openoffice-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From jan i <j...@apache.org>
Subject Re: OOXML
Date Sun, 03 Aug 2014 18:41:18 GMT
On 3 August 2014 19:56, Dennis E. Hamilton <dennis.hamilton@acm.org> wrote:

> Below, Jan asks
>
>   “Does a consumer normally have some sort of conformance sheet
>     (like we have for communication protocols) or is it solely the user
>     that painfully finds the lack of support ?”
>
> I think this is easy to answer.  Where have you found an ODF conformance
> sheet for Apache OpenOffice?  LibreOffice?
>

I have of course not, and always been wondering. My background is
communication protocols and in broad terms ODF can be seen as such, so to
me a statement of conformance is natural. But given your explanation that
many parts are left implementation-dependent (unlike "real" communication
protocols) I understand why.

Im simple words, its a wonder it work, we dont know why, but its a lot
better than nothing :-)

thanks
jan.


> Many choices of what to implement and also deviations of the way features
> are implemented are left implementation-dependent.  In ODF 1.2 there are
> more cases where *implementation-defined* is a requirement.  I am not aware
> how any of those have come up for AOO and LibO and how the
> implementation-based choices are defined, if any.
>
> Here is a serious conformance statement I have found:
> <http://technet.microsoft.com/en-us/library/ff852100(v=office.14).aspx>
>
> Here are some about ODF (scroll down to [MS-OODF], [MS-OODF2], and
> [MS-OODF3],
> <http://msdn.microsoft.com/en-us/library/gg548604.aspx>.
>
> Here’s the on-line version of the one for ODF 1.2 support:
> <http://msdn.microsoft.com/en-us/library/hh695327.aspx>.
>
> It is instructive to expand the sidebar section 2 Standards Support
> Statements and 2.1 Normative Variations.  (I never know what it means to
> say something is not supported.  I believe it is clear that such features
> are not produced, but I have no idea what happens when a not-supported
> provision is encountered in an input document.  All in all, I think this
> is, compared to other implementations, a “glass-half-full” condition.)
>
> In the past there was an on-line database that you could use to review
> compliance with ODF feature by feature, line chapter and verse.  It
> provided for user comments and questions at that level.  It was
> ill-maintained and I can no longer find it.  It looks like the [MS-OODFn]
> documents have taken on that task.  The statements in those documents are
> very much what was to be found on the database.
>
> Cynics will point out that the EUC required Microsoft to describe all
> deviations in its support of ODF.  It is unfortunate that the EUC did not
> consider that such statements would be important from other sources of ODF
> Consumers as well.
>
>
>  -- Dennis E. Hamilton
>     dennis.hamilton@acm.org    +1-206-779-9430
>     https://keybase.io/orcmid  PGP F96E 89FF D456 628A
>     X.509 certs used and requested for signed e-mail
>
>
>
>
> -----Original Message-----
> From: jan i [mailto:jani@apache.org]
> Sent: Sunday, August 3, 2014 00:57
> To: dev; Dennis Hamilton
> Subject: Re: OOXML
>
> On 2 August 2014 22:31, Dennis E. Hamilton <dennis.hamilton@acm.org>
> wrote:
> > [ ... ] There is no strict minimum Conforming OpenDocument
> > Consumer.  A consumer must not object to anything in the document file
> that
> > conforms to the ODF specification, but it is not required to "interpret"
> > all or even any minimum set of features.  There is no producer that I am
> > aware of that produces all features provided for in the ODF
> specification,
> > and most implementations only interpret those features that they are
> > designed to produce (sometimes incorrectly) themselves.  This doesn't
> > matter too much if you use implementations with a common genealogy, but
> > across independent implementations not having any common code base there
> > tend to be unexpected surprises.  There are also many places where a
> > provision of ODF is not rigorously defined and implementation-dependent
> > variation is the result, whether explicitly called out (e.g., for macros
> > and scripts) or not (e.g., for supported image formats).
> >
>
> Does a consumer normally have some sort of conformance sheet (like we have
> for communication protocols) or is it solely the user that painfully finds
> the lack of support ?
>
>
> [ ... ]
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@openoffice.apache.org
> For additional commands, e-mail: dev-help@openoffice.apache.org
>
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message