a 'mooh' point

clearly an IBM drone

ODF 1.0 Errata 01 released

Yesterday the official Open Document Format for Office Applications (OpenDocument) 1.0 Errata 01 was released for public feedback. The email in part says that

The public review starts today, 7 August 2008, and ends 22 August 2008. This is an open invitation to comment. We strongly encourage feedback from potential users, developers and others, whether OASIS members or not, for the sake of improving the interoperability and quality of OASIS work.
The document is available http://docs.oasis-open.org/office/v1.0/errata/cd02/ for those of you wanting to take a peek or contribute to the work. The document consists of 88 corrections to the text with references to both the OASIS-edition as well as the ISO-edition of ODF 1.0.

I am a bit unsure if the Japaneese ISO-NB defect report as well as the comments from the original ODF ISO ballot are included in this errata - maybe they will not be dealt with before ODF 1.2 hits ISO (possibly) sometime this Fall.



Comments (13) -

You say the errata has references to both the OASIS version as wel as the ISO version.

Does that mean that OASIS and ISO agreed on those errata together ?
(It would be rather strange for OASIS to publish errata on an ISO version unless the errate were agreed upon with ISO.)
And will ISO add those errata to their website as well ?

Does that mean that OASIS and ISO agreed on those errata together ?
No, that is not the case. OASIS has collected errata-data for ODF 1.0 through their regular feedback channels and the results of this work is the errata sheet. After the feedback period ends I will assume the errata-sheet will become an official addition to OASIS ODF 1.0.

Remember that ODF is not maintained in any way in ISO, so nothing happens in ISO unless something is submitted by OASIS (I am not sure if national bodies can request creation of a new "Work item" to do some work on a standard that is not maintained in ISO, but maybe Rob can fill in the blanks here). The Japanese NB submitted a defect report some time ago through ISO to OASIS, but I am not sure anything will happen with it until OASIS submits ODF 1.2 for ISO approval.

... or "rubber stamping", if you will Wink

(It would be rather strange for OASIS to publish errata on an ISO version unless the errate were agreed upon with ISO.)

Yes - but by "referencing" I simply meant that for each errata item, there is the following data:

Section - ODF 1.0 Page - ODF 1.0 Line - ISO/IEC 26300 Page - ISO/IEC 26300 Line - Correction

This is an extremely short timeframe.

In addition, it is not very Euro-friendly, since I imagine a huge amount of people will be on holiday (vacation) for the entire window of opportunity.

The French, certainly, are usually out of action for the entire month of August.

Was this a tactical move, or just bad planning.


Still be referencing the ISO specs they seem to be suggesting that the errata apply to the SIO spec as well which should mean they should submit the errata to ISO really and let them publish their own version fo the errata together with the specification. It is not good if people download the ISO standard and the errata to it are not on available on the ISO site because OASIS chose not to submit them to ISO.


I don't really think it is so short a time-frame. We are talking about two weeks to go through a list of 88 editorial corrections to ODF. That should be time enough. Also, I would imagine that ODF TC is feeling a bit of a time-constraint regarding the submissal of ODF 1.2 to ISO in this Fall - possibly in good time before the SC34 plenary in Korea in late September. If we were to wait until all countries in all time zones were done with their Summer vacations, we would have to wait until October to get some work done.

The proper way for JTC1/SC34 to report defects to the OASIS ODF TC is via the ODF TC's public comment list. The errata document that is now under public review includes fixes to several defects which were submitted by SC34 NB's, primarily Japan, in this manner.

OASIS has a process for creating Approved Errata.  This process must be fulfilled before we can consider any further JTC1 procedure.  Similarly, we must finish ODF 1.2 in OASIS before we can talk about it in JTC1.  The dual references for page numbers and line numbers is intended to simplify any future JTC1 processing of this errata document.

Rick Jelliffe

(This is not a comment on the errata in particular, nor a criticism of the delay,  but relates more to moving forward. Whether in retrospect the ODF TC should have prioritized ODF 1.0/1.1 maintenance over ODF 1.2 development, in order to fulfill its commitments, is a matter for reflection.)

For background research, note that the "OASIS Submission to ISO/IEC JTC1 SC34 Plenary May 2006" from OASIS is at http://www.jtc1sc34.org/repository/0761.htm

In the JTC1 Directives, Clause covers maintenance of PAS. "Its maintenance will be handled either by JTC 1 or by JTC 1 designated maintenance group of the PAS Submitter in accordance with JTC 1 rules."
(See http://www.jtc1sc34.org/repository/0764.htm)

Note also Clause 14.4.2 "Maintenance for a transposed PAS is also negotiated in the Explanatory Report. JTC 1's intention for maintenance is to avoid any divergence between the current JTC 1 revision of a transposed PAS and the current
revision of the original specification published by the PAS submitter. Therefore, the Explanatory Report should contain a description of how the submitting organisation will work cooperatively with JTC 1 on maintenance of the standard. While JTC 1 is responsible for maintenance of the standard, this does not mean that JTC 1 itself must perform the maintenance function. JTC 1 may negotiate with the submitter the option of maintenance handled by the submitter as long as there is provision for participation of JTC 1"

Note that this means that in submitting for PAS by a body that is continuing to undertake maintenance, the submitting body is agreeing to take on the *extra responsibility* of conducting maintenance "in accordance with JTC1 rules": while the location and conduct of the maintenance may be outsourced (so that things are done in an OASIS-ish way rather than an ISO-ish way), the actual maintenance requirements still stand. Notable in the JTC1 maintenance requirements is an emphasis on timely resolution for example (see Clause 15.4.6).  

The OASIS PAS submission is explicit that the intent to be "consistent with the rules of both organizations", and that would certainly still be the expectation of SC34.

So while SC34 certainly should use the comments list mechanism for submitting defect reports to OASIS ODF TC (as per the OASIS submission), the OASIS ODF TC needs to be aware that in engaging in the PAS process it has represented to fulfill the JTC1 requirements with respect to dealing with defect reports. PAS is not a rubber stamp.  

PAS is not a rubber stamp

Actually it is.
Effectivly ISO has shown to have totally zero control over ODF.
Citing directives does not help change that one little bit.

Rick,  perhaps you can point me to 2 or 3 PAS or Fast Track submissions which have been maintained according to your interpretation of the rules?  I can easily point to 2-3 recent ones which show that these clauses are in practice interpreted quite differently by JTC1 and ITTF.


Rick Jelliffe

hAl: Of course JTC1 has no "control", it is all a matter of co-operation and trying to help each other: JTC1 gives the external standard an ISO number and status but at the cost of requiring JTC1 "participation" and with the benefit of roping in a different set of stakeholders (i.e. NBs) whose review gives the consortium standard more legitimacy (e.g. from anti-trust concerns.)

If a PAS or fast-track submitting body reneges on their commitment as a PAS or fast-track submitter, JTC1 has no compulsive capabilities (nor should it IMHO): at most all that would happen is that the ISO standard would be withdrawn in the interests of not having divergent versions. But that would be at the end of the line, and more likely by mutual agreement without acrimony where the submitting body notifies or agreed with the SC that they cannot (or choose not to) perform timely maintenance in response to SC defect reports to keep the versions maintained and in sync.

And if it were just an innocent SNAFU (based on limited resources, unawareness by committees, etc) or the defects were minor or there was some other big picture activity that would satisfy the NBs in the mid rather than the short term, you would expect there would be a good attempt at accommodation: it is not a turf war, just a matter of whether the ongoing maintenance effort is actually meeting the goals and expectations/representations of both organizations.

Rob: I would be very interested in seeing your examples, where a PAS or fast-track group makes representations that they will co-operate and then does not. From your past comments, I am sure you are very concerned that JTC1's reputation does not suffer by consortia making glib promises for marketing advantage without being serious about the cooperative maintenance required for an International Standard.

I suppose "JTC 1 may negotiate with the submitter the option of maintenance handled by the submitter as long as there is provision for participation of JTC 1" is the issue: I am not sure that many SC delegates would consider that "participation of JTC1" was satisfied if defect reports were merely added to a general list to be dealt with at lowest priority some years later, for example.

You might be interested in my reply available at:

The mail thread about the handling of defect reports might also be interestring.  
It is available at:

SC34 has tried to influence the ODF maintenance by sending the liaison report
from SC34 to ODF TC (SC34 N847) and and Japanese defect report (http://www.itscj.ipsj.or.jp/sc34/open/0942.htm).

Bij approving the errata (or in fact any changes to come ) in the OASIS proces prior to submitting them to ISO the only thing ISO can do is rubberstamp them anyways.
OASIS creates the defacto standard version by approving changes and having them  imedialty im-plemented by the main implementers of the format.
ISO refusing to rubberstamp the changes from OASIS would effectivly make the ISO standard useless unless for instance governments would really require the ISO version (which I don't think they will).  

I do not doubt that JTC1 has various ways to interpret the directives regarding maintenance of ISO standards, but I think OASIS should pay close attention to the growing dissatisfaction of the NBs in SC34 to how OASIS has chosen to maintain ODF without any relation to SC34. As you know, the Danish NB voted against PAS-submitter approval for OASIS earlier this year. The reason was exactly this and specifically I share the irritation of other members of NBs that OASIS released an updated version of ODF very shortly after approval of ODF 1.0 in ISO - without sending it to ISO as well. Add to that the "touting of ISO-approval"  while not wanting anything to do with ISO than the bi-annual ISO-stamp.

I don't want to tell OASIS how to run their business, but I think it would be foolish not to listen to the concern of the NBs - they do after all get to vote on ODF in the future ... unless of course, ISO-approval is no longer on the "to-do"-list of OASIS ODF TC.

Comments are closed