Re: [oic] Restrictions for meta:editing-duration

[Bernd Eilers <Bernd.Eilers@Sun.COM>]

Hi Bart,

Well your c2 MAY Solution even better solves the possible dilemma while 
still allowing applications to internally use a represenation that is 
just a number of seconds or something similar for keeping the value.

Kind regards,
Bernd Eilers



Hanssens Bart wrote:
> Hi Bernd,
> 
> 
> a) ok, adding the word 'accumulated' would clarify the intention
> 
> b) and it indeed makes sense that a Producer only uses days, hours, minutes
> and seconds because months and years is ambiguous
> 
> (we can safely ignore leap seconds and assume a day is always 86400 seconds,
> and I don't think someone is actually interested in seconds)
> 
> 
> c) I don't entirely agree on Consumers not having to deal with years, months, etc...
> 
> Suggestion: 
> 
> Producer SHALL only use days, hours, minutes and seconds
> 
> c1) Consumers SHALL be able to parse P0Y0M... (Y = 0, M =  0 is a bit useless,
> but valid, and an implementation has to be able to parse this notation anyway if it
> also supports meta:user-defined with a duration...)
> 
> c2) Consumers MAY treat Y = 365 and M = 30 days
> 
> 
> d) more important: what is "editing time" ? Time between opening and saving  a
> document in a non-readonly mode ?
> Or (this makes more sense IMHO) from the first change, regardless if this change
> is "undone", until saving the document. Does merely changing some styles counts
> as editing ? Modifying application specific settings ? Signing ?  (I guess it does)
> 
> Best regards,
> 
> Bart
> 
> ________________________________________
> From: Bernd.Eilers@Sun.COM [Bernd.Eilers@Sun.COM]
> Sent: Wednesday, November 18, 2009 10:04 AM
> To: oic@lists.oasis-open.org; Michael Brauer; Oliver Specht; Michael Stahl
> Subject: [oic] Restrictions for meta:editing-duration
> 
> Hi there!
> 
> Section 3.3.2.17 of
> 
> http://www.oasis-open.org/committees/download.php/33633/OpenDocument-v1.2-part1-cd03.odt
> 
> 
> defines <meta:editing-duration> Metadata as an element with a value that
> conforms to section 3.2.6 of the xmlschema-2 Spezifikation.
> 
> http://www.w3.org/TR/2004/REC-xmlschema-2-20041028/#duration
> 
> I interpret the meaning of that value as the total accumulated time
> during which the document was being edited. Since there can be any
> number of gaps since the document was created where the document was not
> edited and which would thus not contribute to this total editing time a
> value here that would contain subelements specified as number of years
> or months does IMHO not make sense although it would be allowed by
> xmlschema-2 section 3.2.6. Number of years and months subelements in
> such a value being present would only make sense if we would have a
> value for some kind of elapsed time without non counted gaps in between
> which is not the case here.
> 
> I would therefor like to have the following additional restrictions
> added to Section 3.3.2.17 of the ODF 1.2 Spezifikation and as well to
> the Interoperability Guidelines created by the ODF Interoperability and
> Conformance TC (OIC) for ODF 1.1 and ODF 1.2:
> 
> 
> 1.) A ODF Creator MUST NOT use a value here where number of years or
> months subelements are being used although allowed by the xmlschema for
> this element.
> 
> 
> 2.) A ODF Processor is NOT REQUIRED to be able to interpret such values
> here where number of years or number of months subelements are present
> although allowed by the xmlschema for this element.
> 
> 
> There is at least one Office Application (OpenOffice.org) out there
> which as an ODF Processor would not be able to handle
> xmlschema-2/ISO8601 conform duration values at this place which would
> contain months or years subelements.
> 
> Possibly section 3.3.2.17 of ODF1.2 should also clarify that the value
> here is to be interpreted as accumulated total editing time and not as
> elapsed time since document creation.
> 
> What do you think?
> 
> 
> Kind regards,
> Bernd Eilers
> 
> 
> ---------------------------------------------------------------------
> 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
> 
> 
> ---------------------------------------------------------------------
> 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
>