<office@lists.oasis-open.org>
wrote on 11/26/2014
02:15:26 PM:
> From: Chris Rae <Chris.Rae@microsoft.com>
> To: ODF TC List
<office@lists.oasis-open.org>
> Date: 11/26/2014 02:17 PM
> Subject: [office] Proposed wording of
response
to JP2 (change tracking)
> Sent by:
<office@lists.oasis-open.org>
>
> I thought I’d get the ball rolling here – how about the
following
> for a response:
>
> “Change Tracking is a complex feature and
the
TC acknowledges that
> there are improvements that can be made – the Advanced
Document
> Collaboration subcommittee is working on exactly that for
ODF 1.3.
> However, we disagree with the suggestion of adding a
blanket
> ‘implementation-defined’ caveat to the whole of section
5.5. The
TC
> will investigate specific interoperability issues within
the section
> if further information can be provided.”
>
> Chris
We could also argue that putting a blanket
statement
of "implementation-defined" would not be accurate. Although
behavior is not as tightly bound as we'd like it in some
places, it is
not entirely unbound either.
We also need to watch out for the distinction
between
implementation-defined and implementation-dependent. If we
say implementation-defined
we've put an additional requirement on implementors, to
document how that
is implemented. If we do that (and I'm not sure we can in
errata)
then we'd need to make it clear what exactly an implementor
must document.
I guess it all comes down to this: a vague
comment
does not improve the specification.
In any case, rather than saying "However, we
disagree with the suggestion of adding a blanket", maybe
something
like "However, we believe it would not be entirely accurate or
useful
to implementors of the standard to add...". Or something
like that.
-Rob
Doing a merge of your two proposals would result into the following,
which would be fine for me:
"Change Tracking is a complex feature and the TC acknowledges that
there are improvements that can be made – the Advanced Document
Collaboration subcommittee is working on exactly that for ODF 1.3.
However, we believe it would not be entirely accurate or useful to
implementors of the standard to add a blanket
‘implementation-dependent’ caveat to the whole of section 5.5. The
TC will investigate specific interoperability issues within the
section if further information can be provided."
Best regards,
Svante
|