2007 23:29:58 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:19:10 GMT