Re: [office] The desirability of xml:id stability

[Michael Stahl <mstahl@redhat.com>]

On 26/02/13 16:36, Svante Schubert wrote:
> On 05.02.2013 14:26, Michael Stahl wrote:
>> On 05/02/13 04:38, Dennis E. Hamilton wrote:
>> ...
>>> It seems to me that there is a lot more to clean up before xml:id
>>> stability becomes important.  It is my considered opinion that any
>>> such stability should be left to implementations and only if there is
>>> some sort of outcry for the need for interoperable implementation in
>>> a standard-specified way should anything be done about it at the ODF
>>> TC.  And yes, an ODF 2.0 would likely be the place to address
>>> significant breaking differences with respect to ODF 1.x.
>> as an aside, from an RDF metadata perspective a more serious open
>> question is that when some content element is copied, the copy/paste of
>> the RDF metadata referencing that element via its xml:id is left
>> completely unspecified by ODF, which is unlikely to yield similar
>> results in different implementation.  (the problem is that the RDF
>> graphs have no "natural hierarchy" (they are not trees), and of course
>> by the very nature of the feature ODF implementations cannot assume
>> particular semantics of the RDF properties that happen to be used in
>> some file.)
> The copy/paste of RDF metadata is unspecified as the behavior depends on
> the type of metadata (=> meta-metadata?).
> For instance:
> 
>  1. If I had been the original author of a paragraph and the paragraph
>     is being moved, I will be still the original author and the metadata
>     should/will be kept.

what is "the metadata" exactly in the context of that particular
paragraph?  (it must be somewhere in one of these graphs...)

>  2. If the metadata does instead represent the position of the
>     paragraph, it have to be adapted/removed, if the paragraph itself is
>     being moved (copy/pasted).

i'm not sure if the case 2. is of any practical relevance, why would
anybody add metadata about positions when they could just count the
paragraphs instead?   i'd be more concerned about things like hashes
over content (which are a problem not just for copying) or uniqueness
constraints.

> Either only the module/plugin knowing about the metadata can handle the
> metadata correctly, or my suggestion we require some meta-metadata, as
> "moveable" that makes it understandable for third party software, as the
> office itself.

this is based on the assumption that a "plugin" knowing about the
particular metadata exists for any ODF consuming application, or for
that matter, that the application consuming the file supports "plugins"
at all.

> Ceterum censeo, the xml:id is like a public API for a document, the
> removal/change during a load/save roundtrip is similar to changing the
> anchor IDs of an HTML page, breaking all references into the document.

that's certainly a valid viewpoint.

-- 
Michael Stahl | Software Engineer
Platform Engineering - Desktop Team
Red Hat

Better technology. Faster innovation. Powered by community collaboration.
See how it works at redhat.com