RE: [oic] Confused about the purpose of the Interop Profile

[robert_weir@us.ibm.com]

I see this as being in the spirit of PDF/A.  namely a profile of the full 
standard that defines additional constraints that make it more suitable 
for a particular purpose.  The work on PDF/A was largely around 
prohibiting features that were implementation-dependent and the source of 
portability problems.  I think the desire is to accomplish something 
similar with ODF.

As you point out, there is more than one way to do this.  It could have 
been done in the form of a "guidelines" document.  It could have been done 
in the form of a profile standard.  It could have been done in the ODF TC, 
in the form of  a proposal for the next revision.  These approaches differ 
in some subtle ways.  The last option (ODF TC proposal) would be out of 
scope for this TC, but I don't see a problem with the other approaches.

Note that not every implementation-dependent feature in ODF 1.1 is 
necessarily a problem that should be "fixed" by eliminating it from the 
ODF standard itself.  For example, ODF documents may have scripts or 
macros in them.  These are useful for many people.  We don't want to see 
them prohibited from ODF.  But they are implementation-dependent, and 
undoubtedly can be the source of interoperability problems.  So for those 
uses who prioritize portability above other concerns, say because they are 
looking for long-term archiving of documents, then a profile that 
prohibits embedded scripts may be appropriate. PDF/A does something 
similar when they prohibit JavaScript in PDF/A documents.

Also, nothing that this TC does can add or change a conformance clause or 
anything in the ODF standard itself.  We're two different TCs.  The most 
we can do is define a profile standard, which would define a conformance 
clause, and an implementor could claim conformance to that profile. This 
might be done in addition to the ODF 1.1 or ODF 1.2 conformance clause.

I agree we want to avoid confusion.  With PDF, PDF/A, PDF/X, PDF/E, 
PDF/UA. PDF/H and PDF/VT they may have gone a bit too far. 

As for ODF, the OASIS versus ISO versions are intended to be only a 
difference in cover sheets.  And errata/corrigenda do not change 
conformance.  They primarily fix editorial errors.  So we really only have 
ODF 1.0, ODF 1.1 and soon ODF 1.2.  Regardless of whether you do the 
interop points as a guidelines document or as a profile standard, you 
still have the problem where you need to evolve it in parallel with the 
ODF standard.  So if we have an interop "thing" for ODF 1.1, then I'd 
expect our TC would revise that document for ODF 1.2.

Regards,

-Rob

Cherie Ekholm <cheriee@exchange.microsoft.com> wrote on 03/19/2010 
12:53:51 PM:


> 
> RE: [oic] Confused about the purpose of the Interop Profile
> 
> I understand wanting to clarify the 1.1 standard - it often happens 
> that after a standard is released and implementers start using it 
> that areas that need clarification are exposed. I'd agree that this 
> can be a detriment to interoperability.
> 
> I also understand the vertical and horizontal profiles as described 
> in the charter. Horizontal and vertical are marketing terms (or 
> began as marketing terms) that are widely recognized in the 
> industry. The examples used in the charter are fairly good 
> descriptions of the types of profiles that would fall under each domain.
> 
> However, the proposed interoperability profile currently under 
> public review does not meet the criteria for either a horizontal or 
> vertical profile as described in the charter. Nor does this profile 
> fit the industry understanding of those terms. The items covered in 
> this profile do, however, seem to fall under Item 6 in the OIC charter:
> 
> 6. To provide feedback, where necessary, to the OASIS Open Document 
> Format for Office Applications (OpenDocument) TC on changes to ODF 
> that might improve interoperability
> 
> It's understandable to be concerned that the next version of ODF 
> will not be out for a while yet and want to create some best 
> practices for implementers to help create a better environment for 
> interoperability with the current standard. It's common practice for
> international standards to have corrigenda, technical reports, 
> amendments, and even implementer notes or best practices documents 
> written between versions of the standards for exactly this purpose, 
> as well as to correct any incorrect items in the standard. 
> 
> One of the concerns I have here is that with the publication of this
> profile, we will have ODF 1.0, IS26300, ODF 1.1, ODF 1.1 Interop 
> Profile, plus applicable corrigenda, and the submission of ODF 1.1 
> that just went to JTC1. There have been discussions in the ODF TC 
> that it may already be too confusing for regulators to figure out 
> which single standard or combination of standard plus corrigenda to 
> require. And now OIC is adding a profile to the mix that addresses 
> one of those existing standards rather than a vertical or horizontal
> that the committee's charter calls for. 
> 
> If the idea here is to put out a best practices guide or best 
> practices profile as Bart says, why not call this just that?
> 
> Cherie
> 
> 
> -----Original Message-----
> From: Hanssens Bart [mailto:Bart.Hanssens@fedict.be] 
> Sent: Thursday, March 18, 2010 5:17 PM
> To: Cherie Ekholm; oic@lists.oasis-open.org
> Subject: RE: [oic] Confused about the purpose of the Interop Profile
> 
> Hello Cherie
> 
> 
> well, in the charter we have these two activities
> 
> 4) To define profiles of ODF which will increase interoperability 
> among implementations in the same vertical domain, for example, ODF/
> A for archiving;
> 5) To define profiles of ODF which will increase interoperability 
> among implementations in the same horizontal domain, for example ODF
> Mobile for pervasive devices, or ODF Web for browser-based editors;
> 
> Now, while reading the spec itself, and while testing different 
> office suites during the ODF plugfests etc, it became clear
> that sometimes the spec is not clear / not complete or a bit too 
> relaxed, and implementers implemented some features
> differently.
> 
> This can be solved in the next version of ODF, but this could take a
> while, so the idea is to create a profile for typical
> office suite implementations (is that horizontal or vertical or both
> ? not sure, doesn't matter IMHO), by creating new
> conformance clauses on top of the existing clauses of ODF 1.1 
> 
> The profile won't become an OASIS standard, and of course it does 
> not change the ODF 1.1 spec itself, but at least
> some best practices are documented that way (Although perhaps some 
> parts might make it into an ODF 1.1 errata,
> some of them might end up in ODF 1.2..)
> 
> 
> Best regards
> 
> Bart
> ________________________________________
> From: cheriee@exchange.microsoft.com [cheriee@exchange.microsoft.com]
> Sent: Thursday, March 18, 2010 11:04 PM
> To: oic@lists.oasis-open.org
> Subject: [oic] Confused about the purpose of the Interop Profile
> 
> This is probably going to come across as one of those stupid newbie 
> questions, so I'll apologize in advance if that's the case.
> 
> I've been reading through the Interop Profile document and I'm not 
> sure I understand what I'm reading. Is it supposed to be a new 
> conformance class for ODF 1.1? I read back through a lot of the 
> email from the summer and fall when this was being worked on, prior 
> to my joining the committee, and read through the charter again and 
> have only become less certain about what the purpose of the document is.
> 
> Can someone send me some context so I can provide a better review?
> 
> Thank you.
> 
> Cherie Ekholm
> 
> ---------------------------------------------------------------------
> 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 
>